Western Digital Corp. - Analyst/Investor Day

 |  About: Western Digital Corporation (WDC)
by: SA Transcripts

Robert Blair

Hi. Good morning. If everyone could take their seats. We're going to get underway in 1 minute, so if everyone could have a seat. Thank you.

Hi. Good morning. I'm Bob Blair, Western Digital's Vice President of Investor Relations. I'd like to welcome you all here. We're really appreciative of the terrific turnout that we have for our first investor day in quite some time. Many of you have come a great distance, and we appreciate your being here today.

Today's event is being webcast with the webcast link and presentations available on our website at wdc.com, under the Investor Relations section. Additionally, the webcast will be archived on our website and available for replay for a period of time.

This slide contains a disclaimer about the forward-looking statements we will be making in our presentations and comments today. Forward-looking statements will be made concerning growth opportunities on the storage industry, our expected financial results, business model and financial goals and our capital allocation strategy. These forward-looking statements are based on current expectations and are subject to risks and uncertainties that could cause actual results to differ materially, including those listed in our report on Form 10-K filed with the SEC on August 20, 2012. We undertake no obligation to update our forward-looking statements to reflect new information or events.

I just want to spent a couple of minutes or a minute going over a few logistic items. The agenda for the day is in your welcome kit folder at your seat. John Coyne will also be going through it in his opening remarks. We have a 10-minute break scheduled this morning at 10:20. There is a glossary of terms on your welcome kit as well as on the website at the back of the website version of today's presentation, which contains a lot of the terms and acronyms that we'll be using today for your reference. There is also bios of each speaker.

A file containing all of today's presentations will be given to each of the attendees today as you leave the event on a WD passport drive.

As far as Q&A, we've got 2 Q&A sessions scheduled, one this morning and one this afternoon, and we ask that you hold your questions until those sessions in the morning and afternoon. And during the Q&A sessions, we will provide microphones in the audience, and we ask that each of you state your name and firm before you ask your question and wait for the mic, so it can be -- the question can be heard on the webcast.

Also, a reminder, please, for everyone to put their cell phones on a silent mode. And a reminder that no pictures or videos are allowed at the product demos that we have today. We have product demos, as you've seen at the back of the room, and we have a specific time set aside for each of you to spend at least half an hour viewing those demos in the back of the room as well as in the foyer, where our branded products team has a demonstration. And then also, I think many of you have seen our stream machine outside, which is -- and this is the -- today's the first stop of a nationwide tour for that vehicle, showing off and marketing our new router products.

Just logistically, in terms of lunch and the demos, we've got 1 hour set aside and half an hour for each. And those of you with green badges will be attending the demos first. Those of you with red badges should have lunch first and vice versa after that first half an hour.

I'd now like to introduce the Chief Executive Officer of Western Digital, John Coyne.

John F. Coyne

Thank you, Bob. Well good morning, everyone. And my welcome to you and my thanks to you for attending today. We have, I think, a very exciting program, and we intend to demonstrate during the day the passion and the passion for change that has been the underpinnings of Western Digital's success formula in driving consistent and continuous profitable growth over the last many years.

The vision for the corporation is to empower people, to create and manage and experience and preserve digital content. Our core expertise over the years has been the preservation of digital content, but as that digital content has become completely pervasive in our lives and touches almost everything we do, whether in work or play, the sheer scale of that preserved data and that communicated data is requiring that we put our passions and our skills and our experience and capability to work, enabling people to manage simply and interact and access that information in a simple coherent way, and we see significant opportunity in building on our core preservation business to expand our role in the ecosystem and to ensure that we bring continued added value to customers as we move ahead. And you'll see some of the areas in which we're -- how we're thinking about that and how we're acting to achieve that expanded vision for the corporation.

Here we go. So I want to spend a few minutes of retrospection. If we look at the 50 years, the last 40 of Western Digital's life and the upcoming 10 years, you’ll see an underlying focus on understanding change, anticipating change in market needs and reacting within the company to position ourselves take advantage of those changing market requirements on a continuing basis. Back in 1970 when Western Digital was founded as a semiconductor designer and marketer, our first products were calculator chips. And through the 10 years, we evolved. And as we entered the PC age, we were primarily a designer and marketer of storage controller chips, floppy and hard drive controller chips, which were delivered to the market for hard drives as an interposer card. So we transformed ourselves from a chip company to a subsystem board level company and became an 80% market share leader in hard drive control.

We then developed the WD1010 [ph] , which was an SoC that combined all of that board-level logic onto a single chip, enabled it to go onto a hard drive. And in order to offer the market what it needed in terms of a solution for what was now a requirement for all PCs to have a hard drive embedded in them, we acquired a hard drive company and became a hard drive company through the '90s. And then we transitioned from drives to a broader product line throughout the last decade. And we’ll talk today about some of our ideas and our focus for the 10 years to come.

Let me look a little closer at the last 10 years and, again, to look at this theme of understanding market trends, acting in a timely and appropriate way to intersect those trends and to continue to delight customers with the product portfolio and the utility that enables them to succeed in their markets. And over the last 10 years, we started that 10 years as a strong desktop-focused market player. And we built a business model that was focused on quality, reliability, speed and cost leadership. And as we succeeded in that desktop market, we looked at the trends in the market and applied our business model and caught 2 major waves of change in market expansion throughout the decade. The first of those being the proliferation of mobile devices started initially with the laptop PC, and entering that market into 2004, and within 4 years becoming the market leader by applying customer-centric solutions. We listened to customers. We anticipated who the successful guys were going to be in the laptop PC business. We listened to what they needed to succeed, and we delivered that formula, which was a highly desired solution compared with the existing competitive solutions that were in the market at that time, and drove us from entry to market leadership in 4 years in the fastest-growing segment of the market.

Partly driven by that mobility was the need for personal cloud, which at that time was called external storage. And as we listened to customers in that arena, we made investments, very strange for a drive company. We made significant investment in industrial design. And in listening to customers, we understood they wanted things that were good to look at, good to feel as well as functional and utilitarian. We delivered that and created the biggest branded products external storage business in the drive industry, and that business went from -- came into existence and grew at a rapid pace. So anticipating where the market's going, coming up with solutions to intersect that market.

And then as we came to the end of the decade, we conceived of and executed the acquisition of HGST to further broaden the product portfolio, to deepen our technology capability, to expand our people resources, the experience and the passion and the capability of the teams to look forward and address the coming 10 years. So building out the existing Western Digital portfolio solutions that addressed mobility and personal requirements and significantly beefing up our capability and product portfolio to address the other side of that market, which is the infrastructure that supports all of that information transmission and information exchange in the enterprise environment.

So we feel very comfortable about our position as we now move into the beginning of the next decade of the company's evolution. And we see that we have a very solid, very broad platform on which to build a future. And you're going to hear today about the 3 pillars of that market as we see them evolving today, the trends in thin and light mobility in individual devices, the trends in the data center and the private and public cloud arena to service all of that data requirement, and then the personal cloud environment in the home and people's connected lives around the family unit. So you're going to hear about each of those 3 initiatives by the company.

And then we continue to look, as we have for the last 40 years, we continue to look at market trends to seek, to discern directions early and to work on having the right product portfolio, the right services and the right cost profile to intersect and actually enable those market trends as we go forward.

So the agenda for the rest of the day. Steve Milligan is going to follow me and talk about our overall opportunity in the marketplace as we see it and our strategy on a broad basis to address that. Mike Cordano, President of our HGST subsidiary, is going to address the market as we see it in the enterprise and cloud environment. And then Tim Leyden, who leads our WD subsidiary will follow Mike with a view of the mobility market and the opportunities that creates in the Connected Life. And then Bill Cain -- Dr. Bill Cain and Dr. Currie Munce, our chief technical folks in each of the subsidiaries, will do a joint presentation on the technology that underpins the business and the trends that we see both in our technologies and in competing technologies and how we view that affecting the market in the years to come. We'll then break for the demonstrations, which I believe if you spend some time, there's a lot to learn about not only what the products are that we're showing you and what -- but why they are and how they address the market needs that we perceive. And then in the afternoon, Wolfgang will talk about our business model and our capital allocation strategy.

So with that, I'd like again to thank you all for being here. I hope you see as we go through the morning a very capable, committed and passionate team that understands what they're doing and have a significant demonstration of doing it well.

And with that, I'd invite Steve Milligan to come up and talk to you about our strategy. Thank you.

Stephen D. Milligan

Good morning. Certainly an honor to be here. What I'm going to is I'm going to talk a little bit about what we're going to be doing as a company going forward. First off, I'm certainly honored with the privilege of in January being able to succeed John. John has absolutely done a fantastic job building this company. And some of the points that we will get at is that we have assembled, in our view, an extremely powerful platform in a number of different ways to address the challenges that we're dealing with as an industry. A very important point is that we are arguably at an inflection point, not only for ourselves, but within the broader landscape of what we are -- the environment that we exist in. And so we're going to talk about -- I mean, change to us represents opportunity. And that is really the essence of John's setup, is that Western Digital, not only Western Digital but the heritage of HGST, is that change is our friend, and that we've got to figure out what we're doing to address the challenges and the opportunities that, that presents. So let me get into some of the details.

So I'm not going to give you a history lesson. You're all well aware of what's happened over the last several years in terms of the evolution of the computing market. Some of the key points here really are that, number 1, it's all about data. And we know that as it relates to our personal lives. We know that as it relates to our professional lives. More and more of what we do, more and more of what we touch is becoming digital. Not only that, the devices that are either creating that content, interacting with that content, they're changing. And the needs are changing as it relates to that because of the various types of interactions with that data.

The other point is, is that if you look at what's happened in the industry, and sometimes we forget this, but the things that we have done in the storage industry and specifically what we've done in the hard drive industry have helped to enable all of these changes. We don't always talk about innovation, but we have by virtue of what we've done over the last 50-plus years is innovated in a number of different ways to enable that proliferation of devices, growth in data. And given where we're at from an industry standpoint, given where we're at from a broader landscape perspective, back to the point on inflection, innovation, in our opinion, and the ability to be able to adapt as we move forward is becoming more and more important. And it's presenting for us a very unique opportunity in terms of how we're positioned from a competitive standpoint, from a value perspective, within the broader landscape to take advantage of all of those changes.

So drilling down on that a little bit more. And again, most of what I'm telling you is not really news. On the left-hand side, we've got growth in devices and again, we all know this. We've seen a proliferation of devices, different kinds of devices, edge devices, smartphones, tablets, et cetera. We're also seeing, and again, what that’s doing is it's driving or enabling substantial growth in terms of data, and that data has to be stored some place. Increasingly, it is being stored in a lot of different ways, and we'll talk more about that specifically as we move through the day. But -- and it's -- and somewhat being ironic, the death of the disk drive has been predicted for 50 years, but the primary storage preference will continue to be, during this study period, hard drives. And we'll talk a little bit more about why that's the case. But we estimate that by 2020, that 85% of all of the data that is created that needs to be managed will continue to be stored on rotating magnetic storage. And to be clear, we have critically evaluated this. We have not gone into this from the perspective of, well, let's come up with data that helps support what we want to believe.

One of the key points of this conference, I believe, is that we are approaching, or viewing, or evaluating things very realistically, very critically to make sure that we understand, not only where our opportunities exist, but also where we perceive potential risks exist and what we need to do to meet those various different challenges. The other point is, and again, this debate has existed for a long time. I mean, I remember having these discussions when I was back as the CFO of Western Digital. Will SSD, will flash replace hard drives? And in some cases, the answer is yes. But the important point is, is that it's not necessarily an either/or. The storage media or devices that we're using are complementary. We're all facilitating this opportunity for all of us to create, manage, access pervasive amounts of data.

So there's a couple of other things that are going on that I think are critical to understanding the opportunity that we have as an industry, the opportunity that we have as a company, and not only that, the challenges that we have as a company. And what that is, and Mike will talk more about this when he talks about what's happening in the enterprise space and in the cloud. Tim will talk more about it in terms of what's happening in the mobility segment and in personal storage. But the landscape or the ecosystem that we exist in is changing. And it's changing for a number different ways. And we've highlighted some of those changes that are going on. I mean we've got changes in our customer set. We've got changes in terms of the overall standards kind of ecosystem.

It's not all about the Wintel world. The use cases are changing. We've highlighted Internet service providers, social media companies. Their use patterns are different than, say, a traditional data center. Their needs are different. So what do we need to do? What is the opportunity for us? Not only that, there have been changes. There has been consolidation in the supply base, right? That is providing, all of this is providing us an opportunity to drive deeper relationships, deeper different kinds engagements with our customers, call it more customized kind of solutions that help address the unique requirements, the unique needs that our customers have. And again, Mike will talk more specifically about what we're doing about it in the enterprise space. But this is a big opportunity for us. And that opportunity means that we can add value in different ways than what we've done historically.

It's not all about, let's get x number of vendors lined up. Everybody's got the same kind of device, minimal differentiation. Let's go figure out, we bid everybody off. I mean, these customers are willing to do different things. And with that opportunity, it provides us with the opportunity to derive value from an economic standpoint in a different way. And I don't know if that's a well understood concept, but that's a very important aspect of what's happening in terms of the broader changes that we're seeing within the industries that we compete.

Additionally, on the mobility in the personal storage side, we're seeing similar dynamics, not exactly but similar dynamics. We've got customers that are going through substantial changes from their standpoint, right? They're trying to figure out what all of this means to them. Value and what customers or what consumers are looking for is changing. Also, standards are changing. The ecosystem is changing. Again, it's not necessarily all about Wintel. We've got different devices interacting with data in different ways, creating different data in different ways. We’ve talked about the movement to thin and light. What are we doing, what do we need to do to not only address that but in fact to lead that change. And that's the key point that we're going to talk about as well today. And Tim will dive a little bit more deeper into some of this. And then as individuals, as consumers, we all know how the data that we're dealing with or that our families are dealing with. And how do we manage that content? How do we access that content? How do we do with that data and what we want to do with it and do it in a way that's easy.

So what does that mean to the hard drive market? Now, let me sort of caveat this, or not caveat, but let me go back to a point that I made earlier. Again, we have attempted through our valuation of the market, our plans, our strategy, to again be critical with regards to how we view the future. So we'd like to believe, and of course, all of you may have a different view, one way or the other. We're trying not to fool ourselves. We're trying to make sure we are being realistic with regards to where we see the industry going. So right now, we're forecasting through 2017 a growth rate in terms of aggregate spindles being shipped of something like 3%, certainly down from what we've seen historically, right?

But there's a few key points in that where that growth is existing largely is areas where we believe we have the opportunity to add value in different ways, which gets back to the previous 2 slides in terms of the changing aspect of the landscape and the opportunity that, that provides. Additionally, with that change, and not only that, with the size of the industry, we continue to believe, and Wolfgang will talk more about it, is that we're going to continue to derive substantial amount of economic value by virtue of what we do. And so the financial returns, we believe, will continue to be very attractive in our industry and in our space.

So what are some of the opportunities that we're looking at? Now I don't want to steal Mike and Tim's thunder, but John talked about this as well, but we're seeing trends in 3 key areas: mobility, thin and light, all of that area. We're seeing the cloud, public, hybrid, personal, the Connected Life as a third area. These are key areas where we see opportunities, where customers, whoever they may happen to be, are facing different challenges and where we have the opportunity to innovate and bring technology into the marketplace to not only address those concerns, but more importantly figure out different ways to lead those changes in ways that are beneficial to our customers, beneficial to us and allow us to derive more economic value.

Additionally, and let me go back on that -- additionally, because of the changes that are happening within the landscape, within our customer set, we are now -- we have a different platform than what we've historically had. We're being asked and being provided the opportunity to do different things with our customers, and we're fully up to that challenge by the way. But that allows us to continue to look at opportunities to evolve our model in different ways, and not just simply be a component provider, but to transition our model in a financially responsible way into different things, solutions-oriented kind of things, to look at software as a mechanism to solve different problems, to help address some of the infrastructure needs that our customers are dealing with. And so we're going to continue to evaluate and look at opportunities and work with our customers to figure out how we can in a mutually beneficial way and in an economically beneficial way expand our platform to address that opportunity.

And I'm going to talk with a bit more about that as well. So what is our -- the essence of our strategy? And now on the left, we talk about in essence the platform that we have. And I can tell you -- and a bit of history, I mean, when John approached me to acquire HGST, there were lots of discussions that we had. But the essence of the discussion, which oh, by the way 6 months into the combination I think that is only reinforced, is that we believed that we had the opportunity to create a platform that would allow us to address the challenges that we're facing much better than we as individual companies would otherwise have been able to do.

So, yes, it's clearly a financially beneficial transaction, in our view, but the real strength of the combination is the power of the 2 platforms coming together to address the challenges that we're looking at. We have fantastic people. Hopefully, you agree as you interact with some of the folks that are here. The platform that we have, where we sit in the ecosystem, the broader landscape, we are in the middle of all of these discussions that are going on with our customers regarding all of these changes that we're seeing. And so we have the opportunity, we have the technology to address those challenges. And as I alluded to earlier, equally important is that we have a fantastic financial engine that will allow us to leverage those capabilities in ways that we believe will continue to drive value further on down the line.

So more specifically, let's just dive into some of this a little bit more. The 3 key areas that we've talked about: mobility, thin and light, cloud, enterprise, Connected Life. You can see that the numbers there, the 44%, the 46% and the -- yes, the 46% and the 45%, that's our market share today. So we are obviously a substantial player, a leading player in each one of these markets. And we have done and we will continue to do and as the product and technology demonstrations show, we are going to continue to innovate, do different things, help customers solve different problems in each one of these markets to not only maintain the momentum that we've got, but hopefully accelerate it. Because obviously, the better job we do at addressing the issues that are presented by these 3 segments, the better we'll do from a customer perspective and from a financial perspective. But the point is, is that we're well positioned today and we're going to use that position to continue our momentum.

Another thing, and I think this is an important point, and it may represent somewhat of a change at some level from what you have traditionally heard historically from Western Digital. But because of the changes that we're looking at, because of the platform that we have, because of what's happening in the overall landscape, we have the opportunity, obviously, to innovate. We've innovated for a number of different years. We don't always talk about it, but that's what the industry has been all about. But even more important is that we have the opportunity to make markets or influence the outcome of a given situation in a different way than what we've done in the past. We have the opportunity to move the market and move the market in a direction that we think makes the most sense from a customer perspective, from a financial perspective, what have you. And we've already done some of that. So we've got some proof points on here in terms of what we've done historically as separate companies, the introduction of 7-millimeter drives, some of the leadership position that we've established there, some of the things that we have done from a cloud perspective, additionally some things that we've done in terms of the Connected Life. Also, as I alluded to earlier and as the product demos demonstrate, there's more that we're going to do. And hopefully we'll see an acceleration of this because we need to do what we can to meet and lead the challenges and opportunities that we're seeing from an industry perspective.

