EMC's (EMC) Management Hosts Bank of America Merrill Lynch Global Technology Conference (Transcript)

Jun. 4.14 | About: Dell Technologies (DVMT)

EMC Corporation (EMC) Bank of America Merrill Lynch Global Technology Conference Call June 3, 2014 5:50 PM ET


Jeremy Burton - Executive Vice President, Product Operations and Marketing


Scott Craig - Bank of America/Merrill Lynch

Scott Craig - Bank of America/Merrill Lynch

Started here, good afternoon everyone. I am Scott Craig from Bank of America/Merrill Lynch. As you know, I cover the IT hardware storage and distribution sector here and now we are happy to have EMC with us next. Sitting to my right here is Jeremy Burton. He is the Executive Vice President of Product Operations and Marketing. I know he is familiar to some of you guys. He meets with investors. So, well rehearsed in the whole shindig here, but he is going to go through one slide and a few remarks and then we are going to go right into Q&A and hopefully we will keep this interactively, feel free to shove your hands up in the air at any point in time.

So, with that, Jeremy thanks for coming today. We really appreciate it.

Jeremy Burton - Executive Vice President, Product Operations and Marketing

Alright, thanks for the invite. My one slide you might be relieved to know is that one and I should point out I will be making forward-looking statements, which involve risks and these are well-documented in our SEC filings upon emc.com. So, that concludes my formal presentation for today.

And as far as opening remarks, I thought what is usually the EMC portfolio, it’s broad, it’s diverse and I thought it will be worth spending a couple of minutes just cluing you into the thinking what we are trying to achieve as a company and how the product strategy hangs together. At a macro level, I was getting asked what about this cloud thing and I tend to backup from the tech trends as I call them and kind of look at what the business drivers are.

I spend a lot of time in the briefing center. Our pivotal division has got a pretty interesting customer briefing center up here in San Francisco. And increasingly into that briefing center, we are getting all manner of companies coming in essentially saying hey, we want to become a software company. And this has been automotive firms, this has been sports retailers, this has been insurance companies. And these are essentially old school companies. Often, they don’t bring IT with them by the way.

And the real desire is look if we are a services company we want to surface our services through mobile devices and tablets and put our services online and provide more frictionless access to those services. If it’s a retailer, the same hey, we know nothing about our customers. For the first time, we can embed telemetric data, our telemetry devices on to clothing we can capture information about the people that we sell through. We can capture that up on our website. And for the first time, we can really understand who we are selling to. And so we see the business in many, many different industries driving a trend towards new application development.

New applications though they are going to be built I think very differently, they are going to use a different application platform, they are going to use a different data platform and that then is going to drag a different set of infrastructure. So, in the past, we, EMC, typically have gotten involved from an infrastructure perspective, but I think increasingly and as I said pivotal been the tip of the arrow here has really pulled it into these business conversations, where we really see what is driving the change, not just in the datacenter, but in the storage environment as well.

Now, specifically, drilling down the level into the product strategy for a second, we have never believed that one storage product was going to solve all used cases. We have been fairly loud and proud about having a best of breed strategy for storage and expect that to continue. If I look at some of the more disruptive trends within storage, I mean clearly flash we think is going to be disruptive in many transactional or latency-sensitive used cases. We really start to adopt flash in 2008 within our hybrid arrays. We couple of years ago went and acquired XtremIO, which is all flash array. And we start to see VDI and again latency sensitive database workloads, that’s starting to play very, very well. And then most recently and probably at least known to many, we acquired a small company called DSSD that really is playing in the world of service side PCIe flash.

And we haven’t really disclosed the whole lot of detail around that presence, but I think when we do it’s going to be fairly shocking, there is just a fantastic team working on that. And so as much as we think this flash thing is going to have a big impact, the smart guys IDC tell us that through 2017 approximately 3% of the data are in the enterprise is going to be flash, which means that 97% is going to be sitting on hard drives technology. I don’t know extreme that kind of multi-hundred petabytes and exabytes scale. We think a lot of those display systems are going to run against commodity storage hardware, why I mean think about it if you got an exabyte of data if you can save $0.10 a gigabyte that’s $100 million you save right there. So commodity drives that kind of scale do make a difference.

