Seeking Alpha
We cover over 5K calls/quarter
Profile| Send Message|
( followers)  

Executives

Simon Biddiscombe - President and CEO

Jean Hu - SVP and CFO

Analysts

Mark Moskowitz - JPMorgan

QLogic Corporation (QGLC) JPMorgan TMT Conference Call May 16, 2012 11:20 AM ET

Mark Moskowitz - JPMorgan

Good morning everyone. My name is Mark Moskowitz. I’m the IT hardware analyst here at JPMorgan. And we are pleased to have with us today QLogic. We have both the CEO and CFO. So, Simon and Jean, thanks for coming. I appreciate it.

Simon Biddiscombe

My pleasure.

Mark Moskowitz - JPMorgan

What we'll do is start out with about 15 minutes or so of questions from me and then this event is really for the investor audience. So, we will open up the last 15 minutes or 20 minutes to investors. So, please be comfortable asking your questions. We do ask you to step up to the mike so the webcast listeners can also hear what you are asking.

But for now, I will kick it off here. So, Simon kind of a loaded question, but I think pretty topical point right now just given some of the anxiety across the tech investor base and beyond with respect to enterprise spending. My question is kind of loaded so just spare with me, but our JPMorgan CIO Survey recently suggests that offshore services companies, printing companies, networking and communications equipment companies were going to have a much tougher growth prospects relative to software-as-a-service, software application companies, storage, server virtualization type companies. And this past five or six weeks of earnings have really confirmed those survey results, most notably with Cisco disappointed recently. So, I want to get from your vantage point, when Cisco is taking about enterprise spending is that maybe just relative to their end-market exposure that they are not really exposed to the faster growing or maybe sturdier growth markets today, whereas maybe 10 years ago they were?

Simon Biddiscombe

Yeah, it’s an interesting question, Mark, and when I think about how QLogic participates in the key areas of the datacenter within the enterprise and increasingly outside the enterprise in places like Web 2.0 and cloud type environment as well, there is no doubt that the traditional enterprise part of the market whilst to the headline level is not necessarily as robust as it started have been. When you think about the areas where QLogic is placed specifically it actually remains relatively strong.

And so, if you start with the storage market, whether you are talking about growth in storage capacity across the enterprise or whether you are talking about growth in storage capacity across every other environment that continues to be very robust. And fundamentally QLogic is a company that connects to storage and how well we're looking storage area and networking and whether its Ethernet-based network or fiber-channel based network this is the world for us about storage and datacenter networking at this point in time.

So, storage market I actually think it’s relatively robust at this point in time. I think it continues to be one of the bright spots within the overall IT spending environment. We actually done very good job carving at new positions for ourselves we will talk a little later within the context of the storage market.

And then on the server side, clearly we are going through something of a refresh cycle or we're at the very earliest part of the refresh cycle as the Romley-based processors become available from Intel and servers-based on those processors become available. To a certain extent, they have already started shipping. To a lesser extent we are going to start shipping later this quarter from other OEMs.

So, you're not at a point where that part of the market is at a full run rate at this point in time, but the server market continues to be relatively robust at this point in time both in terms of the enterprise, but more importantly in terms of also what’s going on in the cloud and Web 2.0 type environment.

So, yes, at the headline level it’s not often to the races necessarily, but once you get down to the parts of the market that QLogic truly cares about, which is server and storage, the market is a little bit more robust than the headline numbers are.

Mark Moskowitz - JPMorgan

Thanks for that overview. For investors who aren’t as familiar with QLogic can you maybe just summarize your key addressable markets and what are the attached or associated growth opportunities there over the three to five years?

Simon Biddiscombe

Sure. So the company has historically been very Fiber Channel centric. The Fiber Channel market has been around for like 15 years at this point in time. We're going through the fourth transition of technology. So as we move from 8-gig to 16-gig over the course of this year, QLogic enjoys a dominant share in Fiber Channel connectivity; we have about 57% market share. Now that's the highest market share we've ever had by the way in that market and that was for the March quarter.

We continued to execute very well in that market. And that market actually continues to have growth associated with it but we shipped more ports as an industry last year on the Fiber Channel side than ever before. So despite the minds of people have hold towards the Fiber Channel over the course of many, many years, we continue to ship more ports on a year-to-year basis, and that continues to be a very robust business.