Additionally, and very near and dear to all of our hearts and certainly near and dear to my heart being an ex-finance guy, is that at the core of what Western Digital has always been about has been operational excellence. And not only that, with operational excellence, but strong and I mean, great value creation from a financial perspective. That tradition, that element of what's at our core will absolutely continue. So you can see that on the left. Over the last 5 years under John's leadership, the company has generated an enormous amount of economic value. I hope that I have the opportunity to continue that and certainly will work my tail off to make that happen.

Additionally, if you look at what we've done from a capital allocation perspective, now I'm not going to steal Wolfgang's thunder when he talks about capital allocation because I know that you're all anxiously awaiting that. But the capital allocation strategy of the company historically was actually very deliberate. Might not have always appeared that way, but John knew prior -- and not only that prior management knew, the executive team knew that we, Western Digital, continued to have opportunities to fill out our platform. Obviously culminated at least most recently in terms of the acquisition of HGST.

So the capital allocation strategy followed that strategy of the company. And that's allowed us now to grow that pie, right? So we're going to generate more value because of the strategy that we have from a capital allocation perspective, and that's allowing us to do different things.

So we announced back in July an expanded stock buyback authorization of $1.5 billion, so we've got the opportunity to increase the cash return that we're providing to shareholders. And the circle on the right, just so everybody knows, is just kind of representative. It's not -- don't try to guess what it all means, right, and send it to your trading desk. Don't do that yet. It's just representative. But we -- but there are several key points on this thing. We have demonstrated over time an ability to intelligently allocate capital, whether it be CapEx, R&D, growth, M&A, et cetera. And we're going to continue to use that expertise to allocate it in ways that make most sense not only in terms of what we're trying to do from a business perspective, but also from a value creation perspective as it relates to our shareholders. And so we'll talk more about that this afternoon.

So in summary, so what are we really trying to do? I mean, when you really summarize it, I mean, the first thing is, is that again we will continue to rest on the rock of operational execution as an organization, as a company. Both companies have demonstrated over time. Certainly Western Digital has a longer history of it. We do a lot of good stuff at HGST to get ourselves to where we got to that point as well. Great team, great people, great capability. We're going to keep focused on that, and that will be the cornerstone of what, we do.

You're going to see more, and you are seeing more from us in terms of innovation and market making and gearing things in a direction that we think makes sense and allows us to derive value in different ways going forward.

And then lastly, we're going to use that opportunity, we're going to use that change in terms of the platform that we've got to continue to look at different opportunities to evolve our business in different ways to not only continue to succeed in terms of our existing platform, but to figure out what are we going to do in the next 10 years? How are we going to fill everything out and beyond that? Because as John has ably reminded me, I need to carry on the tradition that he set up, so I've got a high bar for myself.

So with that, I will wrap up my presentation and turn it over to Mike Cordano to talk about the opportunities in the cloud. Thank you.

Michael D. Cordano

Good morning. Before I get started, I'll make my excuses in advance. I'm dealing with a bit of a head cold, so I hope my voice holds up. And if I say anything that's too absurd, I'll blame it on the cold medicine. So what I'm going to try to do with my presentation is to deal with this concept of the cloud that's often talked about and frankly, is complex and confusing. I'm going to try to strip that down a bit such that we can derive what it means for us and what those opportunities will be for us going forward. So hopefully, I'll be successful in delivering that to you all today.

So first, let's talk about the market itself. I'm going to drill in a little bit on the cloud segment specifically. And when you think about the petabyte or exabyte growth here, it's being driven by a number of aspects. So the most obvious one is stuff that's either stored in or moving through the Internet. And I'll get to that in a moment with a graphic on the right. But it's at massive scale.

The other thing that's driving it is compliance and regulatory action. So those factors are enabling a massive amount of additional storage and retention. And it's also creating new and different types of markets for us to serve. And then when you think about quality of service across all this infrastructure and you think about a content delivery network as an example, you have 2 to 6 copies of this content being deployed around the world. So all of that is driving massive growth.

You then look at the graphic to the right, and this is an example of every minute of every day what kind of a scale and activity is happening across the Internet. And it's just massive. 2.1 billion people connected at any one moment and actively doing things. And at the bottom of this graphic, it talks about the data is growing with no end in sight. And this is attributed to a web analytics company. This is not our own data. But that's their interpretation of this. And we certainly stayed in the underlying demand, growth of data and opportunities that are in front of us.

Now let's talk a little bit about the cloud. What is it? NIFT, which is a standards group, has very tight definition of that. Frankly, things are moving so quickly. We have a more simple definition of what it means to us. It really means anything that's accessed across the network either by a browser or an application. So think of it that simply. That can be in an enterprise, so that's more of a private cloud situation. That can be, of course, in a public cloud and through the Internet. It can even be in the home, right? Tim's going to talk about the personal cloud, but that's the same situation, right? The trends are absolutely the same here.

So the other thing that we want to do is look at, in fact, the types of clouds that are out there. Starting first with the thing that everybody sees, which is the Internet. The consumer interprets that as things like Netflix, things like Google. That's how they interact. And certainly, those companies and their products are representations of cloud infrastructure being deployed.

But then on the left you see other names you recognize that are really there to build upon, right? And those are people that will either allow startup companies or software-as-a-service providers to load their value-added applications on their infrastructure and deliver it to the market. It also lets enterprises scale their applications outside of their 4 walls. So we look at them independently. They have different objectives. But what ties them all together are the services, the infrastructure and the storage that is required to make them happen.

Then we look at the cloud. What does it mean? What does it mean to infrastructure? And basically what's happening and why we're seeing so much turmoil and change is that the hyperscale that's being given by this massive amount of activity is breaking the old model. So the old monolithic IT model no longer copes with the requirements of what's happening here. So we see a number of things happening. Steve alluded to this. It's no longer just exclusively a Wintel world. Well, part of that is people trying to deliver differentiated capabilities to deal with the scale and size of this cloud and Internet trend. Data center gear is becoming more specialized. What that means is providers of these applications have a very clear idea about what they're use case is. Very clear. It's very narrow and it's a very large scale, so they can custom-design the hardware, the software and the whole solution to optimize their result.

So it's no longer a general-purpose world. That's the way we need to think about that. And that again gets to an opportunity for us to participate in creating value add. And in all that scale, there's a huge focus on total cost of ownership, and it's not about just exclusively cost per gigabyte or cost per terabyte, which is how everybody wants to think about disk drives. There is a number of other dimensions, which I'm going to get into, that deal with total cost of ownership. And we as a hard drive and SSD provider have a lot opportunities to influence them very directly.

There are also things happening relative to the underlying technology with a RAID world, again from a protection perspective, but that's all changing. Replication is, it is now becoming more predominant. That's why you have this 2 to 6 sort of replications across the web both for protection, but also for quality of service and performance. And then you're seeing new technologies, which our technology team will allude to like erasure code and other things that are further enabling this. So lots of change, which is our opportunity to interpret that, work with our partners and deliver value-added solutions.

Let me touch on the cloud. The cloud is a panacea. Everything go into the cloud. Well, I think you could have a theoretical conversation about why that might be so, but there are a number of challenges both in the data center as well as being users of that cloud as an external basis. So you think about device reliability. I've talked about the scale in which we're doing things. When you have a reliability issue, the multiple effect on the cost dimension is huge. I'm going to talk about this more specifically later. The other one is power. One watt per drive savings translated across this kind of scale is millions of dollars of operating cost saving. So when you think about our ability as a device provider into the space, maybe we can do unique things that bring that sort of change, it's very valuable to our customers.

From a user perspective, there's a number of issues. Tim's going to talk about some of these as well. But it's not possible today to move to a thin client world where everything’s in the cloud and we're going to all have dumb terminals or tablets and access all 100% of our content and computing in the cloud. You have upload bandwidth constraints that's why edge cache devices make sense. You have data portability.

Well, what happens when that service changes. As a user, that's a problem. I might lose that data. You even have companies that are new and interesting that may not be here in 2 or 3 or 5 years. So all those things are driving a corresponding challenge relative to the use of the cloud. But obviously, there's lots of innovation going on in both ways. We have opportunities as Western Digital to create opportunity, to create differentiation on both sides.

Now let me talk a little bit about what private, public, hybrid. Tim is going to talk about personal, so what is that from our interpretation? First, private. The easy way to think about this is private is inside the firewall. That can be in the Fortune 1000. That can also be in my home, right? So the version -- a personal cloud is really a private crowd for a consumer. That's the way to think about that, right?

Public cloud is outside of the firewall. There are concerns and issues that need to be dealt with around security, but it's also a pay-per-use sort of model. Even when you think about Google, although you're not paying for it, you're getting advertising delivered, right? So how does this translate? There's pluses and minuses of both. And what we see happening, the most likely case, it's not going be one or the other is going to prevail. There's going to be a hybrid use case. And when you think about this in the context of a corporation, you do first processing. You don't want to build your capital infrastructure to manage the peaks of your business necessarily. You're going to bring it to an optimal level, and then they're going to do surge capacity at externally on applications in which you feel comfortable doing that. So the combination of the 2 allow, in this case, the CIO to make decisions about where information is and what processes are happening.

The other one -- rather an example is in retail. We all know that the calendar fourth quarter is when the highest velocity of retail online transactions happen. Well, do I want to build out infrastructure for the fourth quarter and just have it underutilized for the other 9 months of the year? Probably not. I want to have flexible capacity to take advantage of that surge, but I don't need to have it operational year round. So there will be variety to this, but we absolutely see this is as the final outcome. Certainly, there's going be private and public, but in combination they will create what is often talked about as the hybrid cloud.

All of this, translating it further to what it means to us. We talk about storage tiering and there was these sort of simplistic, monolithic models of computing in the past. And what's happening is what we now know, with the advent of storage or computing virtualization and other technologies, is the models are no longer monolithic. Meaning you don't have an application, single application running on a specific defined set of infrastructure. So that's no longer the way these things are being deployed with these technologies out there. And so what we see happening is we see a balance between performance and capacity and applications being -- that are common, represented by the different colors here, having different components of what they're doing, needing to reside on different types of hardware to deliver what's required. So using a social networking example, there was some recent press around Facebook moving 100% of their storage to flash. Well, interestingly enough, there are certain analytics that they need that are very high performance that will be flash based. But the bulk of the content, the vast majority of the content resides on rotating magnetic media, and in this case, the most cost effective, highly reliable part of that. But that is all one integrated application. So it's about how they optimize the performance and where they put what critical activity.

So at this point, we're now already servicing a vast majority of this marketplace. We have a leading position in fiber channel and SaaS, SSD, so we're already at the high-end, high-performance part of this marketplace. We have been for some time providing 10,000 and 15,000 RPM drives that are right in the center of the performance band. And then we've also been growing in recent years and having a substantial contribution in the capacity enterprise space. So we've got broad coverage already. Steve talked about our market share here. But there's more opportunity because what we're seeing is increasing fragmentation and increasing need to optimize the workloads that are happening in these data centers.

So the first one we see as a growth area going forward is to continue to go up the performance axis and participate in the very highest performance, for example where high-frequency trading happens. I think you guys in the room are familiar with that. We need to continue to evolve our SSD strategy and participate beyond where we are today, which is giving us great learning, but we need to add to the portfolio our PCIe product in the future. We're not announcing a product today, but I think we are directionally committed to cover this space going forward, so you'll have to watch this space in the future.

But why do we want to do that? Let's address another topic, another buzzword: big data. What's interesting, I was in Boston at one of my favorite customers last week. And one of the quotes [ph] they talked about is the understanding of big data by the broader market is about 2 or 3 years behind where the cloud is. So that 2 or 3 years ago, people were looking at the cloud. There's a lot of talk about the cloud, but nobody understood it. I think we have a better understanding now. Big Data's at that stage. And what is it? There's sort of a simple way to distill it down, talk about the 3Ds of big data and variety, which is about data sources, data structures, data types. And in the old world, in the database world, you had very organized structured data so you knew how to deal with it. In this world, you have certainly that structured data, but you have semi-structured and unstructured data that all need to be dealt with to refine potential value.

The other thing is scale. We used to be terabytes. Now it's petabytes, going to exabytes. And the number of installation, of exabyte installations, is growing. So that is just a pure scale challenge and within that scale, you have, of course, the different data structures. And the last thing is the velocity of the data. It's no longer batch mode processing. It's real time. You want to be able to interpret that data real-time, so all these 3 vectors create challenges and complexity, but if solved will generate absolutely significant value. And for us, what does that mean for us. There are a lot of software companies out there trying to do Big Data analytics. That's great. They’re our partners. But they need more sophisticated, more capable infrastructure to be able to deliver on that promise. So I've already talked about high-performance SSD for analytics, very critical. But it's not just an SSD play. We have to be able to deal with the volume of that content simultaneously, which is why rotating magnetic media. And in the future, which I'll get to in a moment, new classes of rotating magnetic media will further enable this. So it's creating opportunities, participating add value at the device level, and even look at new architectures which allow us to consider more value-add up the stack, which I'm going to talk about in a slide or 2.

The bottom end of the performance capacity, there is a new emerging area that disk has not covered historically. It's typically either been not covered at all or covered by tape, obviously backup and archiving, and I'm going to get into that the next slide. So this is the next area we believe we have a great opportunity to innovate. We announced a product platform this morning. It is the first glimpse of what we're thinking here. And obviously more to come.

So what is cold storage? Another new word that we're talking about a lot. It's not single use case. There's a variety of potential use cases. There's probably as many as there are people in this room. But I think you can simplify it into sort of 2 general groupings. The first one looks a lot like backup. And it's about you're accessing infrequently either writing to or reading from. And then all of the rest of the time, it's idle or off, right? That's the cold, up at the top of this. So it looks more like a tape market historically.

Then there's cooling storage. We’ll use that word. And that's things that are frequently accessed at the beginning of time and overtime will decay in terms of the rate of access, but will never go to 0. Both of those groupings are important. They both are required to underpin what our customers and partners want to do. And these represent huge opportunities for us to do differentiated things going forward.

Let me try to put a little bit of sizing on what that might mean for us. So just tape replacement, and so you get an idea of scale. The tape market, you can see it here in exabytes, is large. If we look at this year, maybe it's 1/3 to 1/2 of the size of the enterprise disk market. So that alone is a nice adjacency for us to participate in, but we have to develop products that allow us to do it with the right metrics.

Then there is this notion of unstored data. This is very intriguing to us because if look at the slide on the right, that little dark green sliver at the bottom is the entirety of data stored on enterprise-class drives, 3% of the market. All we have to do is generate an additional 3% that is useful out of this unstored data, and that doubles the size of our market opportunity. And what is this unstored data? Most of it is media, video in most cases. In some cases, images. But I'll give you one example. Social media, big social media guys down-res all the images that come into their infrastructure, and that's driven by cost. But I'll tell you, they are losing some users because users don't want their content eroded or down-res. So it would be in their services' best benefit to be able to retain that original image. But today, they can't do it based upon the cost of their infrastructure.

So if we were able to work with them and on a total cost of ownership basis, we're able to deliver something acceptable, that is a new market that doesn't exist today in terms of exabytes. So that's just one small example, and there are a number of them like that. So this is an area where we will do a continued work with our partners to understand how we can create products that allow us to capture more of this unstored data going forward.

Cold storage, is it the future, is it reality? We already have a proof point of this happening in the market. We have Amazon, AWS, announced their Glacier product. And without saying anymore, it's a disk-based product. And we feel very strongly that this is validation of this general market segment, and there is absolutely more to be done going forward. But it's not just a wish and a prayer and Western Digital's desire to create a market, it's already happening.

And to do all these interesting things, we have to evolve our capability and enhance our influence on ecosystems. Steve talked about this already. This graphic shows sort of 3 generalizations of the engagement model for different market segments. We have the traditional server market, which has historically been a Wintel bastion. Our ability to innovate is really represented by the blue. So of course, the HDD -- the HD interface and some work at the controller level. But the rest of the value-added stack, we were really not participating in. That was the province of the file system guys and the systems guys.

As we move over to the one more to the right. In storage systems, you're dealing with companies that control their own file systems in most cases. So in our partnership, we have a little more opportunity to do interesting things up the stack with them to optimize solutions and bring more value to the party. And then furthest to the right, in the cloud infrastructure, is where there are so much innovation going on that we have an opportunity to engage all the way up and down the stack. And while we can bring value add it’s being invited to the party.

So this is a very new world for us. And we have to through different kinds of engagements that are more consultative, more deep technical relationship, have skills, we're going to bring value to the party. We can't just show up with our pencil and paper and say, what do you need? They're not interested in partners like that. We need to have an opinion. We need to have some value added in that conversation to have a constructive debate that ultimately translates to a better outcome.

Part of what we need to do is build new capabilities to accomplish that. So I'm alluding to knowledge around infrastructure and architectural focus. That's something we've been investing in for some time. And we're investing on top of our historical capabilities. As hard drive companies, they're a fantastic foundation, but they're not enough for where we need to go in the future. So we need to have more technical knowledge around how data centers are built, the painful [ph] I've talked about and what we can do in our products and solutions to solve those issues and bring benefit to our customers. And all of this is going to be things you'll see us continuing to invest in from a capability standpoint as we move forward to take advantage of these opportunities.

And one outcome of that would be our participation in Open Compute, which was an open source hardware project. We actively participated in that. Open Bulk [ph] is the storage version of that. And that's where we've been active in trying to add value-added thoughts to the final outcomes here. So these things are happening very quickly. These are places we haven't been participating directly in the past and we are increasingly today.

So this is interesting, but in the end, it comes down to how do you enable cost and how do you enable these services in a way that makes sense? So when you look at this graphic, this is not -- this is representative of where data center, hyperscale data center costs are allocated. So servers and storage is just that exactly what it says, the infrastructure and the capability. Then you have power and cooling. Everybody knows that's an increasingly high or a large component of the cost of running these scale data centers. You have facility bandwidth, networking and administration. And this is the breakdown. And these are all the things we need to think about how do we influence them because this is what our partners and the end users of our product are having to manage relative to their cost performance.

So reliability. This is an interesting one. We've been focused for some time on having very high-reliability products. Well, it's obvious why administrative costs might be lower because you don't have to have a technician go out there and pull that failed drive out and replace it. So better you are there, your there, and administrative costs are lower. The one that's sort of harder into it is bandwidth. Why does disk reliability help bandwidth? Well, I talked about replication. Remember when they replicate, they replicate across their WAN. That costs them money. So if you have to restore across the WAN, in this example, that's a cost to our partners. And the fact that we can avoid that and have a low instance of failure actually helps them save on their bandwidth costs.

We've developed in our 5 disc [ph] enclosure a purpose-built HDD for capacity enterprise. That has allowed new capabilities in the raw storage in terms of terabytes per spindle at high reliability. But it's also allowed for increasing density, which helps the facility. And that's not density -- aerial density in the drive, that’s spindle density at the system level. So we influence those 2.

