mercredi 29 janvier 2014

Cavium Management Discusses Q4 2013 Results - Earnings Call Transcript


Executives


Angel Atondo - Marketing Manager


Syed B. Ali - Founder, Chairman, Chief Executive Officer and President


Arthur D. Chadwick - Chief Financial Officer, Principal Accounting Officer, Vice President of Finance & Administration and Secretary


Analysts


Blayne Curtis - Barclays Capital, Research Division


Sundeep Bajikar - Jefferies LLC, Research Division


John Spencer Ahn - JP Morgan Chase & Co, Research Division


N. Quinn Bolton - Needham & Company, LLC, Research Division


Anil K. Doradla - William Blair & Company L.L.C., Research Division


Richard E. Schafer - Oppenheimer & Co. Inc., Research Division


Sujeeva De Silva - Topeka Capital Markets Inc., Research Division


Sanjay Chaurasia - Nomura Securities Co. Ltd., Research Division


Dean Grumlose - Stifel, Nicolaus & Co., Inc., Research Division


Hans C. Mosesmann - Raymond James & Associates, Inc., Research Division


Ruben Roy - Mizuho Securities USA Inc., Research Division


Betsy Van Hees - Wedbush Securities Inc., Research Division


Gary W. Mobley - The Benchmark Company, LLC, Research Division


Vijay Bhagavath - Deutsche Bank AG, Research Division




Cavium (CAVM) Q4 2013 Earnings Call January 29, 2014 5:00 PM ET


Operator


Good day, ladies and gentlemen, and thank you for standing by. Welcome to the Cavium Inc. Fourth Quarter 2013 Earnings Conference Call. [Operator Instructions] This conference is being recorded today, Wednesday, January 29, 2014.


I would now like to turn the conference over to our host, Ms. Angel Atondo, Senior Marcom Manager. Please go ahead, ma'am.


Angel Atondo


Thank you. Good afternoon, everyone, and welcome to Cavium's fourth quarter 2013 financial results conference call. Leading the call today are Mr. Syed Ali, President and CEO of the company; and Art Chadwick, Vice President and Chief Financial Officer. Before we begin, I would like to remind you that various remarks that we make on this call, including those about our future financial results, including revenues, gross margins, operating expenses, design wins, product plans, our competitive situation, market trends and our anticipated growth and profitability, all constitute forward-looking statements for the purpose of the Safe Harbor provisions under the Private Securities Litigation Reform Act. These forward-looking statements and all other statements that may be made on this call that are not historical facts are subject to a number of risks and uncertainties that may cause actual results to differ materially. We refer you to our most recent Form 10-K and Form 10-Q filed with the SEC, in particular, to the section entitled Risk Factors, and to other reports that we may file from time to time with the SEC for additional information on factors that could cause actual results to differ materially from our current expectations. These forward-looking statements speak only as of the date hereof and we disclaim any obligation to update these forward-looking statements.


In addition, Cavium reports gross margin, operating expenses, net income from operations, net income and EPS in accordance with GAAP, and additionally, on a non-GAAP basis. Management believes that non-GAAP information is useful because it can enhance the understanding of the company's ongoing economic performance, and Cavium, therefore, uses non-GAAP reporting internally to evaluate and manage the company's operations. Cavium has chosen to provide this information to investors to enable them to perform comparisons of operating results in a manner similar to how the company analyzes its operating results.


The full reconciliation of the GAAP to non-GAAP financial data can be found in our earnings release issued earlier today. And we ask that you review it in conjunction with this call. Additionally, the information we provide on this call regarding sales by market segment involves certain management judgment as to which market each sale is assigned, and you should consider that when analyzing such information.


I will now turn the call over to Syed Ali. Syed?


Syed B. Ali


Thanks, Angel, and thanks to everyone for joining us today. Q4 was another excellent quarter on major metrics that we track, including top line growth, non-GAAP gross margins, non-GAAP operating margins, cash flow and design wins.


In brief, Cavium's fourth quarter revenue was $81.13 million, up 3% sequentially and up 22% year-over-year. This was our seventh quarter in a row of sequential revenue growth. Non-GAAP gross margins were 66.3%, and non-GAAP net income was $17.4 million or $0.31 per diluted share. GAAP net income was $187,000. Art will discuss our Q4 financial results and Q1 guidance in more detail shortly.