Approximately five years ago, we recognized the opportunity that existed in converged networks, which means bringing together the Fiber Channel capabilities that we've always had with the 10-gig Ethernet network within the data center, and we brought forward products for converged network adapters that essentially moved the 8-gig Fiber Channel traffic on top of a 10-gig wire. And that market obviously very logically took us into the 10-gig Ethernet market. So we follow a progression of technologies and we are now the number two participant in the 10-gig Ethernet market. So, at adapter levels, Intel is number one, we are number two.

So we have done a very good job making sure that we follow the technology roadmap, the trajectory of technologies within the data center by delivering next generation Fiber Channel technologies where we have leadership positions, and then by building on that as we move into the 10-gig Ethernet world, where we've also carved out the number two position for ourselves.

Mark Moskowitz - JPMorgan

And when we think about the converged network adapter opportunity can you maybe weigh in and kind of discuss what QLogic's incumbency in Fiber Channel means in terms of really locking in kind of transition you have to work for you? Because the reason I ask this is, I often tell investors that Emulex and QLogic is more than just spinning a name, so that you guys, both companies have really deep engineering expertise around software, firmware, being able to really breakdown the kernel, if you will, the operating system and the data to make it those transmittable across the network. So because of that software edge does that kind of give you a lead here?

Simon Biddiscombe

Yeah, absolutely that's what drives the incumbency. Historically, and as we move forward to the next generation technology, the way we deliver our solution is essentially a QLogic ASIC on a QLogic adapter. This is a host centric conversation and I would hope that switch in a minute, but it's QLogic ASIC on a QLogic adapter and the QLogic snack sitting on top of that hardware capability okay.

There are many companies who got away with all to bring to market ASIC based capabilities and adapter based capabilities. But there is nobody who has successfully brought to market a snack at any point over the course of the last 15 years other than ourselves and Emulex. And there are very good Fiber Channel companies as you will know, Mark, who have tried to participate in this market four years ago.

One of the fewest people we've had was Brocade introduced adapters and we should be fearful that Brocade would be able to eat away the markets we serve. Four years later they still have essentially zero market share. Why? Because it’s the robustness of the QLogic software stack and the data center that enterprise customers have become very comfortable with over the course of 15 years.

So regardless of which server it sits in, regardless of which hypervisors are operating, regardless of which operating system environment is there, regardless of which switch it attaches to and regardless of ultimately which storage is there, we have the wherewithal to deliver robustness of Fiber Channel capability that the enterprise has grown to rely on. And that translates instantly into the same market share on the converged network adapter side, it is the same market share by the way 57%, 55% because you have become has an enterprise customer extraordinarily comfortable with what you buy from QLogic, and its all but the software in that regard.

Mark Moskowitz - JPMorgan

Okay, great. And then as far as the market size for the network convergence opportunity can you maybe give us a sense of the sizing? And also, what are the key drivers behind this network convergence?

Simon Biddiscombe

Sure. So if you take the total markets that QLogic serves at this point in time on the host side the Fiber Channel market is a about $700 million and then last year the 10-gig Ethernet market was about $400 million, okay. The Fiber Channel market is expected to grow at very low rates over the course of the coming years call it plus or minus flat essentially, and then the 10 Gig Ethernet markets obviously expected to have a very strong growth associated with the launch of the service that is on going at this point in time. And the backdrop areas when you bought a server three months ago it's really one gig connectivity you buy a server today you get 10 Gig connectivity. And QLogic was not a one Gig company; it is a 10 Gig company.

So what we see as an opportunity that expands from about $400 million in 10 Gig Ethernet to about a $1 billion in 10 Gig Ethernet over the course of couple of years and that’s really one gig transitioning to 10 Gig in the data center. So I think everybody is comfortable with the size of the market over a short period of time.

And then to answer your question specifically how much about this is associated with Fiber Channel reconnect, right. So the current estimate is that probably a third, somewhere around 30% of the market for 10 Gig E has an FCoE element to it. And that’s clearly the part of the market where we enjoy the very highest strength of incumbency.

So, we’re comfortable that we’ve got the right technologies serving that part of the market at this point in time. As I said, we’re comfortable with the stack that we spend 15 years investing is enabler of success as you move to that FCoE part of the market.