With the addition of our SaaS product and our participation there, again, we're delivering a different performance point in the tiering that I've already talked about. So that's enabling capabilities that were previously not there.

And finally, we announced the sealed-drive platform this morning. Interestingly enough, it certainly does the normal things. It helps us with density. It helps with capacity. Helps us with a number of very predictable things. But the one that is most pronounced here is power and cooling. So we're going to -- and Currie will talk about this later. But flying in helium allows us to significantly drop the power per terabyte that this offering delivers. So not only do we deliver increasing areal density and the other things that we've historically done, we now deliver a very specific advantage in the area of power and cooling. So we've got in aggregate -- here we go -- in aggregate, across our portfolio, we're influencing a number of these key cost elements. And these will be things we continue to focus on and continue to bring value on moving forward.

And for my last slide, I really want to -- I talked about a number of things. All the new applications and new things happening, some of the ecosystem trends, which are now more diverse, more complex. Certainly the new TCO measures, which I just talked about, all of those things, moving the way they are as aggressively as they are, are creating opportunity. And as Steve talked about, our customer engagement is proliferating. It provides complexity, but it also provides opportunity. And as this ecosystem evolves, storage is very clearly right at the center of it. We have a very good position to influence outcomes going forward, how we will be investing as we move forward to make sure we are delivering and creating value at the center of this ecosystem evolution. I think we're in a very unique position. And our plan is to continue to evolve our position in the broader ecosystem.

And with that, I'm going to wrap up and turn it over to Bob. Thank you very much.

Robert Blair

Okay. We're running a few minutes ahead of schedule, which is good news, so we're going to take advantage of that. We're going to break now, but we're going to re-adjourn at 10:25 instead of 10:30. So we can take advantage of that extra time in our upcoming presentations this morning.

A couple of things I do want to mention before we break. For those of you wishing to access the wireless network here, if you haven't figured it out, it's the Ritz-Carlton wireless meeting network, and there's a drop-down menu for conference. And the password is the word "WEST," all uppercase, W-E-S-T.

Just a couple of other things. I misstated the time that we'd have for demos and lunch earlier, it's a total block of 2 hours, so an hour for lunch for each of you and an hour for the demos. And then finally, in addition to providing you with the passport with the presentations on it as you leave today, we'll also sometime tomorrow be posting all the presentations in PDF form on our website. Thanks.


Robert Blair

Okay. We'll resume our program. I'd like to introduce the President of our WD subsidiary and a longtime member of the executive team at Western Digital, Tim Leyden.

Timothy M. Leyden

Thank you, Bob. So I'm going to talk to you today about levering change into market opportunity. So Mike spoke to you about data centers in the public and private clouds. I'm going to talk to you about the Connected Life, which is enabled by the -- all the devices -- the proliferation of devices and networking and the ability to access the storage from anywhere, anytime, anyplace. And also going to talk to you about the increasing trend towards mobility and thin and light devices.

So we've looked at storage demand and we see it as a function of homes, people and spending power. And that spending power driving exabytes of storage into devices, into data centers and across the networks. Then when we peel it back further and we look at the different needs of the developed world and the developing world, we see that the developed world, it values convenience and in valuing convenience is willing to use -- buy and use multiple devices and also use the networks. So consequently, they have the income and they're willing to pay for that convenience. However, when we look at the developing world, it's a different picture. We see that the developing world values versatility. So consequently, they want as much functionality as possible in a central device in the home. And they didn't want to be able to expand and extend that device as their incomes increase or as they earn more money.

So you should be familiar with this chart by now. It shows the size of the opportunity. You've got billions of people using billions of devices driving lots of storage demand. And as Mike talked to you about it, that won't all be accommodated in the cloud, in the private cloud or the public cloud. It will actually -- some of it will be accommodated on a personal basis, whether that be personal cloud or client. So I'm going to talk to you first about the personal cloud.

So the personal cloud, it enables the Connected Life. And Connected Life is the ability to access your data across multiple platforms, multiple applications from anywhere at anytime. And the 3 critical components of that capability is storage, ease-of-use and network speed. So let's talk first about storage.

Direct attached storage obviously is on the peripheral of the personal cloud, but it's a particular strength of WD. We've built up a large business. It has great potential with one-to-one attach rates, both on computing devices and handheld devices. It has -- it'll be further enhanced with Windows 8 capabilities. We do provide ease-of-use through our backup applications and it's got great potential in the developing markets as an expansion device as I talked about earlier in developing markets where they want to expand their central device. We're going to continue to innovate and we're going to continue to be a leading -- the leading provider in this particular segment.

So next, talking about network attached storage, which is a great enabler of the Connected Life. It's a huge opportunity for WD and for the rest of the industry, what we feel we're ideally positioned in order to capitalize on that. We already have significant market share. It creates [ph] --senses a network attached device. It provides wireless connectivity across multiple devices in the home and on the go and as well as that, we have software applications, which enable ease-of-use from both in the hand and on the go.

So the glue that holds all this together is software and it enables the ease-of-use. We have a suite of applications, which are aimed directly at improving the ease-of-use. And you can see them here whether they'll be WD 2go, WD Remote, WD Photos and our computing apps, backup utilities and security. We continue to invest in this area. It's an area in which we feel we have big advantage in this market, and we'll continue to enable the capabilities and evangelize this particular capability because it isn't an experience that's familiar to everybody. So consequently, in taking a leadership position, we're going to work on creating that awareness in order to capitalize on what we believe is a huge opportunity.

One thing I didn't mention was we are cross platforms. So consequently, we provide the capabilities across the iOS, Windows, BlackBerry and Android. And we will continue to provide that capability across platforms that have scale. So consequently we are agnostic to whatever platform is being used.

I'd encourage those of you that haven't done so yet to visit our demos. We've got demos in the hotel and demos outside in the new WD stream machine [ph]. And as you go in and look at our products, you'll see how well they operate both on their own and in concert with other devices and across platforms.

So now turning to mobile devices and the trends in mobile devices. The PC's size has evolved over the years. It's not a new concept. And HDDs, hard drives, have evolved in order to accommodate and enable that system evolution. You can see here where we have recent entries into the marketplace in order to enable the further acceleration towards light and thin and mobile devices. We've advanced from 9-millimeter products to 7-millimeter, 2.5-inch. We're continuing with a 5-millimeter, 2.5-inch. And additionally a 5-millimeter hybrid, 2.5-inch drive. So again, this is just a reminder of the potential, just talked about the personal cloud. There's an opportunity for us in clients to use innovation and enable our customers to drive a larger TAM in the client space.

I'm having some technological problems, I'm sorry. So a proof point of how accepted that innovation and enablement is. We've already shipped more than 50 million 7-millimeter devices. There has been strong market acceptance and we're the market leader in providing those products. The 5-millimeter device will be approximately 50% of the volume size of the 9.5-millimeter drive. So this is an indication of how we enable our customers to be successful and to be able to provide appealing and compelling products in the marketplace.

Our 5-millimeter drive can fit in 15-millimeter devices and even below that. And just by point of reference, the first Apple iPad was a 12-millimeter device. Our 2-disk 7-millimeter device is an ideal product for the developing markets because it provides that expandability that I talked about earlier by providing an upsell for our customers in the marketplace. And it can fit in a 15-millimeter to 17-millimeter device. But size alone is not enough in this marketplace. Customers desire low power, higher performance, less noise in addition to the advantages of an HDD, which offers higher capacity and good price. So with our hybrid devices that use the best of both technologies, we're able to offer a great value to our customers in the marketplace in order to be able to provide compelling products to consumers.

We're not the only one that has that view. Behind me is a chart courtesy of Intel, which shows the value that was gained from their first-generation time-to-market Ultrabook where it had a sole source device which was SSD. And if you look at the blue section of the chart, you can see the value that was generated by moving to best of both using small amount of cache and then HDD in order to be able to provide better value to the consumer at a lower price point. And the collective innovations that were done between generation 1 generation and generation 2 enables a price point reduction of $300, which is a -- and price points drive volume in the marketplace. So it's a significant enabler to our customers.

So again, I'd invite you to go to the back of the room where we have demonstrations showing the capabilities that we have in the different devices. You can see the comparator benefits of capacity, cost and performance. SSD obviously offers higher performance but lower capacity, higher price. The hard drive offers good price and higher capacity. The hybrid product offers in-class capability at good performance and high capacity, so it offers great value in the marketplace.

So to sum it up, what we are doing is taking advantage of the change that's occurring in order to capture more mobile opportunity in the marketplace and increase the TAM, and in the Connected Life space, we're providing ease-of-use networking and also storage in order to provide better functionality to the customer and to the consumer. Thank you very much.

Carrie Munce, [indiscernible] research at HGST.

Currie Munce

Thank you, Tim. So now we're going to get a little bit of technology behind us. I will be giving this presentation jointly with my colleague, Bill Cain from WD. This will be a little tag team here. So we'll go back and forth. But we're going to get in to a lot of electrons and spins and nanometers and gigabits per second. And I just want to give you -- you don't have to remember all that. But that should help you understand, I think, the 3 key points we're trying to make here and really understand the motivation and the grounds for what we want to say. And first is that no one storage technology meets all customer needs. The world and the ecosystem is tiered. We provide tiered solutions. We always have and we will going forward. We do that by combining products with different attributes and combining technologies on and together such as an hybrid drive. But we also heard from Tim and Mike talking about changes in applications, in big data, the cloud, mobility, social media. We're seeing different attributes, different requirements evolving. And as Bill and I go and tear this down into the technologies and the products that required to support that, I hope you see the great challenges, the great opportunities for us to innovate and create even more compelling solutions as we go forward through innovation. And the last point is this is a very competitive industry. Only the technically most advanced and innovative companies will survive. But Western Digital has a deep heritage in its people, in its patents, in its partnerships and a proven track record and technology and products. And so while we'll talk about some of that, we really do invite you to come into lunch time and see some of the new innovations, the new technologies, new products we are displaying just kind of show you what we think the future is.

Let me start with going back to the chart Jim just showed looking it from a customer's perspective. What are the requirements? The requirements are diverse. They're very broad. From weight to cost to capacity. No one technology is capable of meeting all these requirements. And these requirements are where a user's making-decision varies by that consumer, by the application they want to run, by the location where they want to put it and even by time, these need to be dynamic systems that we can balance going forward. But if we really look at this, the 2 lowest, most important priorities go from the bottom up in general. Availability and reliability, those are really necessary to play. If you don't have that, you're not going be a storage product that's going to sell in volume.

The next 3, though, capacity, price and performance are the areas which are most interest in terms of tiering, the most interest in terms of balancing requirements. So let me get a little bit quantitative here, give you a graph, the y-axis, the vertical axis is performance, higher performance at the top. The x axis is capacity per dollar or bid costs, cheapest device to the right-hand side. If I were to go back 15 years ago and normalize these 4 technologies to the cost of SRAM at the time and to the performance of SRAM, performance being measured in terms of enterprise-type workload requirements. We get a graph that would look like this.

Ideally, I want to be in the top right-hand corner. Fastest and cheapest. There's nothing up there. Everything falls generally along a diagonal that we see. And as you look at roughly the slope of that line, you see for every 100x increase, you want to get a performance, you have to pay 10x more in cost. Another important thing to recognize is that, that these get reasonably spaced out. It's hard for any new technology to get in because it's hard for a new technology to penetrate against a large incumbent, which has a large install base.

But the most important thing about this graph, I want to make you, is it helps us understand where we are, how we got to where we are and most importantly, allows us to see where we think we're going forward in terms of technologies and requirements as we move forward. And that's because as we look at each one of these technologies, we can track trends in terms of how the performance and how the cost is going to change and as those change over time, the relocations of these bubbles will change and give us guidance to what's going to happen as we go forward.

So if we do that, we fast forward from 15 years ago, what will we see? We'll see -- most importantly, you'll see that HDD and tape dropped down in performance because their mechanical devices and their performance is not scaling with semiconductor speed.

In doing that, the large performance of gap opened up between DRAM and HDD. So what’s happened? What is happening and emerging? We see NAND technology, NAND flash technology coming in terms of SSDs entering to fill that point, but following again on the same diagonal of 400x increase in performance, you have pay 10x more in cost. SSD has been around for 20 years. There's nothing new particularly about SSD. We've seen evolutions, we've seen advancements with MLC, we've seen advancements with the controller technologies getting into single chips. But the key thing that drove it was the performance of HDD did not scale, it created a gap. And the SSD is coming in to fill that gap as one more tier in this architecture, not as a replacement for HDD, as another tier adjusted to DRAM with a tier to HDD. And our HDDs today are shipped with DRAM on them to provide some kind of cache buff on the front end.

So where are we going to go forward? I'd said we can learn about the future. Let me just take a second before going into that to really get into the 2, which are the biggest growth here, which is flash and magnetic rotating disk. So let me get in a little bit of the technology here. HDD and SSDs rely on magnetic recording and on the flash cell. Both use the electrons to store the information, store the zeros and ones. In the case of HDD magnetic, we use the spin property -- electron spins, the electron spins around its axis. In a magnetic crystal, there's 2 preferred directions, the easy access in which you want to spin, which are 180 degrees out of phase with each other. If I orient those magnetic crystals on a disk, it'd be pointing straight up or down with the spin, we call that perpendicular magnetic recording. And so as that electron spins though, it can be influenced by the presence of magnetic field. So if I bring a recording head here, you can see an implied magnetic field, I can cause that spin to flip down. If I apply a magnetic field, again, in the opposite direction and cause that spin to rotate back up again.

So I can use an extra magnetic field to influence the spin up and down roughly corresponding to 0, 1 is a little more complicated than that. But as I go over to the flash side, I now have a silicon substrate at the bottom, on top of that I have a floating gate and above that I have electrodes, the electrons and I use the charged electron, the location of that charge to store the information. The electron sitting in the floating gate is a one. The electron, if I apply a positive voltage in the silicon substrate, though, opposites will attract, the electron will move down to the silicon substrate and that will be a 0. If I apply a positive voltage on the electrode above, it will attract the electrons back up, go back up on to the floating gate.

Now I showed one electron in really case of flash, there's tens of electrons in the floating gate. If I move all of them in unison up, and all of the unison down together, there will be single level or SLT NAND flash. There are techniques that have been developed to be able to move portions of those, so I can get more information stored and get multi-levels in the cell MLC, which is conventionally is only referring to 2 bits per cell.

But there's 2 key distinctions here I want to point out before moving forward is, one, if I look on the case of HDD and we're using magnetic field, you have a bar magnet, you can attract the metal object from a distance, I don't have to touch it. That head does not have to be touching that disk surface. It can move relative to the disk surface. That will be important [indiscernible] in a second. In the case of flash, they're all interconnected. All have to be packaged together as one.

The other important attribute to understand is that there's a gate oxide that sits in the floating gate and the silicon layer. That's a normally a nonconductor. It's a nonconductor because it has to keep the electrons from moving between the 2 without me applying those voltages. The process of moving electrons through there is actually sort of a non-natural process. It's actually referred to as quantum tunneling. How I make that go through? And every time it goes through there, it's a finite probability it gets trapped or causes damage. Someone asked me one time, can you explain that even more simply? I said if you see all those chase movies, action movies and they're always on those tall buildings and they always run across the roof, and they come to the edge of the alley and they got to jump across the alley to the next rooftop? That alley doesn't jump across, that's you jumping across, that's the electron transitioning. Every once in a while, it's always the bad guy, of course, who doesn't make it, falls down the alley and that's the damage in the trapping.

So if we look at leading edge MLC today at 20 nanometers, the number of times you can do this re-write before you wear out that cell because you have too many electrons have been trapped or damaged is only 3000x, so this will be an important thing. It's important to the endurance or the wear-out mechanism in NAND flash, and that will be a challenge and we'll talk about later when we get into how we implement these in products.

So now how do we take this technology and move it in to create products out of it. Well, both again is very similar. Capacity is equal to surface area times the density. Now this is a density on the surface or sometimes we refer to as aerial density to differentiate from a volumetric density. So how do I do that in the case of HDD? Well, what I'm going to do in HDD, which is very unique, I'm going to use only one element to find the density, and that's the head element. So I do a lot of sophisticated processing, a very fine lithography on one 200-millimeter head wafer, which can yield about 80,000 heads. And I take one of those heads and I can scan it mechanically across a very large, very cheap, very inexpensive to make disk, a very large area. So I only have to put the density in one element and I get a very cheap surface area scanned across from the mechanical advantage.

And the mechanical advantage, again, goes back as magnetic field, so I don't have be touching, I just have to be in close proximity as I apply across the surface. When I do that, that long headway will produce 40,000 terabytes of storage. Now when I go over to NAND, I said the NAND all have to be packaged together. So that surface area that I have scale up to get capacity all have to be finally featured. So it becomes a very expensive NAND wafer and a 300-millimeter NAND wafer. Today, I will get about 2 terabytes of storage of -- bytes out of that, which is 20,000 times less than that smaller headway per produce. But the advantage is they're all interconnected, so I use wiring in rather mechanical motion, I can access data faster, I can gang them together and say die together behind the controller and I get parallelism to get the highest speed. So I'm getting the speed from the NAND by the interconnected wires and parallelism. I'm getting the cost out of HDD by the mechanical advantage to be able to scale one element. So today, the cost differential between HDD and SSD is about 10 to 1, 10 times more expensive for NAND flash or SSD.

Now how is that going to scale as we go forward? In general, our advantage in moving forward in time is that we keep the surface area and the cost of that surface area roughly stays constant and what we do is we take the density up. As the density goes up, the capacity goes up. For the same dollars I get, $1 per gigabyte goes down and I continue to scale. So what's important to understand about that 10x ratio is we're -- looking at what's going to happen with those aerial density trends in HD versus the NAND flash. And Bill will come up in a minute and talk about how we see those trends and what they will mean going forward.

So let me just kind of wrap up my first section here. No one technology does it all. We have to build products out of tiers. We already built products out of tiers. I think the key message here is that we now start to see a new tier which hasn't classically been in there 10 years ago in there with SDD and flash. That the width of this triangle tends to represent the capacity of the number of bytes that we have in very fast devices at the top which are very expensive, and we tear them down with technology and more bytes towards the bottom, which are slower but are more cost effective. And the key issue as we go forward is how do you continue to manage and build these systems are dynamic, and dynamic and manage the data, which you need very fast up to the top tiers where you can get the speed of it and move the data that doesn't need to be up there, down to be more cost-effective storage down. And again, a lot of the discussion Tim and Mike referred to is more focused on software to go forward to how we play more in that interface between how do I move this fast data up and take the data that doesn't need to be up there further down.

But with that, let me introduce Bill Cain from WD.