In Q4, growth was driven by chip sales into our core service provider, Enterprise and Data Center markets. Sales into the broadband and consumer markets were down double-digit. In Q4, we saw growth in the Enterprise and Data Center markets with strength in enterprise switches, routers and security equipment. While Data Center growth was driven by our LiquidIO cards. In the service provider market, we saw continued strength in wireless infrastructure market being offset by weakness in wireline and telecom market.


Sales into the broadband and consumer were down as sales into discontinued legacy consumer segments trail down.


In Q4, our top customer was once again Cisco Systems, which came in at 19% of sales and was up 8% sequentially. In Q4 2013, Cisco was up 36% compared to Q4 of 2012. Growth was primarily driven by RAMs of newer designs going to production.


I would now like to provide an update on Q4 design wins. Q4 capped a record year for design wins for Cavium with wins across the Enterprise, data center and service providers segments. Design wins in the quarter were specially strong in wireless infrastructure, security and our new 28-nanometer OCTEON 70 and 71XX products.


Overall, in 2013, we saw a reacceleration in design win momentum, driven both by our existing products and the newer products in our product line like our 28-nanometer OCTEON III, LiquidIO, Fusion small cell and NEURON products. Design wins in 2013 in our core products were up by greater than 50% year-over-year compared to 2012. We believe this shows the strength of our product line up and our strengthening competitive position in our served end markets.


Now I'd like to give an update on other new highlights since our last earnings call. Our new 28-nanometer, OCTEON III 70, 71XX products, have been sampled extensively to a very wide range of customers and applications, shipping in the thousands of units already. These processes integrate high-performance MIPS64 cores, with full hardware virtualization, deep packet inspection, packet processing, security and QoS capabilities, along with very sophisticated power management in a highly integrated system on a chip, with leading performance per watt and performance per dollar metrics. This product line has been our best product line to date in terms of total customer pipeline and speed of ramp up of design wins post product sampling.


We already have over 20 customers for this product and have won over 35 design wins in a very short time period. We're especially pleased with winning a number of sockets that in the previous generation was being serviced by incumbent competitors.


As I've already pointed out in previous earnings calls, this product line adds a substantial new serviceable TAM for Cavium. And we believe that we have a winning product line, which will enable us to gain a significant market share in this segment. We are also seeing a significant increase in design win traction for our OCTEON and NITROX products for robust network security, driven by the latest hacks and overarching concerns about securing networks and the rise of multi-tenant cloud data centers. This is driving our customers to add robust line rate high-performance security to their equipment. This in turn is driving increased content design wins for next-gen systems at both existing customers like VA Networks, Cisco, Dell and others.


We're also seeing our products being designed into other Tier 1 customers that we have never sold to before. We see security as a long-term secular tailwind for us, as we have a very strong leadership presence in this market. We're also seeing an acceleration in traction for our OCTEON Fusion small cell products. Issues related to coexistence with macro base stations, backhaul and physical site procurements that slowed the deployment ramp are being resolved rapidly. And we are seeing a significant uptick of interest by service providers to deploy them.


The value proposition that we are bringing to market compared to our competitors is high-performance, high user count up to 128 users, and an extremely robust fi [ph] and layer 2, layer 3 software stack along with low-cost, form factor -- small form factor reference designs that have performed excellently in our ongoing trials.


We also expect design wins for our NEURON family to accelerate in 2014, with the release of new network processors with native optimized support for our NEURON chips. We also completed [indiscernible] V valves [ph] at 2 new Tier 1 customers who have now certified and ready for use as a legacy standard TCAM replacement.


Project Thunder developments are continuing in full swing and are on track on silicon software and system development schedules. Project Thunder provides a highly scalable family of 64-bit ARMv8 processors incorporated into a highly differentiated SOC architecture that integrates high-performance compute, networking, security and storage, along with targeted workload application acceleration and high-speed industry standard I/Os fully optimized for next-generation cloud and data center applications. We have a number of customers developing software on our Thunder SDK. The developed software will be able to run out-of-the box in our silicon.


Customer feedback continues to be extremely positive. Our expectation is that our solution will be the most differentiated solution compared to other announced ARM-based offerings in this space, which are all pretty similar to one another.