Mark Moskowitz - JPMorgan

Well, that's a very exciting opportunity in terms of going from $400 million to say a $1 billion with 10-gig E opportunity. Obviously, that's going to probably attract a lot of other participants. Maybe you can just share with investors your view in terms of who you see as the current stable of competitors and who could be the potential stable of competitors as we go out in the next three to five years?

Simon Biddiscombe

Yeah, it's an extraordinarily long list of 10-gig Ethernet companies. So it starts with Intel, Broadcom, ourselves, Emulex, Mellanox, and so on, and then there's another dozen smaller private companies who have 10-gig based technologies each of whom has a specific point product that was introduced to solve a problem for an OEM. So maybe it's low-latency solution, maybe it's a base-T solution.

So over many years many startups have entered the 10-gig market with a view as to how they can carve out a position for themselves. I think ultimately, the offerings from ourselves and certain of our competitors will become all encompassing. Anything you thought you can get from the niche provider, we will be able to offer as part of our standing product, and we've seen that successfully happen over the course of an extended period. So I think a lot of those smaller participants in the market will disappear over time.

The good news there is the market is actually highly fragmented at this point in time. There is a lot of revenues that go to those smaller participants. We're number two in the market, and we have 15% market share. So you can tell that three through n, the lowest have very small market shares individually, and that is available to us as we move forward in terms of market capture.

Mark Moskowitz - JPMorgan

And then as investors try to think about how you can even become stronger in this market, are there certain features or functionality pieces that QLogic is trying to develop now that will be available in the next generation products that you are aware of?

Simon Biddiscombe

Yeah, absolutely, and we don't talk about them specifically, but when you think about NIC partitioning type of technologies, every OEM has a different way to think about how to partition a NIC such that it looks like multiple NICs to multiple servers or multiple NICs to multiple VMs and so on. So beyond the ability to just deliver a 10-gig chip there is hedges QLogic differentiate through incremental technology value such that it's not fighting the fight against frankly some companies that are just slightly bigger than us and buy some more chips right.

So, how I'm going to continue to deliver a higher value solution centric view to the 10-gig world as opposed to just trying to compete in a highly commoditized element of it. So beyond the principal advantage we bring, which is of course 5, 10, Fiber Channel Ethernet, then there are whole series of different technologies that we are working on on a day-to-day basis that try to differentiate from every other 10-gig technology in the market.

Mark Moskowitz - JPMorgan

Okay. Well, why don’t we shift gears to the near term, in terms of just the March quarter? Can you share with us in your view of what the derivers were behind that quarter in terms of maybe some of the disruption that occurred? Was it Romley related, was it Macro related, competitor, just kind of rank order and discuss some of those factors or other factors that may have been contributing?

Simon Biddiscombe

Yeah, so the good news is it wasn't competitive because we captured share in certainly the Fiber Channel part of the market. We haven't seen all of the numbers for the Ethernet part of the market yet, but clearly the capturing of share indicates it really wasn't a competitive issue.

I think people got caught even though we'd laid it pretty clearly in advance by the timing of the Romley launch. So, Romley was meant to launch at any point in the second half of last year, various points in the past and clearly didn't. Then went on March the 9th -- Intel, the 9th processor, I think there was an assumption that meant everybody who had a server technology based on that process would be shipping. That's not the case.

We've said many times and if you look back over previous server launches, its kind of a six month window from the first guy introducing the first server to the last guy introducing the last server is about a six month window and obviously you start with high volume stuff and over time you introduce all of the lower volume servers with a new processor technology as well.

But I think people just expected that great Romley is launched and everything will be off into the races. It's not the way the industry works as we've said before. There is a long ramp to the launch of all of the servers. So that was part number one.

Part number two was, we definitely saw a pause in demand for servers or we saw a pause in demand for our products which was predicated on servers toward the end of the quarter. And if you look at what Intel said about the pause in Nehalem demand, if you look at what Cisco subsequently said about pause in demand for UCS as people waited for next generation servers, clearly we're seeing exactly the same phenomenon as everybody else in the market.

Mark Moskowitz - JPMorgan

And then as far as Romley?