And we think a lot of the new applications the new web scale applications that are going to be build, the way they are going to be architected is different and they are going to be able to be more easily deployed on that kind of an infrastructure. Because we didn’t think of that new architecture for storage should be managed through the silo and the same way is the mainframe is really managed with the silo within IT, we really had this kind of vision that the existing storage of state wouldn’t it be nice if you could manage that right along the side this new architecture that’s going to run a lot of the new applications. And that really was the genesis of the ViPR project.

And so ViPR really is going to allow you manage your existing storage of state and provision it in a very, very similar way to the new storage of state potentially running on commodity storage. And so we think it’s going to provide a great bridge between the world of today and the world of tomorrow. Now one of the unique things about ViPR is not just that it is a control plane and a management of a (blocking) layer, it’s also a new storage system, it’s left 20 years, it’s been about blocks and files in the world of storage. We think the next 20 years is going to be increasingly about things like object storage and HDFS storage.

And so ViPR actually provides a new object and HDFS storage system that can be run against commodity hardware. ViPR is also the thing that is going to allow us to plug very, very seamlessly into the VMware software defined datacenter. So VMware one of our federation companies today you can manage all of ViPR from the VMware environment. So if you are a VM app and you want to use VSAN and you want use vCAC, you wand use a VCL, you can access everything that is in ViPR today right through the VMware environment.

By the way you can also do it through the OpenStack tools as well but that type degree of integration with the VMware software defined datacenters is that today. And we think that’s going to be the basis for what a private cloud environments. We also think it’s going to be the basis for a lot of public cloud environments including vCHS which is really the – a federation level service provider entity that ourselves and VMware apposite on in order to provide this public cloud services to folks who want to move workloads off premise.

And finally, once you got that cloud based infrastructure I will kind of finish where I started, people are going to want to build a new set of applications and that really is where Pivotal comes in. So Pivotal where the people are building an application and deploying is in cloud foundry I think cloud foundry is one of the most strategic assets that we have in the portfolio or whether they are looking for help on building these next generation applications that was really the rationale between the acquisition of Pivotal Labs. So when you think of Pivotal is helping people build applications on a new platform, deploying them to cloud environment, hey we would love it vCHS, VMware EMC and then really staring vast quantities of data on the new storage architecture that we think we can provide from within EMC. So I hope well that gives you a flavor of how the various pieces of the federation hang together. And with that I will turn it back.

Scott Craig - Bank of America/Merrill Lynch

Yes. It’s great. Let me kick off with a few questions to start with when I am out meeting investors obviously the feedback on the product portfolio is generally quite positive. And yet I am trying to connect that with the fact that the storage growth just hasn’t been there from a revenue perspective and there is obviously a bunch of different buckets you have got the high end and the medium end, then you have got the new initiatives emerging markets, well, whatever you want to call it. And so I think people are struggling with why we are not seeing more growth even out of that core storage if you want to call that. So, I guess they are very vocal at EMC World this year more so than in the past that some of the issues that the industry is sort of wrestling with right now. So, maybe take us through your thought process on what we are seeing from a near-term perspective and how we reinvigorate the revenue growth back in the company and hopefully you get the stock price going?

Jeremy Burton

Yes. It’s a combination of cyclical and secular type of issues. I mean, I think on the cyclical side of things, we have this big spending boom after the recession, where I think there was a lot of pent-up demand. I think since then the certain industries that have just never returned to their former spending levels, look I would say financial services is one and the federal government is another. And those are two pretty big sectors. I would also say that even outside of those two industries, because there has been somewhat of an uncertain environment with the economy, we have seen kind of how it evidenced that folks are sweating their assets more and that’s not just using more aggressively using data reduction technology compression dedupe and so on. It is also the deferring tech refreshes that used to be religiously that folks would refresh their infrastructure after three-year time horizon. We started to see that stretch out to four years, five years. So, people are sweating their assets. At the same time and I think you guys would probably appreciate this as investors in the storage or infrastructure business in general, there is a lot of noise.

I talked to flash earlier, that’s a point of thought I think for a lot of folks in IT infrastructure. The software defined and commodity storage thing, well what role is that going to play, what workloads can I run, is that a viable strategy? Should I start forming relationships with an ODM in China to go buy my own drives, converged infrastructure, should they even bother buying storage at all? Should I go straight to server storage network and buy it as one converged block or should I go public cloud? And I think we don’t see at least evidence in the enterprise, which is really our core market of unmatched people moving to go deploy interest applications in public clouds, but the fact that it’s out there is absolutely a point of consternation and I think is causing people to pause. They want to make sure that if they are going to embark on a strategy for the next X number of years that they are not going to go down a dead end. And so the pause is twofold. I think one has been economic or macro related. And I think at some point we know in the storage business that people have to buy at some points. The other one that I think is playing at the same time is this kind of consternation over the secular trends and folks really trying to understand what is the right way forward and what is viable at this time.