William Cain

Good morning. So now that Currie has shown you how the differences in the technologies lead to the tiering in the storage system, I'm going to go focus in on one of the key attributes to any mass storage technology, which is its capacity.

So on this chart, what I'm showing you is the key enterprise technologies, and that's hard disk drive, it's flash and it's the 2 forms that Currie talked about, single-level as well as multilevel cell and tape technology. And this is showing actual aerial density data from 1999 to 2012 and looking at how the technologies have grown capacity over that time. So a couple of key things that you can take away from this graph. And one of the first things, is you see the blue line, which is representing hard disk drives is actually representing the very highest aerial density which, I think, to some people, is a little bit counterintuitive. You think that the solid-state technology has a higher density, but in fact it's actually the magnetic storage technology. And then the other thing that this chart is representing is that this has been explosive growth or a geometric growth. And it seems hard to see because it's in a straight line, but it represents a 41% compound aerial growth rate, which means that the capacity or the density has doubled every 2 years. So over the time frame that I'm representing here, that's 100-fold increase in the density or in the capacity. And when you looked at the charts in the previous presentations, they were shown on a linear scale, and so that does look explosive and it's an exponential growth. I plotted it on a log scale here, so you can see that the other key feature is that the lines that represent hard disk drives and the line that represents flash are in fact exactly parallel. And so what that's showing you is that the capacity growth rate, the relative capacity growth rates over that 10-year period are the same for those 2 technologies. And back to the point that Currie had made earlier is that keeping those lines parallel means that the cost from bytes has come down at the same rate as well. So the differentiation in the cost per byte has remained the same. Both technologies have come down at the same rate. And the other point on this chart is that the tape technology has actually grown a little bit smaller at about 28% in terms of capacity or density growth, which represents 3 years to double.

So the next chart is now forward-looking, so anticipating or looking at what we think the growth rates will be for these 2 key technologies, hard drives and flash going forward, and again, the punch line there or the takeaway is that we see these growth rates continuing at a parallel rate although slightly slower than the historical rate. And if I get into a little bit of the details about how both technologies achieve that, we look at the incumbent technology today in flash as floating gates as Currie showed you, and in the magnetic recording, it's perpendicular recording, as he also discussed. So both of those technology do have extendibility. So there are more products that will come out in both of those as we scale density, and that will be done through traditional scaling, as well as some extension technologies in the flash side. That can be things like charge trap devices or dual floating gates. And on the hard drive side, through technology, such as the shingle magnetic recording where we change the formatting of the bits on the media to increase the density.

Although as we continue to extend those technologies, at some point, extension of that technology becomes challenged and we actually need to deploy an invention to really change the rate of growth and keep on this historical growth rate. And for flash, Currie talked about that floating gate and the number of electrons on that gate. So in the 2015 to 2016 timeframe, as that cell scales down to something on the order of 10 nanometers, the number of electrons that are on that floating gate become very difficult to detect reliably. And the other challenge, of course, is making a feature that is very small requires a significant capital investment in new photo lithography tools such as EUV. The results of those 2 things are driving the flash industry towards 3-dimensional flash to really be able to utilize an existing capital structure and extended the technology.

On the magnetic side, now we also have a challenge as the bit becomes smaller, that magnetic field that we used to write the information is no longer sufficient to do it alone and so we need to add an additional source of energy in the form of heat for heat-assisted magnetic recording that allows us to continue to extend technology.

At the very far right side of the chart, look at technologies even further into the future beyond 2020, both industries again need to innovate, deliver some invention to be able to extend density further, and both industries are putting substantial research into those extension technologies. And while I don't want to imply that this is easy to achieve this for either technology, I can say that in the hard drive industry over the past 50 years, where we've delivered that explosive growth in terms of capacity, about every 5 to 10 years, we have had a major new technology that has been employed to achieve that.

The next slide now talks about, as we introduce these new aerial density technologies, the application into which they are deployed is really going to be dependent upon the relative maturity of the technology. And while I'm using NAND flash and SS as the example here, this is absolutely true for hard disk drives, too, where we deploy new aerial density technologies first into our client products followed by our enterprise products.

What I'm showing here is the last 3 major flash nodes which was 34, 25 and 20 nanometers, and showing that from the time that we have an initial product announcement or availability of that technology to the time that there are consumer grade products, it's roughly 2 quarters or 6 months. To the time that we have client-based SSDs that offset from announcement is typically about 1 year. And till the time we have an enterprise class SSD, the offset in time is roughly 2 years. And if I focus in on the 20-nanometer, which was the most recent technology node introduced, you could see that, that technology was originally announced in mid-2011. By the end of that year, there were consumer-grade products available. But the first client SSDs were just announced last month, and we don't anticipate availability of that technology, 20-nanometer flash, into the enterprise class products until late 2013.

So when you think about something like 20-nanometer technology, obviously, that's been around for a while and we have a tendency to think about the cost per bytes in terms of that consumer-grade application. But you really have to think about how this technology needs to be matured to deploy into storage applications.

Okay. So now that we've sort of established the technology basics for you, we're going to move on to discussing the opportunities that both Mike and Tim talked about in the different areas. And Currie is going to come back up and focus on the data center and the cloud.

Currie Munce

Great, thank you, Bill. So as I said, it's a tag team, I'm back. Thank you again. So I talked about technologies. Now we're going to talk about product because not only do we have to build tiers within technologies, we have to build tiers within products. We have to look at product design attributes and how do I scale those attributes to represent. This is a chart Mike had in his presentation, talked about the diversity of requirements ina cloud and the applications. He showed you the types of products that we have in here. You can look and see in the enterprise, data center, HDDs, we have products optimized for performance. We have products optimized for capacity. We also bring in complements on a high-end performance SSD. Above that being a different technology, but also leveraging the same firmware support, same unique understanding of customer requirement, same compatibility testing as the HDD, so again trying to in some way blend away the technology difference and make the interface look the same, but we have to design these products differently.

So in just the period of time that I have this morning with you, I want to kind of go 1 and 2 about what we're doing uniquely to try and drive towards the cost and the cold solution -- cold storage solutions that Mike talked about, then I'll come back and talk about some of the unique things we have to do in designing for that very high performance. So let's start by high capacity. What do we do differently to design high capacity. Well, you remember before I said it was capacity and surface area times density. So I wanted to make the biggest possible surface area I can. So I go to the largest possible disk diameter I can get to get the largest area that fits within the form factors and power constraints at the 3.5-inch disk. Then I want to try and put as many of those platters together as I can in the form factor in our leading edge product being 5 platters to get the maximum surface area, to get the maximum possible capacity. Now if I explained that correctly, I won't say that, straightforward, that's simple until you realize the mechanical precision required to achieve this and how to continue to scale it.

The disk in these products, the typical disk is spinning 80 miles an hour. Now not that I want any of you go out and driving miles an hour, but if you were driving 80 miles an hour, imagine your window down and sticking your arm out, and now the equivalent precision that you need to hold the tip of your disk or the tip of your finger to be equivalent to what we have to do inside here is you have to hold that steady to 1/100 of the diameter of a human hair. Imagine doing that 80-mile an hour. So it's a tremendous challenge as we continue to design forward.

So we also look at spinning the disk a little bit slower, 72 RRPM, in some cases, 54 RRPM depending on the trade-up between power and performance requirements. So again, we can change the RPM go a little bit lower, little less power, little less performance. Also have 72 RPM versions that have little -- higher power but better performance. And remember when you do high-capacity and you have lots of bytes, you better make them very reliable or else you'll be pulling them out of the data center all the time. So it's important to have high reliability in each products. And so we have some very leading-edge products, 5-platter enterprise.

The industry’s only 4-terabyte with 2-million-hour MTBF reliability. So this is -- the customers love these products for the cloud applications that Mike talked about. Really exciting leading edge stuff. We're getting to a point though, we're challenged how do I keep going, how do I keep scaling density and what I need to do next about the pinpoint Mike talked about power? Well, what if I took the air out of there and put helium in instead? Helium is a gas, which is 1/7 the density of air.

First, is a lot of the power that's being consumed in a driver's due to sharing that air, 80 miles an hour spinning in there, sharing a lot of air. You reduce the density, power goes down significantly. And additionally, I talked about you're putting your arm out there. Imagine reducing the density of that forces by 1/7 hitting your arm, you'll get a lot more precision. We see the power reduction here is 23%, but because of that increased precision in helium I've been be able to replace the 5 disks with 7 disks, gives me 40% more capacity, gives you a 23% reduction in total power. Overall, I get a 45% reduction in watts per terabyte, almost a 2x increase in capacity. Now you can put it in a data center for the same amount of power, very compelling solution.

So what's -- and we have the demonstration of this technology in the back of the room for you. So you'll be able to go back there and see and compare it to a conventional 5-platter drive.

Now what's innovative about putting the helium in there? Well, it is innovative to put it in there, but that idea has been known for a while. The real innovation is not how to put it in, it's how to keep it in. Here is a very small atom. It tries to get out of everything. I think we all have familiarity with helium balloons and how quickly they go down compared to air. So the real challenge is how do I seal it in there and keep in there for 10 years and guarantee that product's going to work. Now today, that's still proprietary, and I'm not going to discuss it here. We'll skip forward to this chart here.

But innovation driving, getting more scaling allows us to go more capacity in the future and kind of going after power as we go forward. So we use mechanical design, we do other things that are simply pushing density as we go forward.

Let me go to the performance axis. We're doing SSDs. We launched our SSD business several years ago, really bringing our expertise and understanding the systems, understanding the customers, understanding the unique firmware requirements, understanding how to test and do system compatibility testing with them, working with a partnership with Intel. We bring in the NAND expertise and reliability, very, very powerful. But what do you have to do? You should see a similar graph again. Every graph we seem to have is performance on a y axis and capacity and cost on the x axis. You see different kinds of NAND because of that endurance. If we took conventional, consumer-grade MLC and put in some enterprise applications, they would wear out in less than 3 months because of that only 3,000 times you can write that, a very heavy duty stuff. You're going to wear the NAND out. So what do you have? You have different tiers of NAND capability.

Robert Blair

This is better.

Currie Munce

All right. Great. Thank you. We see at the bottom-end here, TLC, 3 levels. It’s really only writes the wear out, a few hundred times. It's really only useful in consumer, electronics applications. Is that really relevant to any kind of high-performance computer application? MLC consumer grade can be used for very read intensive. I don't write often, but I read when I'm streaming data out. There's a new class of eMLC. The e refers to enterprise, has 5 to 10x higher endurance. The build side, it tends to be shifted and delayed relative to the consumer in terms of when it comes out, and it comes out at a higher cost point. And finally, the highest performance, highest endurance but the highest cost is SLC. So again, to provide solutions that meet our customers' requirements, we have to build products that tier and utilize the different technologies.

So we have SSDs using both SLC for the extremely write-intensive applications, 35 petabytes of random writes. We have eMLC offerings to reduce down to 7 petabytes of random writes but at a lower cost point. And we'll continue to work on over provisioning and control the technologies to continue to try and use the NAND to pull it up because, as Bill said, every time we shrink the lithography, every time the density goes up, the number of electrons goes down. That wear level comes down. We have to do even more to continue to pull that up to meet the customer requirements.

So even within these technologies, we have to tier to continue to build. But let me go to my last slide here before turning it back to Bill. Coming back to this chart again, looking out to the feature, trying to represent, again, where we see future opportunities. I showed you HDD actually scaled. Bill showed HDD scaled faster than tape. And that 15 years of transition, HDD is starting to encroach. As we do more to design and use helium and other things to get more surface area and as we're driving our cost points down towards tape, enabling new opportunities like cold storage to go address. But what would have had to be done on tape or data that was thrown away, we now have opportunity to bring that into HDD.

Flash is emerging. Flash is coming. But we can combine those technologies. And Bill will talk more about hybrid, combining the flash with a rotating hard disk drive to get a balance of cost and performance between those technologies. And in the higher end, Mike talked about, as we go up, either the higher end of flash performance is going towards PCIe interfaces to get lower latency buses to allow us to have even higher performance and move it closer to the processor and the system bus rather than on its storage bus. And beyond that, we continue to look out 5 to 7 years and try anticipate trends as John talked about, see where things are going, anticipate them and provide solutions and start working those beyond that.

The one technology we see on this graph, today, which is most challenge that's giving us is actually DRAM. DRAM have this capacitor in it. The capacitor's difficult to scale down in size. So we think the technology, which we'll have the most difficulty will be DRAM. There's a number of technologies being pursued as DRAM replacements. The 2 most mature are MRAM, magnetic random access memory, and PCM, phase change memory, have a little bit of difference in cost, performance, endurance trade-offs. These are certainly still 3 to 5 years away from introduction in mask products that are in niche markets today, but in terms of really truly entering mask market here. But these big attributes, they're non-volatile, so as we introduce DRAM replacements that are non-volatile, we see even further changes in the ecosystem and understanding and anticipating that, as we go forward, gives us great opportunities to continue to even look beyond the future as we go out.

With that, I'd like to turn it back over to Bill to talk really about the influences of mobility and the personal cloud.

William Cain

Okay. This will be the last speaker switch I promise. All right. So now I'm going to go ahead, and -- turn this round, so it's going in the right direction -- and build on the themes that Tim had originally introduced in terms of mobility and the personal cloud. So first up, I'll start talking about mobility, and this slide is another Intel investor slide. It was illustrating the key attributes of the new mobile computer, of the Ultrabook. And as Tim indicated, Intel has stated that really the best solution to be able to meet all of these attributes is a combination of SSD cache with a hard disk drive, which produces a hybrid solution. And the next set of builds really are our data on the proof points associated with each one of these attributes, and the first 2 are focused on the performance or the responsiveness. And here, I'm showing you an HDD, an SSD and a hybrid and the relative performance to each one of those things. So the first one is a benchmark, a PC mark advantage, and the second one is this Rapid Start to rapid resume. And you can see that the hybrid technology is very close to the ACD, SSD technology, which is significantly better than the HDD.

From a form factor perspective, by going to a 5-millimeter drive, we have doubled the volumetric storage efficiency of the hard drive, and this really enables the sub-15-millimeter designs. From a cost perspective, now here, the hybrid actually approaches that of an HDD, well, from a $1 per gigabyte perspective, which is significantly better than an SSD. From a battery perspective or a battery life perspective, again, the hybrid is very close to SSD. And then finally, if we think about the fact that under this high usage eventually that flash does wear out inevitably. The -- your data is still safe and secure and you have peace of mind of knowing that, that information is safe.

So looking forward -- or talking now about the form factor and the work that we've done on the 5-millimeter design. And again, we've stated a couple of times, volumetric efficiency is significantly improved, and that's shown in the graph in the center. But when we set out to design the 5-millimeter drive, we didn't simply take our 9.5- or our 7-millimeter design and scale it down. We worked very closely with our OEM and ODM partners to really understand the application that this was going into, this ultra-mobility type of application, and we reimagined how you would design the drive specifically for that application.

So one of the things that we identified early on was that the legacy SATA connector was actually taking up quite a bit of space on the PCDA. And so we worked with our partners to develop a new connector that is 70% smaller, and then doing some more work on the PCDA, we really looked at what we could do to leverage the latest in terms of packaging technologies to be able to create the smallest board we possibly could. And actually, that enabled us to move the board to the very edge of the drive where traditionally the PCDA extends over the entire bottom of the hard drive. This saved us 2 millimeters in total thickness and allowed us to significantly increase the space inside the drive to be able to account for the shock kind of environment that we see in a mobile application.

And then finally, one of the other key attributes in a mobile application is vibration and performance under vibration. And so what we did there was we leveraged technologies that were employed in our enterprise products such as fixed shaft motor and dual-stage actuators to really deliver the best performance under vibration possible in the mobile environment.

Another area for innovation and this is also alluded to in Mike's presentation is in the total stack, and the total software stack that -- the total storage stack. In the past, we really just focused on delivering the device at the bottom of the stack that stored the bits. But over time, we realized that there’s great value for our customers and for the end user if we provide innovations throughout that stack to deliver a better overall solution. And so the data I just showed you on the Intel chart and the demos that you'll be able to see firsthand after the break will show you how we've been able to leverage that kind of benefit into the hybrid performance, where we have a hybrid drive in conjunction with a driver and some OS services that deliver a total system level of performance that is significantly enhanced. And as we look forward, we see that there’s continued opportunity to be able to leverage further into the stack and deliver value add and also support to our customers, not only in hybrid but also on SMR drives, on SSDs, in the enterprise and in our branded products.

On this chart, trying to once again reinforce the fact that the application really determines the storage technology and that one size does not fit all. So looking at a couple of different applications here, and the first one I'm looking at is a cost-focused application. So here, again we’re not really worried about what the performance or what the capacity is. It's been really about the cost, and so an obvious direction to go is a low die count SSD. And what I'm showing there is today, the capacity for a 2-die SSD would be 16 gigabytes. This would be the kind of solution you might find in a tablet today. So not really appropriate in terms of enough is enough, but you say, well, hey, flash densities continues to scale. At some point that 2 die gets to be interesting. So we looked at what timing that would be.

So by 2020, we would expect about 128 gigabytes in that form factor. So again, starts to begin -- become interesting. But you could see a couple of those other attributes listed below is not just focusing on capacity and cost, but things like performance and lifetime are also read for that particular application. And the reason for that is, as what Currie talked about earlier, that the design of the SSD is really based off of the application it's going be used in. So there’s going to be a lot of data transfer like there is in a client PC as opposed to something like a smartphone or a tablet. The lifetime expectancy of a particular technology is going be reduced. So with just 2 die, we don't think that the lifetime or, in fact, the right performance will be acceptable for that application. So a couple of die more are probably required, maybe about 4 die, which, again, now has a cost point that looks more like a hard drive from an introduction perspective.

On the other extreme side, there is applications that require best in performance, and cost is no object, and for those applications, a high die count SSD is a great solution. So today, the typical client SSD ships with about 16 die and offers fantastic performance, albeit at a very high cost. If the application is capacity focused, once again, the hard drive delivers the greatest value proposition as far as delivering in capacity and reliability. And then as we've been talking about, the hybrid opportunity is really a place where performance and capacity and cost all matter, and we think that this is a great opportunity and a great solution.

So moving on now to the personal cloud. And actually, as I start this discussion, took an extreme position on the other side, which was to say, all of the information is going to move to the public cloud. So what are the different trends that dictate the rate at which that occurs? So as we started doing that work, we uncovered a couple of interesting trends. And so one is that the amount of content in the typical home is growing exponentially again, so at about 38%. And there's a quote from Gartner here, which says that the amount of content that needs to be stored in the home will grow sevenfold from 2011 to 2016.