Simon Biddiscombe

Actually, I'll build them up. So having said all that, we were down the same as every other seasonal March quarter, right. It was down seasonally okay. Why? Because on the silicon line people tend to purchase a little earlier that they do on the adapter line primarily because they've got to build the product in expectation of the ramp and so on. And so, things were late -- even though things weren't often to the races necessarily it was pretty close to a normal seasonal March quarter, and we will benefit as we move forward throughout the remainder of the year from the continued launch of Romley based servers, and more importantly from the continued adoption of 10-gig technology which disconnects us from the server cycle, and historically this business has been extraordinarily highly correlated to the server unit shipments.

Clearly, we've got a new technology that we are enabling the market with that should allow us to grow fast in that market.

Mark Moskowitz - JPMorgan

Okay that's a good segway to my next question you kind of already answered, which is that for a lot of us the Romley business and believe me that revolutionary, it seems more evolutionary, but it does seem to be really opening the door for Ethernet. And I want to get a greater sense from you just in terms of how big could this inflection point be. Is that more of a 2014 or is it 2013 event from a calendar year perspective where you could really start to offset slowing in the Fiber Channel, with the 10-gig E opportunity because of Romley?

Simon Biddiscombe

I think to a certain extent you're already seeing it, right. I mean the 10-gig Ethernet part of the business has grown very strongly over the course of the last year be it FCoE centric or be it pure 10 Gig Ethernet centric. And clearly, there’s an expectation that we’re going to see continued growth every quarter as we move forward as more and more Romley server ship with more and more 10 Gig connectivity associated with them. So you’re just going to see more and more and more 10 Gig from this point forward until you get to 40 gig, and where the 40 gig is five years from now or 10 years from now only time will tell but 10 Gig will be the dominant data center I/O connectivity for an extended number of years.

Mark Moskowitz - JPMorgan

Okay. Speaking of technologies, how are other technologies out there such as SSDs impacting your business? Are they changing how networking and the connectivity models are being developed or at least conceptualized for the next generation?

Simon Biddiscombe

Yeah, can do. We see a significant number of opportunities associated with the use of SSDs and solid state technologies generally within the data centers. So whether it’s server side or storage side and whether it’s SSDs or PCIe based cashing type of products on both the host side and the storage side, there’s no doubt that bringing tier-0 tier-1 data closer to the actual processor is going to result in improved performance of the actual applications that you’re trying to run.

So we’re looking at the whole series of different use cases around solid state drives specifically and then PCIe based adapters that will use cash to improve the performance of technologies that we serve into the market at this point in time to improve the performance of applications that run in the data center at this point in time.

So we say there’s been a significant opportunity to actually leverage what we bring to the data center which is I/O, its all about I/O, right? And if I can get critical data close to the processors sooner rather than later that will improve performance application and that’s good for the end user.

Mark Moskowitz - JPMorgan

Okay. And then as a follow-up, is that more of a make versus buy or --

Simon Biddiscombe

Yeah, that’s a good question.

Mark Moskowitz - JPMorgan

Main type of situations, trying to get (inaudible) because there’s so much activity right now.

Simon Biddiscombe

Yeah, yeah.

Mark Moskowitz - JPMorgan

From astute investors.

Simon Biddiscombe

One thing, I mean, one thing I’ve historically said is I don’t necessarily want to be in the SSD business as such. It’s a drive world, it’s a memory world and not necessarily the highest gross margin or they’re not high gross margin typically they’re very competitive at the end of the day and being commoditized somewhat, but I would characterize there’s been ways where we can think about how to leverage our capabilities with other people’s capabilities to bring a solution to market.

Mark Moskowitz - JPMorgan

Okay. I have a lot of questions still because I’ve got to used a lot of times but why don’t we open to the investor audience, I can always come back to my questions.

Question-and-Answer Session

Unidentified Analyst

[Question Inaudible].

Simon Biddiscombe

So we -- so I’ll tell you we don’t actually give you all the dollars. Yeah, I will repeat the question. So the question was if the fiber-channel market is $700 million and the Ethernet market is $400 million, what is the split of my business okay?

Well, I told you I've got 57% market share in the $700 million, and I told you about 15% market share in the $400 million. Okay, so we don’t give you all the dollars, but there is enough there that you can figure out what exactly that is.