Scott Craig - Bank of America/Merrill Lynch

Okay. And certainly to follow-on to that though when I am out marketing and talking about your stock and business, one pushback I get is on the ramp that we have going for the back part of the year here from a revenue perspective. And so take us through that and does this pause that we are going through right now is it starting to come to an end if they have been slating their assets too long, etcetera, etcetera, because again I am getting some pushback on that revenue ramp as I am sure you guys are as well, so?

Jeremy Burton

Yes, two things. And I would separate the two. I mean, Q1, we were pretty happy with the product bookings. So, they were very much in line with what we had hoped to hit the – so, for the year we are actually feeling like everything is in line, nothing moderate. We also in the back half of the year have got some pretty good product refreshes coming. I think everybody is probably the worst cap scenario is going to be a refresh of the VMAX. That always gives us a path. I see no reason why it won’t this time around, but it’s not just the VMAX. We also are right in the middle of a good cycle with the VNX. The backup products continue to do well. Isilon, we have got a refresh coming there. That continues to do well. And XtremIO, we are very bullish on XtremIO. I think that is going to make a meaningful contribution. So, as much as revenue in Q1 was certainly well below what we expected to be for the year, we think there is enough indications in the bookings rate than also in the products portfolio coming down the line that would get us back for the year.

Scott Craig - Bank of America/Merrill Lynch

Okay. And then in the context you guys made a change to your business back relating to quarter ordering and steps like that, some people are still a little bit confused as to that as why now and sort of why at all, so maybe just go through that and how that fits into some of the guidance for 2014?

Jeremy Burton - Executive Vice President, Product Operations and Marketing

Yes, it’s good question. I mean it’s one of those things that I am sure internally focus that we will be saying pretty soon why didn’t we do this five years ago. The real goal is of changing our end of quarter practices was to try and take some of the pressure off of the factory. And actually some of the pressure of the deals at the end of the quarter and never is that more a Q1 than in Q4 at the end of the fiscal year. And you never wanted to do natural acts at the end of the quarter and our sense look if during the year we can build bigger backlog then we can better manage the business at the end of the quarter and also at the end of the fiscal year. And we wanted to give some indication of what impact that was going to be when we did Q4 earnings because obviously that was first time that we are going to be introducing the change. And I think we would have liked to have probably build a bit more backlog, but this is not an exact science, if at the end of quarter you get a bunch of software ELAs or hardware orders that are easy to fulfill, you are going to fulfill them. The revenue is anything was a little bit higher than we expected, backlog a little bit less, but we expect that during the course of this year, we will continue to build the backlog. And again the big hope there is it takes the pressure of the factory and of the business at the 11th Hour.

Scott Craig - Bank of America/Merrill Lynch

Okay, that’s one more and then open it up. Flash, you mentioned is the big opportunity and XtremIO has started out very well, what’s been the key selling point on XtremIO versus some of the other competitors that have already been out there for maybe a little bit longer and what’s been the customer feedback on that so far because you could argue you are probably the number one player in the market there recently right?

Jeremy Burton - Executive Vice President, Product Operations and Marketing

Yes. We would argue definitely I mean look XtremIO we took a little bit longer to get out the door than I think most would have hoped. But there is a big difference between it’s been a like XtremIO was the startup and it’s been EMC. There is a certain expectation of quality and field service that we have got to really hold ourselves to and so we held on the product a little bit longer. And we felt back when we made the acquisition XtremIO of the available players at that time had the best architecture. And by that it’s a scale out architecture. We shared memory between nodes in the cluster which means that when data comes in we can do everything in memory. So if we have to dedupe, if we have to compress, if we have to perform any kind of action we can do that in line in memory at the time the data comes in.

Most of the other guys have a dual controller architecture. They can’t process things in line. They write them to the flash and say okay I will come back to that later. Now that’s fine when the array is not busy, because you can get back to it immediately and it appears as if it is in line. It’s kind of like if I said, hey can you get me a cup of coffee or if you got nothing to do you can process online go get me a cup of coffee. But if 10 people asked you, you put my cup down until I will get back to you. And if people could keep coming to you saying, hi get me a cup, get me a cup, you will never get back to my cup of coffee. And so it’s not processed in line and so then what happens well if that’s deduplication the data store coming in you have got to write it to the flash, but now you are writing non-deduplicated data.