The other trend that actually sort of works against that right now is how fast is the pipe to the public cloud growing. So this is the data from a service provider called Akamai, and they serve about 25% of the web -- of traffic today. And they have this app on their website that allows you to look at the speeds that are available throughout the world. And this is looking at the developed countries, the G7. And what we found there is that speeds of that last mile of the content served to the home is actually growing at 15% CAGR, so that's 5 years to double, significantly slower than the rate at which the capacity is increasing. So you can see that, that actually could start to lead to a unsatisfactory user experience, and anybody who's tried to back up their PC to a web back-up provider knows what I'm talking about.

So we saw that, and then the other thing that's going on is that, again, in a developed world, the number of connected devices is growing very rapidly. So you have your smartphone and your tablets and your TV, all of which want to consume this content, which is growing very rapidly in the home . And we think that the combination of those trends really leads to this opportunity of the personal cloud.

So building on that a little bit more, again, showing the home here. We have all of these different connected devices, and they want to access this growing amount of content. And I think most of you today probably operate in a situation where you have maybe a PC that has a lot of high-resolution digital photos and all of your music library and some video, and then you service that up to the television or has some set of that information on your iPad or iPod or smartphone. And as we were thinking about that, we said, what is this analogous to. And it's sort of analogous to a corporation where you have a lot of different users with their PCs and they're accessing some -- a bunch of central data. And what enables that to occur is the fact that you have this tremendous amount of bandwidth within the building. And likewise, in the home, with a wireless router or a wired network, you have incredible bandwidth inside the home, and it's essentially free because once it's installed, it's available to you. So it really enables the ability to move all of this content amongst these different connected devices and share that information very easily.

And of course, there is a connection to the outside to the public cloud, too. And again, we talked about that pipe as a little bit constrained, but it's available at a reasonable cost or you can say, hey, I'm just going to bypass that directly and go straight to the cloud through a wireless network. But again, that is the most constrained in terms of bandwidth, and it's also the highest cost connection.

So again, we look at this as a great opportunity to put storage inside the home, and so we are delivering products to meet the decision. And so you're going to see some demos on this, but we have our My Net routers, and these are optimized to manage streaming content to help move that information amongst those devices. We have network-attached devices such as our My Book, My Book Live. We also have media devices such as our WD TV Live Hub, which has storage in it and also serves content throughout the home.

And as Tim indicated, actually, the most important piece of all of this is of software that makes it all work together and how you move content and how that is shared and utilized. And so some examples here is we have software for our NAS devices that allow you to back up content from multiple devices into the home onto the single device seamlessly. We have applications that allow you to share your contents that you have in your NAS device, anytime, anywhere, on any device. And we also have applications that enable the hybrid cloud like Mike talked about, but this time, it's a hybrid between the personal and the public. We have applications that allow you to move content between your home NAS device and providers -- and public cloud providers such as Dropbox. So again, a lot of innovation in the area of software like I talked about before. We see this as a real opportunity to continue to deliver value add.

So again, in the beginning, John talked about the history of Western Digital and its ability to evolve and transform itself as the market changes. With the acquisition of HGST, we also now have a rich heritage from Hitachi Data Systems and the storage system, as well as IBM that invented the hard disk drive. And Western Digital has been making appropriate investments. You can see the blue bar growing over time as far as R&D spending to be able to develop this broadening product portfolio and try to capitalize on these opportunities. And with the combined entity going forward, we will continue to make significant investments in research and development to enable the future.

So I'm going to end with a slide that Currie started with, where we talked about the key messages that we wanted you to take away today, and hopefully, we have achieved that. So I think the first one and everybody was already aware of this coming into the room, but the storage ecosystem is really changing and evolving right now. But the thing we want you to also take away is that there's no one single technology, which can service all of those needs. And the storage stack by its very nature is a tiered solution, has always been and continues to be. And the interesting thing is where it continues to break into sub tiers and the opportunities that creates. So this change is creating opportunity.

And again, I hope that we've shown you through innovation, we are really stepping up to meet that opportunity in areas like cold storage in SSDs, in hybrid, in mobility and in the personal cloud. And also, as I talked about as the portfolio expands, the great opportunity is being able to not just deliver hardware but the software solutions as well. And then finally, as we showed you in the investments that we're making, we have a commitment to put the investments in for the people, for the technology and for the partnerships required to execute the vision. Thank you.

Question-and-Answer Session

Robert Blair

Thanks, Bill. In the spirit of Western Digital efficiency, we're 15 minutes ahead of schedule. And I'd like to ask all of the speakers from this morning, with the exception of John -- he'll be chairing this afternoon's Q&A session. Steve Milligan will be chairing our morning Q&A session now with Mike Cordano, Tim Leyden, Bill and Currie. So we've -- Michelle Soffel [ph] will be -- and Tammy Schaefer have microphones, and we're bringing them to folks in the audience. And we'll be happy to start out on the Q&A. And if you could please, if I don't identify you, identify yourself and your firm. Aaron Rakers.

Aaron C. Rakers - Stifel, Nicolaus & Co., Inc., Research Division

Okay, perfect. So Aaron Rakers at Stifel, Nicolaus. I wanted to understand the hybrid strategy a little bit more. It looked like on your chart, you showed roughly about $0.15 per gigabyte as the cost, I think, versus an 8 gigabyte -- or $0.08 per gigabyte for a traditional hard disk drive. Is that the correct way of looking at that? And how are we thinking about the economics of that market, as it ramps through the model relative to a traditional hard disk drive?

Stephen D. Milligan

So, Tim or Bill, would you like to -- either one of you.

William Cain

So that particular number really was dependent on the assumption there, and that was for the 500-gigabyte drive that we have back there with the amount of flash that we have on it. So I think it's really going to depend on the relative balance between the flash you need to enable the particular application. And again, that's not one size fits all either, so you could have a different amount depending on the application requirement, as well as the amount of hard drive that, that flash is supporting. So I don't think there's one single number that will capture that.

Robert Blair

Rich Kugele. Michelle, over here [ph] .

Unknown Executive

Oh, [indiscernible]. Sorry.

Richard Kugele - Needham & Company, LLC, Research Division

It's Rich Kugele from Needham. Just wanted to tap in a little bit more on the notebook side. We often get asked the question what's enough capacity. So if you have any thoughts on have you done any work on that side. And then, separately, given the changing life cycles of some of these notebooks, do customers even look at it that way anymore? Or do they try and future, proof with extra capacity? So any thoughts on that?

Stephen D. Milligan

Yes, that's one of the -- Bill, again that's...

William Cain

Sure. So again, enough is enough is kind of tough. That's going to be very subjective, and the number that gets thrown around for certain applications of something like 128 to 250 gigabytes, but it's 256, but I don't know that there's a single number that really captures that. And it also depends on the market. I think if you're talking about for a commercial PC, because of the nature of the IT system, there's probably an opportunity for enough is enough sooner than in a client space. And also I think it's going to depend on where you are in the world and some concerns about security and are you willing to have the small amount of content in your data stored in the cloud somewhere or do you want to have your content with you. So I don't know that there's one number that you can use for enough is enough.

Stephen D. Milligan


Unknown Analyst

Jerome McCarthy [ph] from Heartland Group[ph] . I apologize if I missed this, but you guys talked about a 41% CAGR in terms of air density growth for both hard drives and NAND over the past 10 years. Have you talked about going forward, you're going to grow at similar rates but perhaps slower rates. But I wasn't sure if you could quantify that, so maybe if you could kind of quantify that over the next 5 to 8 years in current rating the $86 [ph] that you guys talked about this morning.

Currie Munce

Yes, so I think we -- generally, looking at from a hard drive, obviously, we have our own internal perspective as we've done a lot of work with externals looking at the roadmaps from various communities, the inputs from the key NAND flash folks. So we see both probably going around 30%.

Robert Blair

Howard Rosenberg [ph] .

Unknown Analyst

Just to follow-up on that last question, if we were to poll the NAND flash producers, would we hear a similar projection? Or is there a lot of debate as to what that future or the density growth rate is both for hard drives and for flash? What kind of band of projections is out there?

Currie Munce

So I think if you probably get the most aggressive you'd probably get someone like ASML who is trying to sell EUV and went probably as high as 40%. If you were to look at the international technology research group -- tooling group for the semiconductors, they'd say 20%. If you asked the NAND guys, they'll tell you around 30%. So we'll probably range from 20% from the -- so they've typically been conservative. And ASML at 40%, which is being probably typically aggressive, and so I think the NAND guys are saying 30%. So I think you would – they would agree with that.

Unknown Analyst

And would they debate the fact that hard drives will be able to continue to keep pace [ph] .

Currie Munce

I have not heard anybody from the semiconductor background debate the magnetic recording roadmap. Bill, I mean, have you?

William Cain

No, I haven't seen that.

Currie Munce

I think that comes from our expertise, and we feel reasonably comfortable with the technologies and investments we're making in terms of where we're going.

Robert Blair

Go on [ph] . Up here in the front, Michelle.

Unknown Analyst

Monet Debook [ph] with Trust Company of the West. Have you tried to see the relationship between the explosive growth in handheld and tablets and the growth in storage in the cloud? Is there some kind of a relationship? Could you...

Stephen D. Milligan

Can you clarify what you mean by relationship?

Unknown Analyst

What I mean, like if I'm using handheld, I'm using things from the cloud, downloading video, doing -- checking things, I don't need as much local, but I still need to have storage, bigger storage someplace now.

Stephen D. Milligan

Mike, can you add in?

Michael D. Cordano

Yes, let me just try to tackle that. I think when you think about a device like a smartphone or a tablet, we don't see that as a primary storage device. It's a way to interact with the cloud. It's a way to create content that gets to the cloud, and it's a way to take advantage of the new cloud-based or web-based application. So we absolutely see content moving there. So if you think about a photograph, if you take it on your -- if you take it on a camera or a smartphone, well, you're going to have a primary piece of content stored somewhere, likely on your PC ultimately, maybe in piece of cloud storage, but it's the way you use that content, trying to push it up to Facebook. So it's all these new use cases that are driving it. It's not just looking at storage repository and where is it going live. It's about how this content's being used. So there isn't -- again, we've talked about this. There's not a one-size-fits-all model. What we know is the trend is that way in a very aggressive manner because we see new emerging use cases for that, in this case, content. So that -- it's not like, okay, if I'm going to transition where I keep certain amount of bits in the old model to a new model in the cloud, it is not as simple as that. It's not a one-to-one relationship. It's a very complex relationship based upon use cases. But what we know is there is a dramatic growth in the cloud, I mean, depending on who you listen to. It's the petabyte growth rate there is between 50% and 90%, and it's outstripping the growth rate in other places for these reasons. So that's the relationship we understand. It's growing at 1.5 to 2x the rate at a petabyte or exabyte level than hitting in the devices, planned devices.

Robert Blair


Mark A Moskowitz - JP Morgan Chase & Co, Research Division

Mark Moskowitz, JPMorgan. Two questions if I could, maybe direct them toward Steve and Mike. I appreciate the long-term growth outlook for your industry you put up earlier around 3%. I think that's reasonable just given the secular challenges some your end markets space. But the 2 loaded questions here are, one, can you help us understand more about your assumptions behind the PC in terms of desktop and notebook growth and in terms of the notebook growing about 5%, have you – well have you and the desktop declining? In terms of -- how many times do you expect users to refresh their smartphones and tablets in terms of iterations before they refresh their PCs? The question is really is around the useful life assumptions of the PC is relative to smartphones and tablets. The second question, it is around the enterprise. Obviously, that 9% bogey definitely looks attractive. I'm sure Seagate is going to do whatever they can to protect it, but clearly, HGST and WD as a combined entity in the enterprise have to be a pretty formidable asset that I would think the EMCs and HPs and others you want to work with. So how should we think about your growth in that market relative to that 9% bogey for enterprise?

Stephen D. Milligan

Yes. So you want to take the smartphone one, and I'll take the enterprise one.

Michael D. Cordano

Yes. So I think the way we think about it is rate of refresh. What we know is happening with tablets and smartphones, just think about it from a disposable income standpoint. We know the refresh cycle for notebooks is slowing, right? So that's assumed in these numbers, right? So that's a critical part of the assumption. Obviously, we put forward our best view of what it's going be now substantially less than a double-digit growth that we saw before. But also I think embedded in those numbers is you think about form factor, 3.5 versus 2.5 in client applications, there’s some assumption relative to 2.5 having some life in an all-in-one sort of desktop application as well. So that's one of the countervailing metrics underneath the growth rates for desktop and mobile that you saw earlier. Difficult to predict. You know we are also looking at scenarios up and around that. That's our best estimate based upon all those factors but something we have to continue to monitor.

Stephen D. Milligan

Yes. And I think relative to the enterprise space, I mean, we both combined companies have actually done a very good job in terms of increasing our penetration into that segment. Arguably, we are -- I mean, we are as much of a leader as Seagate in that market. You can kind of debate the numbers. The numbers are always debatable at some level, but we are arguably as big and as formidable as Seagate is. And we've done that in a variety of different ways and Mark -- Mike talked about it earlier, but we've done it because we've got great products with high amounts of reliability. We've been able -- and it's obviously not just one product but a setup of products that are addressing the tiered needs from a customer perspective. And also, we've got an engagement model from a customer perspective that allows us to add value in the appropriate ways. And so relative to our participation in that, I would certainly like to think that we're going to maintain where we're at, and if we do a better job on that, we're going to give increase our presence there. So I don't think that it's fair anymore to assume, simply put, just because of historical precedent that Seagate's the leader in that space.

Robert Blair

Cindy Shaw.

Cindy Shaw - DISCERN Investment Analytics, Inc

Cindy Shaw with Discern. I have 2 questions about the high end and the low end. One, in the pure SSD market, not the hybrids, what do you view as your strengths versus the non-HDD players, say Toshiba -- or not Toshiba but not the Toshiba but somebody who’s going to...

Stephen D. Milligan

Factor somebody.

Cindy Shaw - DISCERN Investment Analytics, Inc

Yes, exactly. And what do you think they would view as their competitive advantages versus Western Digital. And then I have a follow-up question as well.

Stephen D. Milligan

Okay. So Mike or Currie? Oh, Mike, you can do it.

Michael D. Cordano

Yes. Let me start and Currie, you can fill in wherever -- where I miss something. I think what we see and Currie alluded to this, when you qualify a device into an enterprise system stack, there is a high degree of difficulty in passing the rigorous testing that they're doing. So one of the things we have, we have a very mature SAS stack here. It's been deployed on our hard drive, and now we're deploying it on SSD. That is a strong competitive advantage. And frankly, one of the reasons Intel wanted to work with us, right? Because it's not something you can just show up on that Monday morning and deploy it in a reliable way that works in the enterprise. So our ability and our experience executing successfully in those kinds of applications with HDD is directly applicable to SSD. To take the counter-side of that relative to what are those guys going to do relative to us [ph], it's about innovation and being nimble, and that's something we have to continue to get better at because a lot of innovation happens in small companies. We're obviously going to continue to survey that landscape, but we have to generate that kind of innovation internally as well. So our ability to sort of sense where the market's going, understand what we need to do to bring value and then execute on it technologically is critical. And we're going to do that both through internal work, as well as continuing to survey the external environment. But that's really the, let's say, it's a non-vertically integrated competitor opportunity and we have to be mindful of that.

Currie Munce

Like I said, if the numbers are not -- the number of enterprise HDDs we sell is substantially larger so we had -- we'd leverage that infrastructure, that support, that learning, the ability to support the broad range of customers, all the unique software features we've developed over decades of experience and we immediately bring to the table and don’t have to train them how to do it. So I think it's really the support and the service of that product and understanding reliability and understanding how they measure performance and how they do things.

Michael D. Cordano

And I guess lastly, don't discount market access. We have all those relationships set up. We get first -- the first visibility within them. When you're a small emerging company, I know because I happen to have one in my past, it is very difficult to get the same kind of market access that we have, so don't underestimate that as well.

Cindy Shaw - DISCERN Investment Analytics, Inc

The follow-up relates to the cold storage. We've heard a lot about aerial density. I haven't heard a thing today about access density. And I'm wondering what these cold storage applications are, whereas you put more and more capacity on a drive, access density has been declining. Is that going to continue declining? Does it matter for these applications, and how is that going to play out?

Currie Munce

Okay. So I mean, I think the real key is -- Mike showed sort of the model there, so you're going to tier for that access density. So if you look at that long-tail accruing asset, you're going to trade off in access density for the $1 per gigabyte capacity growth. So what you're saying is, “I have posted my information on Facebook. It's now 2 years old. I still want to find it, but I'm not going to pay the high cost to keep it on the high access density. I'm going to move it to lower cost, lower access density. So it's just another tier on how I get that information,” and you want 100 millisecond on the information you just posted, you may be willing to wait a second on something you posted 2 years ago.

Robert Blair

[indiscernible] John, from my right.

John Phillip Ragard - Friess Associates, LLC

John Ragard, Friess Associates. Several presenters today have talked about storages -- or I'm sorry, software as being an area that further advancements need to be made. Could you be a little bit more granular? And what other kinds of software advancements that need to be made to really be a winner in the storage systems business in the future?

Stephen D. Milligan

So Mike or Currie or Bill, you want to start with that?

Michael D. Cordano

Sure. [indiscernible] in the hybrid side of this is, as I discussed, is this great opportunity to be able to not just operate at the device level and try to determine what the user is trying to do by sitting there at the block level, but to move further up into the stack to really understand what the data structure's doing, what the OS is doing, what the file system is doing, and then it'd be able to make much more intelligent choices about how you store information within the tiers that you have available to you. Like Currie mentioned, hard drives are queued in the past with DRAM, and rotating media in the future, it might be DRAM flash. Traditional rotating media, it may be single partition, which is even a higher density type of a partition. So again, the more that we know about how the data is being used at the highest levels, the better the job we can do of deciding where that information should be stored and accessed at the device level.

Currie Munce

And I think we look at -- in the enterprise, it's a very similar story. It's a mechanical device. It has certain physics associated with, which we understand very well, and actually inside the drive, we know exactly where the actuator is and where the disc is, et cetera. The OS knows what data it's trying to achieve, what its application, what's the end-user requirements. If those 2 can be better married together and better understood, we'll produce a better overall optimal solution. So historically, we've been sort of pushed towards create a standard interface, standard commands, everyone does the same thing and we’ll push it down to you. What we're finding is we're continuing to challenge on access density, on performance, on capacity. As we talked about some of the technology, Bill talked about a new pushing-your-old-density. We're giving unique attributes. And as we realize we can reach further up into the stack, we can have more influence and get information. We can get more hints about what is the fast data that needs to be moved up because the application knows that better than the storage device knows that. So the storage device knows what its relative performances and capabilities are. And as we can marry those 2 together, then we will have a better optimal solution. And what we find really today is that, as Mike mentioned, in the cloud, where the -- in a more traditional sense, those are very -- systems have taken decades to develop. We say the cloud guys control the whole stack and they move very quickly. So we're seeing a lot more opportunity, a lot more openness to innovate with the cloud providers because they can open up a lot more hopes to us, and they'll turn the whole system over in 2 years or something to implement these things into our drives.