Mark Moskowitz - JPMorgan

Okay, while you brainstorm, I will ask another question here in terms of the silicon business recently it was lumpy it's kind of the lumpy --

Simon Biddiscombe

It's always lumpy.

Mark Moskowitz - JPMorgan

Perennial for the company, but just trying to get a sense. Is that just going to something inventors always be prepared for? And what are kind of the drivers behind that market to that, I think one thing as often overlooked is, what is the TAM there and who is the customer base?

Simon Biddiscombe

Yeah, it's actually an interesting market and I wish we didn’t disclose it separately, I continue to wish we didn’t disclose it separately. So, we don’t generate silicon to sell silicon. We generate silicon to sell adapters and switches. And there are some used cases, where people will buy the silicon from us and they may put it on a module that goes into a blade switch or they may put it on a module, I don’t mean mezzanine card. I mean a module that would go in to Ethernet switch and bring the Fiber Channel connectivity.

We sold one Giga iSCSI chips and those types of things, right. So we don’t build chips to sell chips. We would always rather sell them back to because of the gross margin pool associated with this significantly higher. But beyond that market the problem with any silicon business is the supply chain is much longer, right. So people will buy silicon in order to build their products that they will ultimately sell to end-users, okay, and that can be an extended period.

The adapter business, there is no inventory and there is no supply chain. Any inventory in the adapter business I/O essentially. And we keep our inventory next to the major manufacturing locations for HP, IBM, Dell, Oracle and so on and they will literally take in a adapter out tonight, put it in a serve tomorrow, ship the server, inventory story over, right. So we have always had fabulous visibility into what's going on within the hubs for the adapter business and for the switch business in that regard as well.

The silicon business just tends to be lumpy because people will place large orders in order to build out expected demand and demand doesn’t play the way they expected it to, and maybe it’s higher, and often is; and maybe its lower and it often is, but this not the same consistency to the silicon line because of that supply chain dynamic.

Mark Moskowitz - JPMorgan

Since we have Jean here why don’t we have Jean weigh in on the potential margin and cash flow impact from your 10 Gig E opportunity and how should we see that trajectory going from now to say three to five years out are there economies of scale opportunities for you?

Jean Hu

Yeah, so last quarter our gross margin was 68%. We guided the 68% again. So we have been working really hard to protect our gross margin, but overall in the longer term that 10 Gig Ethernet revenue does have a lower our corporate average gross margin.

So over time, we have always to build a model that once we ramp up the 10 Gig Ethernet revenue we'll have about 100 base points erosion each year on the gross margin level. So that’s in the model for the next several years. Of course, it all depends on the revenue mix. If we can grow the top line much faster, then the margin erosion maybe a little bit faster, but it’s the gross margin pool we really care about.

On the cash flow side right, the QLogic generated tremendous cash flow. During the past five years, we generated $800 million of free cash flow. It's a very highly leveraged model. So the R&D side, the operating expense side, the operations side, it’s quite efficient. So anything we can generate in the top line growth will leverage to the bottom line to the cash flow. So of course, going forward, our cash flow generation still going to be very healthy and will grow certainly.

Mark Moskowitz - JPMorgan

And then coming back to the 100 basis points degradation per year, what does that tail look like? I'm talking each year for the next three years, each year for the next five years. What's kind of like the worst case that scenario you guys are kind of formulated in your model I know internally? Are we going below 60% or is it more towards the mid 60s?

Jean Hu

No, middle 60s, right. If you think about this, that’s by the 18 months we said we'll go from 66% to 63%, but today after 18 months we're still at the 68%. So I think if you migrate it down to 100 base points each year it’s like mid 60s. That’s what we can see right now.

Unidentified Analyst

[Question Inaudible].

Simon Biddiscombe

So, the question is what percent of our business would have to be 10 Gig to get to the degradation that Jean characterized?

The Last time I did the math and the model it was about 25% of our revenues that had to be that number or I will say haven’t done it for some time so.

Jean Hu

It's roughly that run rate.

Simon Biddiscombe

Yes.

Jean Hu

Because our 10 Gig business of ours also include the Fiber Channel over 10 Gig Ethernet. So there is a margin mix. For the Fiber Channel over 10 Gig Ethernet business, the margin profile actually is quite similar to our current profile.