And you get this massive contention for resources which results in many of these old flash arrays going into where I can only describe as a pathological state. I mean they go – latencies go from like a millisecond to like 10 milliseconds, 50 milliseconds, 100 milliseconds. And so our value proposition is very, very simple. When data comes in we process it and we do it in line and we do that for all data. And so whether the array is empty or full, whether it’s busy or idle, we will always deliver that millisecond signal a second latency. But that is not the case with almost every other old flash arrays.

And one of the reasons why we are so bullish on XtremIO now it starts to hit its stride. There is a series of very simple benchmarks we could run against the competition. We fill the array up, we delayed a few things, we have run at 80% and weird things happened. So, we are increasingly bullish on our win rates and we don’t have or have not had all the data services of some of the others out there, but it’s relatively easy to put data services on top of our architecture. What is really hard is to go re-architect your product to deliver that consistent performance. So, yes, we are very, very bullish on XtremIO, had a great start. And I think it’s just going to get better from here and that is another one of those products that we have a refresh coming a little bit later this year.

The one that we said least about is the DSSD. We think all-flash arrays is a big business there, but if you look at the world of in-memory databases, SAP HANA for example and also a lot of the high end big data analytics. We think there is a huge opportunity there. So, think of something which is server side flash. It’s a shared storage. It’s connected into every server in the rack through multi-lane PCIe connections and yes, something which is really going to deliver an order of magnitude improvement in latency over the very best of flash array. And that is what we saw in the team and what they have built. And sometime next year, we will be bringing that out as well. So, we don’t think it’s just a hybrid array play or just an all-flash array play. We think it’s also a server side flash play and which of those that you are going to use is going to depend on what you are trying to do. And in the same way, they are started saying look we have never believed that one size fits all in the world of disc. We don’t believe one size is going to fit all in the world of flash.

Scott Craig - Bank of America/Merrill Lynch

Let’s open it up for questions from out here in the audience.

Question-and-Answer Session

Unidentified Analyst

Could you explain the differences between EMC’s ViPR solution and then VMware’s VSAN and how I guess those – are they competitive with each other or what are the differences there?

Jeremy Burton

Sure. I think the confusion comes in because when VMware talks about software-defined storage, they mention – VSAN is the first thing out of their mouth. And when EMC talks about software-defined storage, ViPR is the first thing out of our mouth and so therefore the confusion. Technically, these things are well-departed. You have got to think of VSAN as an internal storage pool for vSphere. So, think about VSAN as a feature of vSphere. So, there is no concept of running vSphere without VSAN or vice-versa. It’s essentially a feature of vSphere. And it doesn’t expose itself as a storage array. It doesn’t talk fiber channel or iSCSI or NAS or it really is an internal VM storage repository. That’s the best way to think about it.

ViPR is two things. Number one, it manages storage is external to vSphere, so, VSAN internal storage to vSphere. ViPR manages external storage. Now, that external storage could be in a regular storage array, VMAX or VNX or it could be a NetApp’s box or a Hitachi box or it could be one of the ViPR data services, it could be ViPR object storage or ViPR block storage. The part of ViPR that I think VSAN gets most confused with is the ViPR block storage, which is based on the technology that we acquired from ScaleIO about maybe 12 or 15 months ago. Why is that? Well, both VSAN and the ViPR block storage what they do is they pool direct Hitachi storage, alright, so they both do that.

VSAN, as I said earlier, is a feature of vSphere. It really only works in a vSphere environment and it only really will scale out to 32 nodes, which is the maximum size of a vSphere cluster. The ViPR block storage, it too pools DAS from service, but the design plan for it was kind of literally thousands of nodes. And it doesn’t just work in a VMware obviously we connected up into VMware as I outlined earlier with the ViPR controller. So, it scales to thousands of nodes and it will work in not just a vSphere environment, but also in OpenStack and a Microsoft environment as well. So, the confusion usually between the two is the ViPR block data service, which is based on ScaleIO and VSAN. Those two are storage systems, which do similar things. But ViPR is so much more than a block data service, it’s an object data service, it’s an HDFS data service. And it’s not just data services it’s also a control plane for managing all of external storage. So, as I said I can understand why there is a little bit of confusion, particularly between ViPR block data services and VSAN, but the two things taken in its entirety, ViPR is a very, very different piece.