Robert Blair

Rob Cihra.

Robert Cihra - Evercore Partners Inc., Research Division

Rob Chira, Evercore Partners. Two questions, if that's all right. The 1 -- I think with the -- your new drives or new technology, you talked about the 5-millimeter hybrid and the helium 7 disc that you didn't say when those are going to ship, and I know you're not going to, but I mean, are those like months from now or years from now? I mean, just kind of some sort of ballpark. And then Mike, within your cloud presentation, you gave that thing that we've all seen in terms of flash the top, 10K or 15K in the middle and then, so the 200 underneath, and the -- I don't know whether it's a debate or an agreement, it depends on who you talk to. People talk about the 10K and the 15K in particular, getting squeezed out of that model. A, do you agree with that and B, do you really care, or does it have implications from a financial model standpoint, a competitive standpoint?

Currie Munce

Sure. So we did announce availability. So I think relative to the 7-disc drive, that the platform we announced, we talked about shipping something in calendar '13 at some point, mid-second half of the year.

Wolfgang U. Nickl

And then in the 5-millimeter hybrid, we announced -- or we are having engineering samples right now, call samples, in the first half of 2013 and shipments in the second half of 2013.

Michael D. Cordano

And then the second one, the curing is very interesting because everybody's got a unique fuse. So all these guys we're talking to, you hear one person that might say, it's going to be SSD at the high-end for performance and everything else is going to go on high-capacity storage. The reality is, is what's happening is, there's increasing fragmentation and differentiation. You have different people trying to solve it in different ways. And I think what's going to happen is you're going to see a varied approach to it. In fact, we're getting more calls for high-performance disk space product than we had 1 year or 2 years ago. So that's some of the trends that are happening. So I don't think there's a model that says SSD plus FAT drives and all that stuff in the middle goes away. I think it's just not going to happen. I think there may be some people that do that for a particular deployment. But when we look at the aggregate market, we're going to see opportunities to help our partners optimize, and you do that with a range of products. I think, when I talked about it today, these guys talked about it from a technology perspective, the more we can differentiate, we can tune our approach to enable an optimized solution and that's one of the things we're learning as we have deeper engagements with our -- with these end-user customers. It's no longer -- we're no longer distracted from what is really trying to be done. We understand it well and we're learning. And we're now able to innovate and provide unique benefits, where perhaps in the past, it was more difficult for us as hard drive participants.

Robert Blair

Sherri Scribner.

Sherri Scribner - Deutsche Bank AG, Research Division

Sherri Scribner from Deutsche Bank. I just had a quick clarification and a follow-up. In terms of the slides, you showed -- Steve you showed some slides with the capital allocation and you showed research and development going up from about a quarter and the future to being greater than a quarter of the capital allocation. Is that as a result of the combination of Western Digital and Hitachi? Or is that consciously increasing the R&D investments?

Stephen D. Milligan

Well, I think that as I indicated, those are kind of indicative, I wouldn't really focus on the slides. So I think it's just -- I don't know if it's actually bigger or not, in terms of -- but I think that it was not meant to be to scale, is basically what I'm saying.

Sherri Scribner - Deutsche Bank AG, Research Division

Okay. And then just thinking about the...

Stephen D. Milligan

But we are allocating increasing amounts to R&D. Right? We have over time. I mean, as Bill's chart indicated and we're going to continue to do that. And then later on this afternoon, Wolfgang will be talking about our financial model and how that kind of fits into that envelope. But clearly, as we’ve moved into more complexity from a product perspective, all the things that we're talking about, we're going to continue to make sure that we're investing appropriate amounts in R&D.

Sherri Scribner - Deutsche Bank AG, Research Division

Okay. That actually leads me into my second question. One of the previous questioners asked something similar, but as you increase that investment in R&D, how does that next shift of investment in hardware and advancing aerial density change to more investment in software to enable many of these applications?

Stephen D. Milligan

Yes. I mean, I don’t if Currie or Bill have any sense for that. I think that for the time being, I mean, that the -- it's still a relatively small amount of our investment, but obviously it's pretty leveraged to the extent that we're able to be successful. I don't think it really changes the needle in terms of that investment level significantly.

Currie Munce

Our vast investment is really pushing density but the growth -- some of the growth is going there but it's a small piece, but it's growing faster than we're putting into density.

Robert Blair

Steve Fox.

Steven Bryant Fox - Cross Research LLC

Steve Fox with Cross Research. Mike, just going back on one of your slides. You talked about how the level of percentage of data created is in fact up significantly. I think you said 3% and maybe if we could capture another 3%, it will be significant to our revenues. Can you just sort of talk about why now that's going to start to happen? What is the business case for seeing more data wind up that's created by the consumer backed up? And then if maybe, Steve, if you could tie that into the 9% growth rate you talked about for enterprise? It seems to me like if that does happen it would be favoring your most -- the most valuable portion of your portfolio. So what kind of revenue growth could you get out of your enterprise portfolio and see if these trends do start to really kick off?

Michael D. Cordano

Well, let me talk about the why, and why that's interesting. Because today it's not being stored, so why bother, right? And that's the question. Why bother, really goes back to sort of big data concepts and what a number of people in the ecosystem believe is, the more information I have, the more analytical sort of output I have and understanding the data, I can generate value with that. So today, there's a vast amount of unstored and unanalyzed data. So the more of that I can store and analyze and generate value add out of. It's a value creation opportunity. So those 2 things are very tightly connected. And then what the other attribute that's got to be enabled is the cost point that makes it reasonable, right? And that's all around cold storage and some of the things we've talked about. So you combine those 2 things, now you've created an intersection that is interesting and potentially very, very valuable. And I think candidly speaking, and Steve can talk to this, that's not included in one of the forward-looking view of enterprise. If we are successful doing this, that's greenfield space for us because it's just not yet been delivered. And like most prognostication, until we prove that we can do it, it's not in the forward-looking projections.

Stephen D. Milligan

Yes. So I think that what I prefer doing is kind of deferring a little bit more of financially specific stuff until the afternoon when we get into that. But I think that if you look at the enterprise space, I mean, obviously that is a more leveraged area for us from a financial perspective, right, whether that be in terms of revenue or profits. And so growth in that space is all good for us in a number of different ways.

Robert Blair

Actually, I’ve a question via text from Mark Miller of Noble Financial, who wanted to join us but wasn't able to. He asks, how do you keep the helium in the box and does this increase the cost? Does this increase the cost of the drive over one not using helium?

Currie Munce

The first thing is to say is that we believe the value and what we've created is how to keep it in and so we're not talking about that at this point in time. We’ll talk about it our closer to actual shipments of the drive. That is the key challenging innovation in making that. Yes, there is a small cost adder, but it's not that significant. But there is a cost added into it.

Michael D. Cordano

Yes, but that cost adder is insignificant when you relate it to the benefits that Currie had on his slide. So you got to think about it in a total cost level. That's the way we did it.

Robert Blair

Jayson Noland.

Jayson Noland - Robert W. Baird & Co. Incorporated, Research Division

Jayson Noland with Baird. I wanted to ask about HDD technology, also the bits of change has picked up a lot longitudinal for decades, PMR in 2007 and Hammers around the corner and discounted media after that. Are some of these technologies like Hammer being jointly pursued by the industry through an organization like ASTC? Or would we expect to see a race to Hammer? And I had a follow-up question for Steve.

Robert Blair

So Currie or Bill?

William Cain

Sure. And so you mentioned ASTC and that organization is working. A matter of fact, we've got a workshop next week in Monterey discussing the technology roadmap for the hard drive industry. And so yes, I do think that there is a lot of collaborative works, sponsoring a University research and third-party research in those areas to enable it. But I think this is a competitive industry. There’s always a race to be able to deliver the next technology. And I think the good news is now it seems as an industry, we're all pretty focused on what that next technology is. Go back a couple of years, I think there was a lot more uncertainty about what that would be. So I look at that as a positive. And I think that as an industry, we're making good progress towards that next technology. And so I think that's been kind of the timing, that we showed you is consistent with the message you get from the rest of the industry. And you want to add anything to that, Currie?

Currie Munce

No. I think the industry is trying to be more collaborative. We are trying to do -- sharing more and that was the -- Bill and I are 2 of the founding members of creating ASTC. And so the couriers are intend to leverage that more, but we are -- still remain very competitive in working on a piece of that individually.

Stephen D. Milligan

So does anybody want to talk about that some more? Currie? We already mentioned that. We don’t want to comment on that.

Currie Munce

Obviously, you mentioned perpendicular, I mean, we see it’s the next technology. It was on Bill's chart as being shingled magnetic recordings or shingled [indiscernible] to magnetic recordings. So it comes down to the fact that being able to make the media writable, I talked about the Rightfield's [ph] coming out, how they get even a stronger field coming up. So we think that in general is also the next technology coming out in the year or so here. So that will come out before heat assistance. There's been some discussions on the interface although that's -- again, still time to emerge in terms of how it goes forward. So we see that really as more of conventional extension of PMR before we see the Hammer technology coming.

Jayson Noland - Robert W. Baird & Co. Incorporated, Research Division

The PCIe as I -- just to clarify, that was -- that pursuit, that roadmap is internal development?

Stephen D. Milligan

We're not commenting on our approach.

Robert Blair

Ananda [indiscernible] to the left.

Ananda Baruah - Brean Murray, Carret & Co., LLC, Research Division

Ananda Baruah, Brean Murray. I wanted to ask what your view is on, I guess, the cost curves going forward for SSD versus HDD. Do you see anything changing and what might some of the more important influences be of those cost curves?

Michael D. Cordano

We specifically -- we showed you in the presentation in Currie's setup that the cost of the technology scales is the density scales. So since we're showing a forward look of density scaling at the same rate, we expect the relative offset in dollar per gigabyte to come down at the same rate for the 2 technologies.

Robert Blair

Shebly, up here in my right.

Shebly Seyrafi - FBN Securities, Inc., Research Division

I have a follow-up question. You talked 3% CAGR going forward. You also talked about having considered up-and-down cases. Can you talk about the up-and-down cases? And what are the key assumptions that differentiate your up-and-down cases? And also, what's your thought on the probability? Is this something -- is 3% highly likely? Or it's a very wide range of probability?

Stephen D. Milligan

Yes, I'll attempt to answer that and then you guys can kind of chime in. I think that when we looked at different cases, I mean, there were some cases where we looked at moderate negative growth. And obviously there you would see more movement to the cloud, more people, slower refresh cycle, all the things, all the trends that we all know about, but a greater or increased acceleration in terms of that kind of a trend. And so some shrinkage in the market, not dramatic, but that was that. And then a bull case scenario, I mean, that's a little bit more difficult to call in terms of how that might -- how much that might increase. I mean, the growth rate that we're looking in, in terms of 3% is significantly lower than what we've historically seen. So I don't think that we would see a bull case that would be an access of what we've historically seen, which has typically been what, 7% to 10% or something in that area? But now back to your question on probability, that's always a difficult thing to kind of assess. But obviously, the 3% case that we put forward is something that we feel pretty confident about. To the extent that you can feel confident with the forecast, right? Because the only thing we know about a forecast is it's not going be exactly right. But we wouldn't put that number forward unless we have sufficient confidence in it, or the analytical underpinnings that came up with it. And we also -- and there's an important element to that -- sorry, but there's an important element because I'm looking at important scenarios. We also evaluated what that might mean to our business, right? And what choices might we have to make with that. So we don't know what the growth rate is going to be. I mean, let's be honest, I mean, you don't really know. But the important thing is to assess the different scenarios and know what levers or decisions you would need to make from a business perspective accordingly. So that's the value of that scenario planning.

Robert Blair

Sherri, to your right.

Bill C. Shope - Goldman Sachs Group Inc., Research Division

Bill Shope, Goldman Sachs. Going back to the SSD versus HDD cost curve you talked about. Are there any disruptive events or technology that you looked at over time that would change that dynamic? I recognize that historically, it's been a pretty constant ratio -- or a pretty constant directionality. But some of the flash guys will talk about things in the enterprise, like primary de-duplication being easier to do on flash and that narrowing the cost differential. I'm assuming there's some other factors that would shift it in favor of HDDs. How do we think about some of these outside factors that could pop over time that would tilt it in either direction.

Currie Munce

So I mean, I think, yes. And a lot of people have gone out and talked and de-dup and compression. The reality is there's no reason you can't do de-dup in compression on HDD. You might cache it. You might have small amount of NAND upfront to do that caching, but you just use that as a cache and put it on HDD. So I generally believe that counting on a 4x reduction with some people, sort of trying Code 2 [ph] compression and de-dup is not a very good assumption. We've also seen people doing compression. You see great variability in performance and workloads, also with compressions fully [ph]. MPEG, your movies, your pictures are already compressed. They won't compress further. So if you're out doing social media or something, you’re going to not find much compression. If your data’s encrypted to protect it, it won't compress, by the properties of compression. So there's a lot of things that won't compress at those rates and then people run into big variability and that's the other thing about not only I want high-performance, but when I want high-performance, I want very predictable performance also. So as you move up to higher performance, you have to get more predictable, and these things tend to get less predictable. And we see the big opportunity is really in staying in the cloud. We see people actually replicating rather they de-duping as they go out and they spread the information as they scale. So those will be a factor, but I think there's no reason we can't do it on HDD. You may use a flash cache to do -- to enable some of that, but then your primary, you're going to put that back to an HD anyway.

Robert Blair

[indiscernible] in the second round of the questions.

Unknown Analyst

My name is Mark Celobimo [ph] with Teller House [ph]. You alluded a little bit to replication and de-duping in your slides as well. There was a theory out there, or maybe it's a misconception that if say, Amazon has a copy of an MP3 and iTunes has a copy of the same MP3, that they would choose to, in the future, may be de-duplicate that and consolidate the storage. How do you see -- how do you think about that? And that would potentially drive lower demand for storage. How do you think about that trend? Does that trend exist? And how does a replication counteract that? Is that a greater force and then I have a follow-up.

Michael D. Cordano

Let me start, and maybe you can finish. So I think when you think of quality of service, to believe that there'll be a single instance of any object in the Web in some theoretical state and we can get the performance and quality of service against that, it's just not feasible. So we actually see the trend going the other way because you need to be closer to the content. And so one of the reasons replication happens is for data protection. But the other reason and probably maybe even a more pronounced reason for the numbers and copies is quality of service. So -- and I also don’t see competitors in the marketplace as disaggregating and cooperating in that way. But technologically, I'll let Currie talk to it, but no question, we see the trend more strongly in the other direction for the reasons I just stated.

Currie Munce

Yes, exactly as Mike said. The only thing I'll say is you did touch upon MP3, so you are talking about digital rights management. And so there are content protection concerns or issues with that. So many people who own content don't want to have that de-dup going on because they're concerned about who owns what and how it's being shared and move around. Now in the particular case of Apple, I believe they have special agreements with all the content owners, so they are able to keep 1 library in source multiple, but they require a special license agreement from all those content owners to do that. So it's not something that can be easily -- or concerns about replicating that everywhere when you talk about the license content.

Unknown Analyst

And if I could do a follow-up? You talked about understanding the needs of your end-users, for example, understanding the needs of a social-media user like Facebook. Have you looked into -- and I'm sure you have, but could you talk a little about the medical end-user? There's a lot of DNA sequencing that's going to happen. Digital medical records are going to start being produced. These have to be stored, and you alluded to regulatory reasons why these things have to be stored and maintained for several years, high-resolution MRI scans, et cetera. In theory, these would drive tremendous demand for data storage. Are you looking at that end market? Are you working with those end-users? Or is that something that a company like EMC would do instead?

Michael D. Cordano

Well, so when I talked about the cloud build out, so you think about sort of the group of companies, which are recognizable brands that identified as people you build with. In many cases, those are the people that will be servicing or serving that. So specific vertical applications like you're talking about is an area for further development. You saw that on my slide. We're getting them more aware. We are certainly many steps closer to that market than we were in previous generations, but we're not as embedded and as understanding of that as we want to be in the future. But you're right, those are all opportunity areas for us to continue to understand more deeply. But what we do have is a better connection to is the underlying data center infrastructure that's going to service those. So we're 1 or 2 steps closer to it than we were previously. But it's certainly an area for further investment and understanding on our part.

Robert Blair

Joe Yoo. [indiscernible] This will be our last question of the morning, then we will break after we respond to this question.

Joe Yoo - Citigroup Inc, Research Division

Joe Yoo from Citi. I wanted to go back to the hybrid question. Just if you could comment on what the level of interest is from the OEMs on adopting a hybrid since it is an amazing market. It has -- the adoption has been pretty poor in the past. Also, where you're expectations on when that will inflect and take off? And most importantly, your ability to compete at that market since you are a little bit behind your competition.

Wolfgang U. Nickl

First of all, I would challenge that assumption that we're behind the competition because I believe that’s a different type of device and it's a thinner, lighter, faster device. We are generating a tremendous amount of interest from the PC OEMs and ODMs and we had a collaboration from Asus and Acer on the demonstration today. We are -- we believe that there will be hyper drives in drives that's CES for demonstration purposes. And the ODMs, in particular Quanta, is doing a lot of work and breakthrough work, technology-wise, in order to be able to evangelize the product and they're very excited by the prospects. And so, we believe that it's the high-volume in the December -- in the holiday season for 2013. Our 7-millimeter is already high-volume in -- for the current holiday season and we expect that the 5-millimeter and the hybrid will be in high-volume in the holiday season in 2013.

Robert Blair

Okay. Thank you very much for your questions. We will turn now for demos and lunch. If your badge is green, go to the demos. If it's red, enjoy your lunch, and we will be back here to resume in 2 hours. Thank you.

Robert Blair

Okay. I think we're ready to get back underway. And so this afternoon we have a presentation by Wolfgang Nickl, our Chief Financial Officer. And after Wolfgang's presentation, we'll then have a concluding Q&A session with John Coyne, Steve Milligan and Wolfgang, and then some closing remarks by John.

So before Wolfgang gets underway, just trying to point out that this artwork in front of you contains a disclaimer of our forward-looking statements Wolfgang will be making in his presentation and comments today. Forward-looking statements will be made concerning growth opportunities in the storage industry, our expected financial results, business model and financial goals and our capital allocation strategy. These forward-looking statements are based on current expectations and are subject to risks and uncertainties that could cause actual results to differ materially, including those listed in our report on Form 10-K filed with the SEC on August 20, 2012. We undertake no obligation to update our forward-looking statements to reflect new information or events.