Simon Biddiscombe

So, that you can't do the math as simply as your trying to do the math because the margin associated with being in the 10 Gig and [SRE] world is very different in the margin associated with being in the pure 10 Gig Ethernet world, okay.

Mark Moskowitz - JPMorgan

What about the operating margin perspective? Are there any opportunities there where you really have more consolidated R&D resources or sales and marketing in one business or so or is it pretty similar?

Simon Biddiscombe

No, we don’t -- no, I mean, it’s the same product, right. Ultimately, the same product. I'm selling one chip, one adapter and it has different software associated with the capability that means I'm selling the same product to the same customer in an OEM model, right. So, we're very OEM centric. We've got 80% of the revenues coming from handful of OEMs on a quarter-to-quarter basis. So the sales and marketing doesn't change, and the R&D doesn't change significantly either. When I'm writing drivers, when I'm writing software, when I'm writing firmware, however you want to think about it, typically we're doing it with all operating systems all protocol.

Unidentified Analyst

(inaudible) will you be able to maintain operating margins in the 20% to 30% range or will that also be down a 100 basis points per year?

Simon Biddiscombe

So I think the last time we laid out a long-term model, it had operating margins exact between 28% and 32% and that called for some element of revenue growth as opposed to operating expense cuts. We haven't made that that model for about 18 months at this point in time. But I continue to believe that with some degree of revenue expansion based on the opportunities that are ahead of us, we can continue to deliver a high 20s gross margin, okay, maybe it's all the way to 30%.

If you take the last fiscal year, which was relatively flat from a revenue perspective, the operating margin we delivered was 29%. So we got a rich margin model, and we value that rich margin model. We recognize that you value this margin model. But we also recognize that we need to deliver a higher growth trajectory and we're doing invest to make sure that we can deliver on growth and take advantage of opportunities there in the market, right.

So we've three distinct sets of opportunities right now, all of which require us to step up the level of investment and we alluded to those on the last earnings call. But you got the target market opportunities, which is the set of opportunities associated with the connectivity to the storage arrays, from the storage arrays to the network essentially. You've got opportunities associated with some of the next generation technologies that we're going to introduce over the course of the coming months, and then you've got opportunities associated with some of the near term dynamics in the market. And we've got customers who incrementally attracted to our host offerings, based on some of the shifting competitive landscape that we see at this point in time. So there is opportunity and I intend to invest to seize that opportunity and that's what we characterized for you in the model that we laid out on the earnings call.

Mark Moskowitz - JPMorgan

Maybe discussing that a little more in terms of investing for growth, as you alluded to in the last earnings call you talked about the multi-protocol adapter opportunity and why because of that you are investing aggressively in OpEx.

Simon Biddiscombe

Yeah.

Mark Moskowitz - JPMorgan

From a staffing perspective, is that incremental or you're just reallocating resources, or maybe you're laying off folks in one business and hiring folks for another? (inaudible) you are hiring with an view of the company?

Simon Biddiscombe

So, we're hiring. We're aggressively hiring at this point in time, okay, I want to make that clear. We do see the need to have engineering resource to take advantage of all of the opportunity. Now with that said, you don't see all of the impact from an expense perspective because the elimination of the InfiniBand business eliminated a business that was nowhere near at the margin structure that the on going QLogic from today is, okay. So by eliminating a lower gross margin very low operating margin business I’ve been able to backfill into those positions and allow ourselves to continue to deliver on a very high operating margin model moving forward.

Mark Moskowitz - JPMorgan

So speaking of the InfiniBand business obviously the hindsight is always 20:20 but what did you as a company learn about the InfiniBand initial investments and the ultimate decision to divest --

Simon Biddiscombe

Sure.

Mark Moskowitz - JPMorgan

And how did you recommend you stronger going forward?

Simon Biddiscombe

So if you look back to 2005, 2006 so we bought two InfiniBand companies in 2005, 2006, one was a switch company and one was the host company. And at that point in time there was a question as to whether InfiniBand would be the technology that ultimately replaced Fiber Channel and may be even Ethernet within the data center. So we took the opportunity to buy those two companies at that point in time.