Now I would like to move on and give a brief outlook on our served end markets for Q1. For Q1, we expect to see growth in the Enterprise, Data Center and wireless infrastructure markets. We expect wireline and telecom to be soft due to seasonality and some inventory work downs. We also expect the broadband and consumer business to be down in Q1.


On that note, I would now like to turn the call over to Art Chadwick, who will provide a detailed discussion of Q4 financial results and guidance for Q1. Art?


Arthur D. Chadwick


Great. Thanks, Syed, and thanks to all of you for joining us today. I'll first go through Q4 financial highlights and then provide guidance for the first quarter of 2014. As Syed mentioned, Q4 was another excellent quarter for us. But first, I'd like to do a quick recap of this last year. 2013 was a record sales year for the company, with sales of $304 million, a 29% increase over sales in 2012. We had record design wins and introduced a number of significant products during the year. We executed restructuring activities that reduced R&D resources, away from certain consumer products and reinvested those resources back into our core businesses. As a result, we managed to keep non-GAAP operating expense growth to only 7% year-over-year, while at the same time, significantly expanding our OCTEON and Thunder design teams, as well as increasing investment in other new programs.


Non-GAAP gross margins increased more than 300 basis points during the year, increasing from 63% in 2012, to 66% in 2013. Non-GAAP operating margins increased from 7% in 2012 to 20% in 2013, due to our top line sales growth, expanding gross margins and operating expense leverage. Non-GAAP net income more than doubled from $22.8 million in 2012, to $56.2 million in 2013. In addition, we generated very positive cash flow during the year, increase in our cash balance by more than 65% during the year.


We increased our cash balance by more than $50 million, and ended the year with $127.8 million in cash and equivalents.


For the fourth quarter just reported, we had sequential revenue growth, expanding non-GAAP gross margins, expanding operating margins, sequential non-GAAP EPS growth and positive cash flow. Revenue in Q4 was $81.135 million, up 3% sequentially and 22% year-over-year.


Sales into the enterprise service provider and data center markets were $71.1 million or 88% of total sales. This was up 5% sequentially, and up 38% year-over-year. Sales under broadband and consumer were $10.0 million or 12% of sales, down 14% sequentially, and down 32% year-over-year due to our exit from certain consumer businesses. Sales to our largest customer, Cisco, were $15.7 million or 19% of sales, up 8% sequentially.


Non-GAAP gross margins were 66.3%, a record high for the company. This was a 30 basis point sequential improvement over Q3, due primarily to favorable product mix. Non-GAAP operating expenses were $36.1 million, up 3% sequentially. Non-GAAP R&D expenses were $24.6 million, and non-GAAP SG&A expenses were $11.5 million. Non-GAAP operating income was $17.7 million or 21.9% of sales.


Income tax expense was just $120,000, which was lower than normal due to certain favorable international taxes at yearend. GAAP net income was $187,000. Non-GAAP net income was $17.4 million or $0.31 per share, compared to $16 million or $0.29 per share in Q3. To recap, we had strong non-GAAP EPS growth this entire year, going from $0.19 per share in Q1, to $0.23 in Q2, to $0.29 in Q3 and $0.31 in Q4.


As a result of actions we took related to certain consumer products, amortization of acquired intangible assets this quarter were higher than normal due to a $2.7 million onetime, noncash intangible asset impairment charge. In addition, we took a $2.9 million onetime, noncash charge due to the reduced expected life of certain consumer product licenses.


Q4 non-GAAP results exclude $17.2 million in non-GAAP adjustments, which includes the consumer-related onetime charges I just mentioned, stock-based compensation expense and other expenses, as detailed in our reconciliation between our GAAP and non-GAAP results, as reported in our press release.


During the quarter, we generated $17.6 million in cash from operations, $0.7 million from financing activities, and used $3.5 million for the purchase of property and equipment. As a result, our cash balance increased $14.8 million this quarter and ended with a balance of $127.8 million.


Inventory at the end of the quarter was $45.8 million, an 8% sequential increase due to new product builds and the staging of inventory for early 2014 shipments. Accounts receivable were $43.6 million. This equates DSOs of 49 days, which was 5 days less than DSOs of 54 days in Q3.