And in addition to that, we -- this references the non-GAAP financial information that we are making available today. Reconciliations of the differences between the historical non-GAAP measures that we provide during Wolfgang's presentation are included at the end of our presentation and will also be posted along with the presentation in the Investor Relations section of our website. Forward-looking non-GAAP information will be -- that we provide during Wolfgang's presentation excludes items that are not fully known to us at this time and therefore, we are unable to provide a reconciliation to the most directly comparable GAAP financial measures for this information.

And I will now turn the mic over to Wolfgang Nickl, Chief Financial Officer of Western Digital.

Wolfgang U. Nickl

Thanks, Bob. Good afternoon. I'm going to cover in my 1/2 hour several topics. I'll take a quick look back to the last 10 years. John and Steve have both commented on it. I'll give a little bit more color on what the strategy was and how it played out. I will then go into our forecast for fiscal year '13. I will give an update on the current quarter during that section of my presentation. Afterwards, I'll give you a little bit more insight in how we think -- how I see an economic profit is driving our operations and our strategy. You saw it in Steve's presentation before, and it's also the thing that drives our business model and our capital allocation strategy, and I will, of course, finish up with those 2 items.

So let me get started with the last 10 years and provide you this sort of chart, which basically shows you the development of our free cash flow, and that's not a cumulative job. That's a chart by year, how the free cash flow developed for the company. A couple of interesting notes to this chart. First of all, the fiscal year '12 that you see that was the most successful year does only include 16.5 weeks of HGST. So you see how much better that business really was if you will look at the run rate of the last quarter. I also want to stress that we achieved a revenue, a compounded annual revenue growth rate of 19%. That generated this free cash flow, which was really a compounded annual growth rate of 50%.

The other important point on this chart, and you can't really see it, and that's the significant part of it is we weathered significant crisis over these 10 years, and the 2 to point: our fiscal year '09 included the financial crisis, and fiscal year -- calendar year '11 included the unfortunate earthquake and tsunami, as well as the flooding in Thailand. So what it tells you is, you’re looking at a company and the management team that's trying to be incredibly nimble and adapting very, very fast to whatever the market presents in terms of challenges.

The strategy of the last 10 years can simply be explained by a strategy of cost leadership and diversification. And one way how we did this was through acquisitions. We did 2 types of acquisitions. We did do -- those that gave us cost leadership and technological capability that would be the Read-Rite and the Komags [ph], Hoya, and we bought the hard disk control group of STMicro, for instance. But then, we did the strategic acquisitions that gave us portfolio and technology leadership. And in that category we'll have -- we had SanDisk, which forms the basis of the Connected Life the most that you have seen, Silicon Systems that gave us our first step into the SSD world and then of course, the capstone was the HGST acquisition that rounded out our portfolio and gave us that platform that we think is so powerful for the next 10 years to come. While we were doing this we continuously expanded our gross margin. The gross margin essentially doubled, and that's a function of vertical integration, cost leadership, volume effect and the diversification of our business.

Taking another look at this and how -- and this is more an external view, how was it accepted in the market, what we had to offer, in particular, how was the portfolio accepted. And what I'm going to do is I contrast the first quarter with the last quarter of the 10-year history in the 2 charts here, and on the first click, you'll see the TAM. A couple of observations here. First of all, the TAM tripled. So the TAM has grown very, very significantly. I think the CAGR was about 10% or 11% over the last 10 years. The second thing is there were significant moves in how the TAM is made up. You clearly see the mobility theme, and John alluded in his presentation that we helped drive that with providing disk drives for laptops at a very acceptable price point. But there are markets that were created during that time frame, and we played a big role in one way or the other. For instance, the column called Branded, those are our external and network-attached drives. That market didn't exist 10 years ago. It’s a very good market for us now. Another example would be enterprise. Enterprise, 10 years ago, there wasn't a set, a high-capacity drive there and the market created that, and both HGST and WD were very instrumental in getting that job done.

On the second click, you see how our volume developed. On the left chart, you can see we're basically a desktop company 10 years ago. And we saw that trend that desktop wouldn't grow as much as the rest of the market, and we diversified through organic development and through acquisitions. And when you go to the right side of the chart, you see that we have about equal share in every segment of the market. Over 10 years, we almost tripled our share as well. We started out at 16% then we ended at 45%. We believe that today we have the broadest product portfolio in the market, and we believe we have the capability to make markets in the future as well, and you have seen that in the demos and the presentations by Tim and Mike today.

Steve mentioned that we're a great capital allocator over the last 10 years, and I'll show you later. We intend to be a good capital allocator over the next 10 years as well, but the strategy has also worked out for our shareholders.

On this chart, I simply plotted the 10-year stock price development. And in that sense, you can say that the strategy has worked out for the shareholder as well. But the CAGR of the stock price in the study period is 25%. The red line at the bottom is the S&P 500. And that number is essentially flat. Now we can argue that there was a 2% dividend yield there on average. But I think the company's strategy has paid out for the customer suppliers, employees and shareholders over the last 10 years. The one thing I'm going to mention here, and I'll get back to later on is while this seems impressive, we are not satisfied with it yet because if you link the cash flow job that I showed you and the stock price job, you conclude and everybody in this room knows this that the multiple of our free cash flow is somewhere between 4 and 5. The multiple of free cash flow for the red line is somewhere around 10-plus. So we certainly believe that there is a lot of room when we look at our intrinsic value and the value we're traded at right now.

So that was the look back. And I'm now going to look forward, fiscal year '13. As you know, we established the $10 EPS target on a non-GAAP basis on our last call in July. That is still our target. It's clearly still our target. I just want to highlight 2 or 3 things. Number one, again, where the last year included 16.5 weeks of HGST contribution, we're now looking at a year -- a full year of HGST contribution. And you can see that in the ladder on the top, that means more revenue expressed in volume and mix, and it also means more OpEx with a significant net positive. The other thing that you see from this chart is that this means that we'll continue to share, and it allows for us to continue to share value creation with customers. You see there's significant opportunity forecast declines, which will put us in a position to share that value with the customer.

Let me now talk about the current quarter. Right after this is done -- this presentation is done, we'll issue an 8-K and give an update on the current quarter. It will not come as a surprise to you from reading what others have to say, Intel and the Research Institute [ph], that demand is very muted this quarter. We're also seeing a significant amount of inventory rebalancing throughout the supply chain, and we have changed our view on the total available market.

We now expect the total available market for hard disk drives in the September quarter to be 140 million units. That is down from 157 that we guided to in the July earnings call. As a result, our revenue will be lower assuming flat share. And our new revenue range is $3.9 billion to $4.0 billion. However, and, here’s the important part, we expect that our gross margin on a non-GAAP basis will continue to be at 30% like we guided to in the July call.

So to express it again, the market is what the market is. We look at the knobs we turn. We can't make the market by lowering the price. There is no elasticity there. So our job is to adapt fast. Get the gross margin locked in where we need it to be. And you’ve seen the box at the bottom of the chart, we’re focusing on what we can influence directly. We're continuing the investments in all the areas that were mentioned this morning. We're very aggressively working on costs. I won't go through all the details, but we told you in July that there’s significant cost opportunity where we're still dealing with the aftermath of the flood. We've always been pretty real-time in looking at our capital. And we're implementing our capital allocation strategy that I'll present in a few minutes.

Let me now move on in how we look at the business. Our ROIC for many years has been the guiding metric for us and will be the guiding metric for us for the time to come. Basically, you have seen this chart over and over again. We used ROIC to tailor all our strategic plans and our operational plans, and we'll also use it to express our business model. And basically on the y-axis, you see gross margin and OpEx and tax rate expressed as operating income. And on the x-axis you see the invested capital. Also later in the business model, you’ll see cash conversion cycle and CapEx. That represents this. Of course, in different curves, we'll present our cost of capital and our hurdle rates.

The first point I want to make here is I've been here for 17 years and at least for this time, but I know for the last 40 years since the company existed it's always been a deep heritage of the company to be -- try to be excellent at cash management, and I think that's also expressed in how we look at it from here.

I want to make a very important point here. There is no ideal point on this map other than you want to be northeast of the hurdle rate. That means where you're earning more than your cost of capital. But with this first build, I'm showing you that there's different ways to get there. We plotted a semiconductor company. It's actually a mix of 2 or 3 semiconductor companies, and we plotted the distributor at the bottom right. You all know that there's different ways to get there. If you have slow asset turns, you need high operating income, high gross margins. So if you have fast asset turns, you can live on a lower operating margin. And why is that important for us? It's important for us, because we have different business segments. We have an enterprise business, we have a branded business, we have an SSD business, and while that's a lot of the dots for WDC or Western Digital, each one of our businesses has a different positions. And it's important for us to follow as who end the new business and expand businesses because it will change the positioning of WD over time.

The yellow arrow with the 2 tips that I put in there is a hint towards a new metric that we're using in how we are assessing the business. Steve had the job on economic profits there. Of course, economic profit, in our view, is the same thing as ROIC. It's just expressing the difference between the cost of capital and your returns in dollar values. So what you'll see in the future is when you study our proxies, you'll see that, that metric will play a part in executive compensation. So it's another expression that capital allocation, capital utilization are important for the company.

I won't go through the builds on the left side. We use these metrics for value-based management. It helps us to find the levels in the business that create value over time. On the right, I listed many of the actions we are taking to walk on all the levels we have. I'll just mention a few. Revenue, we're investing aggressively in low-profile mobility and cold storage, like we heard this morning. We're also looking at adjacent markets. We haven't talked much about our SSD effort, but we're investing in SSD very heavily. On the cost side, primary focus chart #1 is to get close to pre-flood costs, a lot of work there to do with utilization. It has to do with ahead media sourcing. Fixed capital, we will have the same discipline in deploying capital in a disciplined way there, and CapEx cash conversion side is something that we've always done at or better than our model.

On the WACC, on the weighted average cost of capital, you will see us exploring levels there as well with unused debt capacity, and we're looking at how we can bring the weighted average cost of capital down there. So the message for this chart, ROIC, it's not just a metric that we plot on a chart. It's actually translating into a real management action in our value-based management approach.

Now onto the third chart that you have been waiting for, our new business model. If it works. 27% to 32% is the gross margin level that, A, we think we can achieve; and, B, we think we need to get a return on the investments we make in both technology and capital. There's a range. The range is there because at times you'd see different levels of utilization, and you'll see different segment mix. I told you our businesses have different gross margin profiles and there’s different seasonalities, that's why we have the range on the gross margin level. The question this morning, I believe it was from Sherri whether we continue to invest. Yes, we will. 10% to 12% is the new OpEx range. We used to be 9% to 10%, and that reflects our belief in market opportunities, and we're heavily investing in R&D like we've done in the last 10 years. As a reminder, more than 2/3 of our OpEx goes into R&D.

On the tax rate, when we announced the acquisition a year ago or more than a year ago, we indicated the 7% to 10% rate, which has really to do with the geographic generation of our profit. We still think that's the appropriate rate for now. I'm making an important point here. That is our midterm model. Like other successful, high-tech companies with a low effective tax rate, our profits and our cash accumulate offshore. This number doesn't -- does not include any repatriations, and we will not know how the laws will pan out after the election. But at this point, there is no repatriation planned in this business model.

On CapEx, Steve showed you the 3% assumption and the market gross rate, so more moderate volume growth in the market. Therefore, our CapEx model is lower, albeit the range we extended to 2%, and that's purely to enable investments in technology. You've heard us talk about 8 inch in the past. The technologists this morning talked about Emo [ph] and other things. So we have built in that flexibility in the model. And lastly, we're going to continue to focus on our cash conversion cycle to make sure that our working capital is in best possible shape.

Moving on to capital allocation. We spent the better part of the last 6 months studying this. We looked at many, many, many financial considerations. We got advice from many people including our shareholders. We considered, amongst others, our minimum cash requirements. That's our minimum cash requirements to operate our business. And a fund for contingencies and bolt-on acquisitions. We will continue to focus on CapEx and cash conversion cycle. We have very thoroughly studied our leverage. Last quarter, we end up -- we ended with a leverage of around 0.7. We will always be financially disciplined and conservative. We feel that our ceiling on leverage is investment grade, and our view is that, that's a 1.5x EBITDA.

If you just take the $10 target and if you triangulate into EBITDA, and you deduct the debt that we have on the balance sheet, which was $2.2 billion at the end of last quarter, you'll conclude that our debt capacity is somewhere in the $3.5 billion range, and we would still be investment-grade. That is a factor that clearly went into our deliberations. We have been successful with our acquisitions. We made like 6, 7 acquisitions over the last 10 years, and we're proud to say that they were successful. We continue to look at bolt-on acquisition, technology acquisitions. And we intend to keep it the way that they’re successful. We have established over the year a very rigorous DCS phased process in how we look at acquisitions, and we'll have that discipline embedded in what we go to forward as well.

I already mentioned the executive compensation that is more and more aligned to economic profit, which is an important consideration as well. The little chart on the bottom shows the result of our deliberations on the form, how we would give money back to the shareholders. And this comes back to the valuation. We run a DCS valuation on our own company and even with the gross rates that are much more modest, as Steve showed at a 3% market, we believe the company’s significantly undervalued at 4.5x free cash flow multiple. We want to put our money where our mouth is and as long as that is the case, we would prefer share buybacks over dividends. And that was an important decision when we finalized our capital allocation strategy.

So now I won't keep you waiting anymore, and I'll just show you what it is. And that will be in the 8-K as well that will go out in the next 20 minutes or so. We're targeting to return 50% of the free cash flow to shareholders. That's a -- you do the math on $10 per share. That's a very significant number every year. Like I said, we're looking at our multiple every -- all the time, and we believe that we need to weigh it toward share buybacks. And our board today authorized an additional $1.5 billion in share buyback authorization. You might recall last quarter, we finished with an ending authorization of $1.3 billion. I can tell you we spent about $150 million quarter-to-date. And with the additional $1.5 billion, we have around about $2.7 billion left, which is 25% of the market capitalization today.

We do see the value of the dividend and therefore, we have decided to initiate a dividend, $0.25 per share on a quarterly basis. The 8-K will say that it will go to shareholders of record that’s of September 28, and that's the last day of our quarter. It's payable for the first time on October 15.

With this capital allocation strategy, we believe we have found the right balance between enabling the growth and returning shares -- cash to the shareholders. It will put us in a position to fund our R&D, fund bolt-on acquisitions, and we'll retain our flexibility going forward.

And with that, I would invite Steve and John back up for your questions.

Robert Blair

Okay. We're ready for your questions, and then followed by that, John Coyne will make some closing remarks. Ben Reitzes?

Benjamin A. Reitzes - Barclays Capital, Research Division

Self-duplication is a good policy. I wanted to ask you 2 questions. First of all, just any thoughts on the pace of the share repurchase, how you're going to think about that. It really hasn't been your policy to forecast that because you don't want the market to know, but anything in terms of a long-term pace. And then I just have a quick follow-up.

Wolfgang U. Nickl

I mean, you can do the math a little bit from the 50%. Maybe it will imply our free cash flow when you take the 50% then you deduct the dividend, you'll get pretty close. I think we mentioned it on the last call that, and we reconfirmed that after our last call. The research shows that you can't really play the market and that the returns for companies that have done repurchases was in the S&P 500. It's actually the highest when you do it pretty steadily, and I think that's what you can expect from us.

Benjamin A. Reitzes - Barclays Capital, Research Division

Okay. That's great. And then with regard to the gross margin forecast obviously 29% to 30% mid-range. Is there a certain minimum TAM in terms of at least the current business model, the current mix that you need to be at or certain unit volume in order to hit that like -- or how should we look at that, because obviously there'll be some scenarios like maybe this quarter where you don't have the units you want but your margins are great? So just thinking, is there a minimum to kind of hit that and stay in that range?

John F. Coyne

Well, I mean, there always a minimum, but I think one of the things that we've demonstrated over the years is the discipline we bring and the speed of response that we bring to attempting to recognize reality in the marketplace, and then attuning our manufacturing, our production levels to match what the market needs rather than trying to create a production plan, which is what the factory needs to be full. There's a big difference in approach. And I think we've also shown a very high degree of flexibility in our model. Our models are relatively -- we've only automated our business in areas that require it from a quality reliability perspective. And that leaves us with great flexibility for moving up and down around the projected demand line. So we're very quick to react on the upside. We're very quick to react on the downside. I think we have the most flexible model in the industry in that regard. So we feel pretty good in a reasonable range of outcomes, and I think earlier to a [indiscernible] question, Steve indicated that when we looked at our low-end model, we were in moderate negative growth or -- and then our base case model, which we presented today, is the 3% growth that we see in the – for the longer-term future. We do our best work when we're wrong to the downside and we get opportunity. And I think you can rely on the business to continue to manage conservatively so that we're not at risk and keep a bit of headroom over what is that breakeven point. I would not like to tell you what exactly that number is. You do know it.

Unknown Analyst

Perfect. I'll ask one question and one follow-up as well. First, I want to go to the business model. You guys outlined 10% to 12% OpEx for rev. You also threw in a comment about $550 million per quarter. If I look at the model, it seems like that, that looks a little bit off. So I'm trying to understand, is $550 million the absolute number that you're running through the model? Is it...

Wolfgang U. Nickl

I will clarify that. That's what we mentioned on the last call, and thanks for the question. For at least the next 2 years, we’re going to run at a $550 million rate, which is depending on the revenue outside of the model.

Unknown Analyst

So to be clear on that, you're not assuming any HGST synergies really at all?

Wolfgang U. Nickl

For the next 3 years at least.

Unknown Analyst

And then the follow-up question would be is that on the share repurchase, the 50% allocation that you're giving a free cash flow generation, just an update on where we stand on the arbitration situation with Seagate.

John F. Coyne

That's still being reviewed by the courts. There's no decision at this point. We don't know when they'll release that decision.

Unknown Analyst

Two questions. One is short term and then a long-term comment. Just in terms of the 140 TAM, can you just talk about where you're seeing the weakness? And in the inventory, just elaborate a little bit further on that. And then the second, in that same vein is just -- that’s shipped TAM, right, as opposed to what, in theory demand, all else being equal, would be. And so what do you think -- it’s a theoretical question, what do you think that really the industry, all else being equal, should be shipping in an average quarter? What is normal? At one point, the industry was shipping or targeting 175 million, 185 million and the world is still growing 40%, 50% for data, and so it’s like that. So do we need to get back there? And if it's that's not the case, where did all those units go?

Stephen D. Milligan

Yes, I think if you look at where there's weakness from a market perspective, it's probably easier to talk about where there is not weakness or at least where things are holding out reasonably well. It appears that the commercial side, particularly in the U.S., is holding out reasonably well in terms of the input that we get from our customers. Consumer side of the market is weak undoubtedly. Europe, there's weakness there, and also there's weakness in China. Those are the kind of main areas that we're seeing. And it is very difficult to get a sense for what true demand is because we've got multiple things going on in terms of the inventory rebalancing that Wolfgang mentioned. First off, given, if you want to call it, some of the overhang from the flood in Thailand, some of our customers purchased ahead in terms of their HDD requirements to make sure that they had sufficient supply to service the customers’ needs. We still are working through that. Additionally, you have customers that are planning for the introduction of Windows 8, and so they are working off their inventory of non-Windows 8 product. And so to attempt to answer your question as to what the true TAM might be, I'm not sure I have a real good answer for that. And so I don't know, John, do you have any better guess on that one than I do?