As time went by over the course of the six years that we own those businesses, it became clear to us that it wasn’t going to be a dominant technology in the data center, and it became clear to us that it was very HPC-centric in terms of where the technology would ultimately be deployed. The moment it becomes HPC-centric and its really the pinnacle of the HPC world as well it became a very small market. I think our best estimates is its less than a $300 million market at this point in time. And that just involved taking our revenues adding Mellanox's revenues and its less than $300 million and they have got a bunch of stuff that isn’t InfiniBand as well.

So, we said look, the cost to continue to invest in this market to bring host silicon and host products and switch silicon and switch products to market does not make economic sense within the context of roughly a $300 million opportunity. Also a very different go-to market model in that we were calling on end-users instead of OEMs.

So without the InfiniBand business that would characterize us its been a faster growing his gross margin, higher operating margin business with more cash on shore. So, for us it was a very logical step. And I've received no pushback from investors. I've received no push back from customers, and I've received no push back from any other partner in the eco-system around InfiniBand as to why QLogic made this decision; it was absolutely the right thing to do, okay.

Mark Moskowitz - JPMorgan

And then either for you or Jean, how should we think about the cash usage from that divestiture? Is it more to reinvest in the business either through staff hirings or acquisitions or could we see actual cash usage more for shareholders in the form of a stock buyback or a dividend?

Jean Hu

Yeah, the -- our first priority certainly is organic investment and margin acquisition if there is any, but if we don’t see anything in the EMEA area certainly what we have been doing is return the cash to shareholders through repurchase share buyback program. That's something we been very consistent doing last five years. We've bought back $1 billon dollar of our shares and we generated $800 million cash flow. So, that’s the consistent message we have been delivering.

Simon Biddiscombe

Yeah, we've repurchased more than half the company. I think we 100 million shares outstanding; we've repurchased 120 million shares over the course of the last eight, nine years or something in that timeframe.

Mark Moskowitz - JPMorgan

Coming back -- we have a question, do you have the microphone I’m sorry?

Unidentified Analyst

With more powerful server chips like I guess with Romley.

Simon Biddiscombe

Yes.

Unidentified Analyst

Does that lessen the attach rate we suppose to (inaudible)?

Simon Biddiscombe

No, it actually increases it. This is the question we have been dealing with for many, many, many years. It actually started with the VMs right. People said if you're going to have virtual machines, you're going to need lesser adapters, okay. And a similar question will be if these processors are more powerful you are going to need fewer adaptors, okay.

Reality is that I/O for the first time is the bottleneck, right. So, you got these incredibly powerful processors that sort of the multi-socket, multi-core and you are going to run VMs on those cores, the bottleneck is the pipe, right. The bottleneck is how do I get enough data on and off that server, on and off that processor to the storage arrays to be able to take advantage of what that processor offers, right. So, at various points people have thought that technology evolution would result in less connectivity. Today, we would argue actually more connectivity to take advantage of all of the performance characteristics that exits in those processors. Intel has got a great chart that actually shows us. It’s an Intel chart on I/O not a QLogic chart.

And then more importantly, the explosive growth in storage, right. Everything that people are actually trying to store big data concepts and so on, all drive significant incremental needs to high performance datacenter connectivity and web, that’s Fiber Channel, or whether that’s Ethernet-based connectivity, QLogic’s in a position to provide it.

No, for the first time if you think about servers you got points where the memory was more powerful than I/O, the I/O was more powerful than the processor and so on. Today, I/O is the problem. You need more I/O to feed those servers and feed that (inaudible).

Mark Moskowitz - JPMorgan

Right, we have time for about one more question.

Unidentified Analyst

[Question Inaudible].

Simon Biddiscombe

Yeah, I can do it. The cash flow in the quarter was a little bit lower. I don’t think there was anything was specifically called out?

Jean Hu

No, nothing there is some tax payments, pre-payment, there is -- yeah that’s really what it is. It’s a pre-tax payment for the year.

Simon Biddiscombe

That was actually $15 million or something.

Jean Hu

Yeah, $14 million, $15 million that’s typically it happens during the year for the time, but this time it happened in the last quarter.

Mark Moskowitz - JPMorgan

All right, thank you, Simon and Jean from QLogic.

Simon Biddiscombe

Mark, thanks for the time. I appreciate your time as well. Thank you.

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.

THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.

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

Source: QLogic's CEO Presents at JPMorgan TMT Conference Call (Transcript)
This Transcript
All Transcripts