Scott Craig - Bank of America/Merrill Lynch

We had a couple of questions over here on the fireside. Two questions. One is just related to your strategy for not the web scale service providers but large private cloud implementations which in five years will become a little more predominant in large enterprises your core customer, what are the smaller service providers, what’s your strategy, product strategy related to that IT would Rackspace for example you announced something with Brocade and is that type of – is that gaining traction? And then second question related to software defined storage, what I don’t quite follow about that versus the software defined networking is that there is no open standards efforts that I can see related to that, do you see that coming down the pike or is this going to be closed software defined storage different than most other software defined efforts and other?

Jeremy Burton

So, let me answer your second one because it’s more tactical and I will forget it, but I will come to your first one. Software defined storage and software defined networking I mean there is an open source project around software defined storage. It’s really in the realm of OpenStack. Some of our frustration with OpenStack has been that the software defined storage standard doesn’t look a lot like the software defined networking standard. And by that I mean in software defined network and they have done a pretty nice job of separating out the control plane from the data plane. In software defined storage, that’s all kind of munched together. So, it’s not as elegant a solution which is why I really don’t think there has been as much adoption what you tend to see right now is that folks are providing Cinder and Swift drivers, so that they can make their arrays plug into an OpenStack environment.

We would love to take that kind of OpenStack definition much more down an SDN like approach which is really what the architecture of ViPR is. And who knows if we made progress there I think we would probably be gaining even submitting cautions of ViPR to what’s making it a much more clean standard. So you are right that today it is not well clearly defined as SDN. I think that is in part because the software defined storage standard within OpenStack is not as well written. It’s kind of the (munched) the control plane and the data plane, but that can change. And assuming that has been a topic of vigorous debate within EMC.

On your first question around hybrid cloud or around private cloud environments I mean two things. We actually have been working for some time with VMware to specify in an incredible amount of detail what would – what are the steps involved in taking existing datacenter environment from what it is today a virtualized environment to a private cloud and being very prescriptive about that. And you might have seen if you are at EMC World last month, we talked about building a hybrid cloud in five days. And so until building these cloud environments becomes that repeatable and that quick, you won’t see the mass adoption. So, as with all these technologies being prescriptive and providing the series of steps and the implementation details that are required not just for the customer directly offer to the system integrators, we think once we have done that, that’s when you really start to see a take up from the internal private cloud deployment, so lot of people call that virtualized environment a private cloud. For us, yes, sharing of resources and pooling of resources, that’s the beautiful thing and that’s part of it, but the real value for a lot of folks in IT, it’s the higher level automation, it’s the self service delivery of these resources to the people that need them. And that’s what really requires the implementation from the either EMC directly or from one of our partners.

The service provider environment, what we find within service providers is there is often two very distinct parts. There is a product that looks very much like enterprise IT, which is running traditional existing workloads. And if I look at a big chunk of our business with service providers, it’s essentially been for them to build out those, what I would call, hosted private cloud environments. We have been very successful there. And I think for the last seven or eight quarters, it’s been the fastest growing part of our business. In a lot of these service providers, Rackspace included, but also many others that we have dealt with, there is another part of that business, which is what I would call true public cloud. And that’s the part of the cloud environment that is running a lot of these next generation applications.

And what you find is that the architecture there is very different. A lot of the traditional enterprise applications that have been built over the last 15, 20 years, I mean folks would pickup an application server, BEA or WebSphere or something like that, an Oracle database or DV2 or SQL, a bunch of Java programmers and boom, they have not got an application. I think these next generation of applications are going to look very different. Right, it’s going to beat the mobile devices. I think you are going to find a lot of in-memory potentially no SQL databases in there. You are also going to see for analytics, it’s going to be more of a Hadoop environment than something like a Teradata. And the underlying infrastructure that you are going to find is going to be different. A lot more of the resiliency is going to be built into the application. Therefore, it can exist on much more of the commodity infrastructure. And so, our play in there, one of the lead plays there, I mean, this is really why we built ViPR. I mean, it really is a true software-defined architecture that contemplates commodity storage and these new storage systems such as objects such as HDFS. And you are going to see us that’s going to be our lead (indiscernible).