I'd now like to provide some more specific guidance for the first quarter of 2014. We expect revenue will be between $80 million and $83 million in the first quarter. By segment, we expect continued growth in sales and enterprise and data center markets, and a decrease in sales into broadband and consumer. We expect non-GAAP gross margins in Q1 will be approximately 66% plus or minus 25 basis points.


Non-GAAP operating expenses are expected to increase sequentially by between 2% and 3%. So at the midpoint, this would put non-GAAP operating expenses at approximately $37 million. Interest and other non-operating expenses are expected to be approximately $500,000.


Income taxes in the first quarter are expected to be approximately 4% of non-GAAP income. And taxes for the balance of 2014 are expected to be between 5% and 7% of non-GAAP income. The non-GAAP share count in Q1 is expected to be approximately 56 million shares. And based on those assumptions, non-GAAP EPS in the first quarter is expected to be between $0.27 and $0.29 per share. So in summary, we had an excellent 2013, and are looking forward to a great 2014.


And on that note, I'd like to hand the call back to Syed for a few closing comments on what we see coming in 2014.


Syed B. Ali


Thanks, Art. I would like to take this opportunity to provide a quick outlook on what we expect our growth drivers will be in 2014. We expect continued year-over-year sales growth in 2014. We expect growth across all of our core enterprise, data center and service provider markets. We expect to see very robust revenue growth over the year in the wireless infrastructure market across macro base stations, radio network controllers and EPC at all our major customers, including MSN, Samsung, Huawei and ALU, as some new model systems rollout starting in Q2 of 2014, and large infrastructure builds in Asia and others across the world ramp over the year.


In the Enterprise and Data Center segment, we expect growth to be driven by routers, switches, security equipment, server load balances. In our newer product category, we also expect extremely strong year-over-year growth from the group of newer products, including LiquidIO, small cells and NEURON products, with LiquidIO leading the charge, leading to meaningful revenue contributions in 2014. In the wired -- service wired segment, we expect growth from equipment such as core and edge routers, CMPS head ends and transport equipment. We also expect 2014 to be the most significant year for new product introduction and sampling ever in our history. We expect these new products to address major new market discontinuities and growth opportunities in the cloud, SDN and enterprise and service provider infrastructure. We believe that as 2014 progresses, we will have an extremely strong product line, addressing these new high-growth segments with breakthrough solutions, which will help drive continued growth for us.


With that, I would now like to turn over the call to the operator for questions. Operator?




Question-and-Answer Session


Operator


[Operator Instructions] Our first question comes from the line of Blayne Curtis with Barclays.


Blayne Curtis - Barclays Capital, Research Division


Syed, you've seen some pretty strong trends in Enterprise and Data Center. You mentioned meaningful revenue from LiquidIO, I'm assuming that's still a small number, but if you could maybe frame where that is today and how meaningful that could be and then what was the other drivers in Enterprise and Data Center that's strong in the December and March quarters?


Syed B. Ali


As in my prepared comments, Cisco is a good bellwether for the enterprise market, and they were up pretty nicely sequentially for us 8% quarter-over-quarter. And then, other markets like server load balancers, customers like F5, Citrix all contributed to that. Overall, for our newer products, we are targeting somewhere in the high single-digit to double-digit revenue contribution for these new generation of products, which is kind of LiquidIO, small cells and Fusion to add that much more to our revenues in 2014.


Blayne Curtis - Barclays Capital, Research Division


Okay. And then, on the service provider side, lots of people are seeing weakness in wired, maybe where -- what particular areas in wire are you seeing and do see any signs of that coming back, and then, if you could just give a little color, there's been a lot of focus on wireless, it seems like it's picking up a bit, are you seeing -- what's the trajectory of wireless into this year, are you starting to see those orders out of Asia, and if you could just provide any color, that would be helpful?


Syed B. Ali


So overall, again, wireless infrastructure seems pretty strong and I think we will see in strengthen also across the year. In terms of the wireline, it is slow and it was fairly decently broad-based, but I believe that with the numbers that we've baked into guidance for Q1 and in having some conversations with our customers for 2014 forecast, we expect that to start upticking in Q2.


Operator


Our next question comes from the line Sundeep Bajikar with Jefferies.


Sundeep Bajikar - Jefferies LLC, Research Division


First, can you tell us what portion of total revenues in 2013 was generated from applications related to software defined networking, and how should we think of that mix changing in 2014?