John F. Coyne

No, we think there's a significant component of inventory in the number -- in the 140. But we think that the inventory is not completely resolved until the -- I mean, there'll be some effect next quarter. But then I think we're into a more -- a clearer idea of what the true end market demand is doing as Windows 8 launches. Because the one very positive sign from everything we hear from customers is that the Windows 8 launch is going to be a chase-up launch because they've been dealing with pre-Windows 8 systems and pulling down, trying to bleed off that inventory. And they've been delaying the start on the building out the new Windows 8 and new ultra books formats. And so I think we're going to see a pretty clean to low inventory market in the next calendar year.

Wolfgang U. Nickl

George [ph], I wanted to add, one piece of inventory rebalancing that's not going to happen. And that has probably happened in prior quarters where the demand was weak. It’s not going to be a rebalancing from us into the channel. We're extremely disciplined in maintaining a very, very low weeks of inventory number in the channel, and it will stay that way throughout periods of demand weakness.

Stephen D. Milligan

Does he have a long-term question too? I hope that wasn't your long-term question. You might want to think of another one.

Unknown Analyst

Could you elaborate? We talked about units. You seemed to imply from the earlier chart that the capacity of the data storage is growing at much faster rate. Have you seen that happening? And what is the reason that you dropped the capital, and I think you mentioned the local [indiscernible] talked about volume and capacity that you don't need to do something to spend...

Stephen D. Milligan

I'll take the budget [ph] question, and…

Wolfgang U. Nickl

Before you do, I mean our revenue over the last 10 years was 19% in average. And that's -- we did this -- per year. And we did this with a 7% to 8% CapEx budget. The market is growing at 3%. Even if it's growing a little bit more, we won't invest the same amount in capital expansion. That's why we took it down. The reason why we took the range up is because we make technology investments at certain times, and we've got to have the flexibility. I'm going to make one additional comment to the OpEx comment that I made earlier. And that is -- we intend to be in our model most of the time, but for the time being, if you remember, we are still walking through the tail end of the flood recovery. So if you have a special circumstance, you might find a quarter when you're outside of the model. But basically, the volume translates into a lower run rate CapEx spend.

Stephen D. Milligan

Yes, and I think just to add a little bit of comment. I think on your first -- the first part of your question in terms of capacity and -- or what's happening with the mix, if I understand your question correctly. I mean if you look at, and Wolfgang will know the math better in terms of the actual numbers, but if you look at the TAM decline quarter-on-quarter, we're assuming that share is going to be flat. But then, you look at the forecasted revenue decline, our revenue is actually declining at a lower rate than the TAM is declining, which means that we're mixing up our business, right?

Unknown Analyst

Do you expect that trend to continue, is there a trend toward mixing up, and is the mix up more profitable for you as highest capacity discount. Are they also applicable?

Stephen D. Milligan

Well, generally speaking, I mean it should be more profitable yes, as a general rule, yes.

John F. Coyne

But I think to your point, also Mona [ph], the technical boy has made the point that aerial density is going to grow at 30%. And the exabyte growth that we showed is in the range of a little over 30%. And there are multiple inputs there. We're not the gurus of that number. There's a range from 30% to 45%. Our 3% unit growth would imply that, in our view, the data growth that is stored is somewhere in the 35% range. And with 30% aerial density growth, our units will therefore, go up in the 3 to 5% range.

Unknown Analyst

So we should -- when we're building the model, we shouldn't offset the units as much as looking at the overall, is that...

John F. Coyne

I think the fundamental is that -- then the other fundamental is turning those units into valuable -- the value of the units that we sell. And I think the revision to our model indicates the expectation that we will continue to deliver higher value add to the market than we have traditionally done in the standards based market. And that we will continue our efforts on our legacy of operational excellence, so that we keep -- we get more value at the customer level. We drive better cost in the areas within our control and by doing that, deliver a solid bottom line.

Stephen D. Milligan

Yes, and we consistently remind our employees and others that we pay them in dollars, right, not in units. And so... Harry Rosenberg [ph]?

Unknown Analyst

I think about a year ago before the floods hit, you had estimated that there was a total of 7 weeks of inventory throughout the supply chain with about 4 weeks in the channel, a couple of weeks in jet [ph] hubs and about a week at your -- on your balance sheet. Since then, depending on what you think true underlying demand is, whether it's 150 million TAM per quarter or not, it seems like, at least, 50 million or 60 million drives have been removed from the system over the past year. Could you try to give us a better sense for where those got sucked out of? I imagine a lot of it came out of the channel, and a lot of white box manufacturers have gotten less volumes than they used to get and so on. But maybe any kind of color you can give us on that would be helpful.

John F. Coyne

Yes, I think you're right in that regard. I think the area that suffered most from a post-flood recovery thing was actually the external retail sales, which very particularly in retail volumes are very highly correlated to price points. And immediately post-flood if we looked at allocating drives, we have to be careful to deliver on our responsibilities to our long-term system customers where it wasn't the issue of a $59 sale or a $99 sale at retail. That could have translated into a $5,000 sale for a particular system customer. So we had an obligation to support our system customers, and we did that at some expense to our own direct retail business from a supply perspective. And then because of the constrained supply there, 2 things happened. We moved up the price points that moved out of the volume envelope, and a lot of that business is driven by commitment to certain price points in volumes to stimulate traffic in store. With our allocation situation, we weren't able to commit to supply. And so the retailers moved promotion dollars -- their promotion dollars to other products, and it's taken some time to rebuild that. So that's where a chunk of the disappearing drives went. We don't believe that we've done any permanent damage to that marketplace. We continue to believe that there's massive opportunity in the Connected Life, personal cloud environment. But there -- we did knock a brief hold in it because of allocation in pricing in that market. On the retail -- or on the distribution side, similar thing. Again, pricing in distribution because of extreme allocation, pricing went higher than was typical in the OEM environment and that disadvantaged the white box guys. And there was a slowdown in the -- a chunk of that channel business also fuels small, regional retail players and that slowed down and replacement, which also slowed as people looked at, "Do I want to upgrade my capacity? Not at that price." And both things are coming back in line, and so I think we'll see some of that stabilize as we move forward here. But they're the, I think, the main reasons for what we're seeing in the last 6 to 9 months. And then, of course, overall there's a macroeconomic thing where growth rates in the go-go countries have been moderated, and there’s a particularly difficult situation in Europe. So all of those things, I think, have affected the underlying TAM. and then the specifics I mentioned have addressed the kind of short-term shock to the system.

Robert Blair

Nehal Chokshi, right to your left, Michelle.

Nehal Sushil Chokshi - Technology Insights Research LLC

Yes. I have 2 clarifications actually. Nehal Chokshi from Technology Insights, by the way. So the shortfall on the TAM there, can you just verify that -- is that all limited to a client space or is there also weakness in the mission-critical and airline and branded, can you just give that clarification?

Stephen D. Milligan

Yes, I mean, obviously, just given the magnitude of the variance most of it’s in client, but we're seeing weakness in really almost all the segments that we deal with right now.

Nehal Sushil Chokshi - Technology Insights Research LLC

Okay. And then my other question is a little bit more long-term related. You mentioned your 20% ROIC of hurdle rates. Now is that based on the past capital structure or is that based on the new capital structure? What will be now optimized in terms of...

Wolfgang U. Nickl

It's based for the future, and one of the things that you consider that it's a technical thing. But most of the time, people report on an ROIC based on book values of equity. I mean, but you really need to consider market values, and that's why you see that we set the hurdle rate that's higher than the WACC based on a book value, if that makes sense.

Unknown Analyst

The build down that 3% forward 3 -- 5 years-ish CAGR number, how does it translate into rapid growth? In other words, the ASP going forward, thinking the mix impact in your certain enterprise growing faster, personal storage has grown faster. And the fact you were talking about you're going to add more value into your product.

Wolfgang U. Nickl

Well, I'll take a stab at it. First of all, it helps to establish history again over the last 10 years. We grew the revenue by 11% -- I'm sorry, the volume by 11% every year, but we grow the revenue only by 2% or so every year. So as you can see from the gross margin model and as you can see from what Mike and Tim have shown this morning, we are also here trying to put more and more value in the solutions that we deliver. And I hope that we can translate into the appropriate ASPs, and our focus is really on maximizing the revenue based on the TAM that we have. So we don't know for sure but hopefully, it's higher than that gross rate.

Robert Blair

Michelle, we have a question up here. Just behind Vincent.

Unknown Analyst

Matthew Gozi [ph] from Carin [ph] Group. My question is really regarding the multiple around 4x. That's obviously an incredibly low multiple, so a couple of things there could be true. One is that the fact it could be very cheap. You guys are obviously honest about your view when you're buying back stock. It showed that you actually do believe that. The other explanation as well that's actually that ease for the cash flow is going to deteriorate significantly or go away, and we're dealing with kind of peaked cash flows, and the mid-cycle is probably much lower. That is probably the broad perception in the market. What do you think needs to be done on your part, or what kind of things do we need to see for that perception to change?

Wolfgang U. Nickl

I'll give you a financial perspective, and I’m sure Steve and John will chime in. But you're right, when we -- I mentioned that we run our own DCF models and the 3 things that [indiscernible] the valley in the DCF model. I mean, first of all, you want to have a business model that translates business into cash, so your ongoing ROIC and free cash flow. The second thing is your cost of capital, which for us mostly is equity-based. And depending on what side of the COE you are, you can lower it to an optimal point with some leverage, and we talked about that, and we're probably going to do that. And the third and probably the one that we have to do the most clarification in the market is the gross rate. And there was the 5-year gross rate. And I think we feel really, really good about that. I mean with 3%, there are scenarios around that, but we feel really good. But all of you do a DCFs all the time, the value’s in the terminal value. And what we have to do is to explain how we're growing beyond 2017. And my view in doing these models is twofold. Number one, if you listen to what Bill Cain said this morning is you've got to look at in 2020 for 2 die SSD solution, which was probably -- is probably the one that could either weigh the most from a TAM. What reliability do you get, and what cost point and price point do you get? And that's a long way to go because before you go to this, enough is enough. The second point is more important point is throughout the presentations, you've seen that we changed, and there's different elements of storage. I don't know how many engineers we have, 8,000 or 9,000 engineers, and we have shown that we adapt towards the market, and we continue to do that. We can't tell you exactly how that will look like in 2020, but we're in the storage business, and storage is relevant and also as the business.

Stephen D. Milligan

I think one of the other things that we talked about today, which I think is important, and we spent a lot of time talking about this is that because, and you guys ultimately will be "judge" of this, but we’re not stupid people, right? And we read everything as well, right? We read a lot of information. We read what you guys are saying. We talk to a lot of different people. And as I indicated earlier, we wanted to make sure that we took a very critical view with regards to how we viewed the future of our industry and the future of our company. And you're seeing the output from that analysis, right? And rightly or wrongly, I mean, obviously we have a view rightly or wrongly, we think that we continue to have, and I don't mean this defensively, but we continue to think that we've got an enormously strong future. And as demonstrated by the capital allocation decision, we have a view on where we think our valuation should be, and we've made our decisions accordingly. But it's not a matter of us just simply wishing that this is the conclusion because you guys meet with a lot of companies, and I'm sure every company tells you that they're undervalued. And -- but we've underpinned this with what we believe to be very rigorous analysis, and we've reached the conclusions that we have. And it's reflected in what we've talked about here today.

Robert Blair

Katy Huberty?

Kathryn L. Huberty - Morgan Stanley, Research Division

Katy Huberty from Morgan Stanley. This morning, you talked about a new channel to market, which is the ISPs and social media customers. Can you just compare and contrast what those customers are like working with versus the OEMs in terms of how quick is the qualification cycle, what type of lead time on orders do they give you, how accurate and seasonal are those orders, how much they push you on price? I just want to understand how the business model and how you run the business might change if the shift to the cloud happens faster.

Stephen D. Milligan

Yes, I think that -- and Mike kind of alluded to this, but I think that -- and all customers are different. I mean, it's easy to categorize them altogether, but I think that when you look at some of our nontraditional customers, if you want to call it that. I mean, they have a lot more flexibility. And they can do things differently. So if you look at qualification as an example, generally qualifications would be quicker. They're also more comfortable as an example with not necessarily going through a traditional multi-sourcing kind of an environment. Their comfortable with sort of locking in a design depending upon what they're trying to do. And so I think that -- we love all of our customers, right? And that's important. But it does certainly provide some advantages, some of the different kinds of customers that we're working with now in terms of how it helps our business and how we can help their business.

Robert Blair

Rob? We’re coming.

Unknown Analyst

I'll do the one near term and one long term, but my near term’s a full year so [indiscernible]. And with the near term, and sticking with the $10 EPS target for fiscal '13, in broad strokes, how much of that is expecting revenue to bounce back maybe post Win 8 rollout? And inventory rebalancing, how much of that is coming more from higher gross margin?

Wolfgang U. Nickl

I'll give you what we said when we announced the target. I mean, we said in July that we are assuming a 5% growth of the market from 600 to 630. We're not having a great start into the year, but we don't know. We don't know how it will land, we don't how successful Win 8 will be. We don't know what the effects of the rebalancing is. We don't know what the economy is doing. What we know is the knob. With many knobs to get to EPS and like I said during the presentation, we're heavily focused on the cost now, and we'll set the target. We think we can get there, and -- but the assumption that was in there was a 5% market growth for the year.

Unknown Analyst

Okay. And then from a long-term standpoint, if you look at diversification, which has been one of your, as you said, sort of last 10 years drivers, one of the ways to diversify is -- or it seems like one of the challenges has always been as you diversify too much forward, you risk competing with your customers. And you sort of walkedd that balance and things like that and that sort of thing over the years. Does that change at all in a sort of post-consolidation world now where there are fewer drive guys than there are customers? I mean, does it get you more ability to conceivably compete with your customers because there are fewer of you than there are of them, or is that really unchanged?

Stephen D. Milligan

Well, I think that -- I mean, I don't think you can sort of draw a bright line in that regard. I think that part of what's happening is, and I alluded to it in my presentation, that the landscape or in a broader landscape, right, not just a storage landscape is changing dramatically and the whole environment, right? And so you're seeing -- there's a little bit of historical answer to your question, but you're seeing lots of people competing against each other that didn't compete against each other in the past. And you're also seeing in some cases, call it, strange bed fellows, right? And I think that we have to recognize that reality, right, and understand -- Because what's happening, and Mike alluded to it, and John was teasing us when we were going through the stuff, he didn't necessarily care for the word, the tectonic shifts. There are dramatic things that are going on in our environment and we're part of that. And so that is going to mean that there will be different relationships from the feature than maybe what we've had in the past. Now the key thing is, is that how we deal with those changed relationships. It's not the fact that they've changed, it's how we deal with it. And we're not signaling anything in particular that we're going to go do an X,Y and Z because it's something that we have to continue to evaluate. But we're not going to draw a bright line and say we're not going to cross this line and because it's going to – somebody’s going to be upset about it, right? Because we all have to evolve. And as we evolve, relationships change.

Robert Blair

Rich? We'll wrap up. This will be our last question, and then John will conclude.

Richard Kugele - Needham & Company, LLC, Research Division

Just one quick question on head technology. Can you just update us on where you stand with some of your new fabs on your internal capabilities and when you expect to get back to your previous model on internal versus external?

John F. Coyne

I think we -- several things to make clear. Many, many years ago when WD invested in direct ownership of head technology and capacity when we bought Read-Rite, at that point, we developed a strategy to which we've been true ever since of having a substantial minority of our sourcing from the outside. That strategy was designed to ensure us against transition risks and technology transitions, as well as to support ramps of new technology that are significantly difficult in fab utilization. And therefore, splitting the ramp across multiple sources is a very good thing to do. And we had a very live demonstration of the value of that strategy immediately post-flood where our external partner, TDK, did just a tremendous job in helping us to respond to the challenges of having our own slide or fab wiped out of Thailand. We've made very good progress in recovering from that in terms of our internal capability. We’ve brought up a much more risk-balanced profile. We have – we’ve restored capability in Thailand. We've created capability in Malaysia. Working with TDK, we've created contract capability in the Philippines, processing WD wafers and of course, we have also brought the HGST supply line into the fold. I'm tremendously grateful for the support that TDK gave us post-flood and continue to give us. They'll continue to be a part of our key strategy going forward. And as we mentioned, one of the knobs we have to turn is as we bring that back towards the long-term model, which we'll do through this year, we have some opportunity there from a cost perspective. But we are very committed to that relationship, and we're very committed to -- even more committed today to the value of that partnership because we've actually tested it in the fire and it worked really, really well.

Robert Blair

Thanks for your questions. I'm now going to turn the podium over to John for some concluding remarks.

John F. Coyne

Okay. So I hope in our day together that you've seen that the title of Passion for Change was not misplaced and that you -- I hope you are convinced that our focus on intersecting markets that are developing and moving in new directions is indeed a formula for sustained profitable growth.

And I'll just remind you of the key pillars and trends upon which we're focusing the business in the next 10 years: the evolution of thin and light and responsive performance in the client side; the consequent creation of increased demand for mass volume storage, both in the private and public cloud arena and in the personal cloud arena. And we believe we've got a very, very good portfolio of technologies and portfolio of product solutions and roadmaps to go address those new requirements and do so very effectively.

And I hope you saw from the -- both the presenters here and the other folks from the leadership team that you’ve had a chance to interact with last night and through the day that we have a tremendous debt of bench in terms of leadership talent within the business. And it's that, that gives me the confidence to call it a day relative to my personal input to this. But I have every confidence in this leadership team to recognize and address the changes that are around us. And in the context of the fundamental thing that we do, which is to preserve and access and create and manage digital content, we have a tremendous opportunity set. And these folks are the best people in the business. I think when you look back, you've got to endorse that statement and highly motivated to continue to succeed into the future.

So I hope you've seen that today. I hope you take away that we have been thoughtful about our business model and our capital allocation strategies, and that we have been in the past. We will continue to be in the future. And I believe that there's great value for investors in this business. There's opportunity out there. There's a team of people who know how to grasp opportunity, turn it into customer delight through great products and then turn that business into extremely solid financial results.

So I want to thank you for taking the time. Most of you, I know, have come a long distance to be with us today. And I want to take -- thank you for taking the time to do that. I hope it's been useful to you, and I look forward to seeing you all again before I finally go play. And again, my thanks to the team, a good job today. And let's make sure now we have a good job for the next 10 years. Thank you very much.

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.


If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!