I would also tell you as well that even these big data analytics environments, which have been typically kind of hard drive based and kind of offline batch processing, we do see a desire for many people to take those systems in the direction of being more and more real time. Meaning there is going to be a dataset within this potentially hundreds of petabytes of data that they are going to want to process more quickly. And we think that is going to be a play for something like DSSD. So, if you like DSSD becomes the hot edge and then a lot of what we have done with ViPR becomes the capacity tier of the cold car. And if you think about it for a second, it’s not dramatically different to what you have in a storage array today. It’s just that within a regular storage array you have to flash tier and the state of the capacity tier all in the same frame. I think at massive scale for all of these new applications, those things end up being disaggregated probably exist within the same rack, but the functions are essentially the same. They are just architected to operate at an order of magnitude bigger scale than the systems of the past. So, it’s kind of a long answer, but hopefully it gives you some insight as to where we are headed with that.

Scott Craig - Bank of America/Merrill Lynch

I think we had a question upfront over here.

Unidentified Analyst

Hi. Could you comment on the competitive pressure from pure storage, where are you seeing them the most and how are you responding?

Jeremy Burton

Yes, I think if I will start it’s out there, pure have probably got definitely got the biggest mouth. We don’t actually see the math much. I mean, that business is still relatively small. If I look at our typical revenue in a quarter $6 billion, they might be doing $25 million, $30 million. So, it’s still in the grand scheme of things relatively insignificant, what I would tell you is that the kilo workload for us out of the gate with XtremIO has been VDI and I think that’s now competitive. They are really – they really got the wrong architecture, it’s a dual controller looks a lot like the hybrid array, it looks like a VNX or something like that. And so it’s got dual controller architecture. It does go into this pathological state when you make it busy. And their advantage is they are not the – they probably got out that was a big voice first. But we have got a pretty interest in refresh go in middle of the year where I think the database workloads will be well within our grasp, because we will add things like snapshots and replication and compression. And we got the best sales force in the world. So, I think these guys are picking the fight with the wrong guy. And you don’t want to be bringing a slick knife to a gunfight.

Scott Craig - Bank of America/Merrill Lynch

Maybe I will ask one more here then Jeremy, you have touched on VMware a little bit and there are certainly questions from investors as to why it’s necessary to own VMware or a question the federation sort of model, so you have the unique perspective of running the products and operations from EMC’s perspective so how important is owning VMware versus how different would it be dealing with them if they weren’t in part of this federation?

Jeremy Burton

Yes. It’s a great question. It’s been topic recently as well. Look, there is certainly mass that you can run and you guys are probably the best in a business at it which would show it being a better bet to sell the whole thing up and keep it. But from a technology standpoint you kind of live and die in this industry by the control points you have. VMware is a control point and the assumption that everybody misses I think is if you sell it off there is no guarantee it remains independent because for the right price somebody else can buy it.

And so the idea of selling off to me would be suicide because it is one of the most precious control points in the industry and we wanted to remain part of EMC. And I think it gives us tremendous leverage in our story. But by the way I see the same thing around Pivotal. And on Pivotal earlier in its life and is not yet established itself as a control point but that application and data platform not only flips once every 20 years. And if we can establish a control point there it’s successful as Pivotal might be I think we would want to always have some level of control because it gets us into a set of conversations maybe that we wouldn’t be in. So I think there is kind of zero chances to selling VMware off.

This federation strategy it does – there is tension because one of the premises if you like of the federation is that release each entity at some level of ability to work with others. It’s not given that VMware is always going to work with EMC that Pivotal is always going to work with VMware and that does create tension. But because that part of this entity on certain things we have to get along. But for the idea is we should be seeing the world through the lens, we may not fly as the core flash from point A to point B. But there is an access of reasoning amongst these federation entities that particularly with the big accounts. I mean I will look at the big global accounts. They don’t want to hear from three companies, they want to hear from one. And for those accounts alone the prices are big if we can align and telco hearing story and also have the customer understand that look if anyone of these components doesn’t work out, you have got to get out a jail free card. You have got choice it’s not a vertically integrated lock-in.

So it does create tensions, it’s not always easy. But I don’t know of a better way to provide the focus on the opportunity and at the same time align that focus with an overall strategy. And candidly, it’s one of the reasons I came to the company in the first place.

Scott Craig - Bank of America/Merrill Lynch

Okay. Alright, well, with that we are out of time, Jeremy. So, thanks a lot for your time. Thank you, again.

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!