Syed B. Ali


Regarding at the end, the main product that we're going to put out there is LiquidIO. So that one just started to shift in Q3 and kind of ramped a bit in Q4, and it should ramp across the year in 2014. So that is -- from a standalone product viewpoint, that is our primary exposure to SDN. Obviously, our processors also go into SDN-enabled equipment, but that would be processors like the OCTEON, which have also shown very healthy growth in 2013.


Sundeep Bajikar - Jefferies LLC, Research Division


Okay. And then, amongst the SDN applications that Cavium is shipping into currently on the -- is it primarily into the service provider side, and if it's within service provider, is it more weighted towards wireline or wireless infrastructure?


Syed B. Ali


These particular products, primarily, the initial ramp is in data centers. Obviously, the product has been sampled by a number of customers, but I'm basically highlighting that the revenues are coming from data center customers and even customers who are deploying their appliance in the cloud itself.


Sundeep Bajikar - Jefferies LLC, Research Division


Okay. Great. And then, last one for me. In terms of wireless infrastructure contribution in 2013, was it somewhere between 20% to 25% of revenues, and any color you can provide on the size of said bucket now in 2013 would be great.


Syed B. Ali


So if you take a look at the overall kind of 2 big buckets, the Enterprise Data Center and the service provider. They are fast approaching parity. These to be, if you remember, a year or 2 years ago, at least 2 to 1, between Enterprise Data Center and the service provider, that gap has narrowed down fairly considerably. So the approximate revenues that we're generating are in that ballpark range if you will.


Operator


Our next question comes from the line of Harlan Sur with JPMorgan.


John Spencer Ahn - JP Morgan Chase & Co, Research Division


This is John Ahn calling in for Harlan. Great quarter. I guess, what I'm trying to figure out here is your 28-nanometer OCTEON III sounds like you're gaining some traction, I think, you mentioned 20-plus customers, you got some new sockets that you're winning from your competitors. I'm just trying to get an idea of the mix of the applications that you're seeing there. I mean, is it primarily in the base station area or is it in the enterprise, any kind of color you can give there would be great.


Syed B. Ali


Yes, it's actually the application area is pretty widespread. So it can go anything into -- from applications such as a line card processor in a big piece of equipment or it can go into a low-end SMB or VPN or router or WiFi AP. So it is a very wide range of applications, very nicely mixed between both control plane, which the one of the data center is -- I'm sorry, which the one in the line cards controller will be. So it's control plane applications, data plane applications and complete being becoming the main CPU in the lower end, low-performance box.


John Spencer Ahn - JP Morgan Chase & Co, Research Division


Okay. Great. That certainly helps a lot. And a question for Art, in terms of gross margins, I mean, you've made tremendous progress in 2013, as you've alluded to before, as well as in the operating margins, just trying to get a better idea of how you see your gross margin, as well as your operating margin profile going into 2014?


Arthur D. Chadwick


Sure. Well, we talked about this, I think, on our last call also. There's a high correlation between core count and gross margins for us. And as the world moves to higher and higher core count product, that's kind of the underlying driver towards higher gross margins for us. Having said that and looking at 2014, we have a lot of new products coming out. And those mass sets get amortized over a 12 month period and get amortized through our cost of sales. So that will temper some of that, otherwise, gross margin expansion. So what we've been saying for 2014 is that margins would be relatively flattish, they could be up a little, they could be down a little, but relatively flattish going from 2013 to 2014, primarily because we've got so many new products coming out and a lot of mass sets that are rolled through cost of sales.


Operator


Our next question comes from the line of Quinn Bolton with Needham & Company.


N. Quinn Bolton - Needham & Company, LLC, Research Division


I just wanted to come back to Blayne's question, make sure I heard your answer right on the top products, particularly LiquidIO. Did you say that by the end of 2014, you thought that the new products in aggregate would be roughly 10-ish percent sales?


Syed B. Ali


Yes. I'm talking about it for the year as a whole. So as you go towards the end of the year, obviously, that percentage should be higher than that.


N. Quinn Bolton - Needham & Company, LLC, Research Division


And the new products include OCTEON III or is it really just LiquidIO, Fusion or NEURON?


Syed B. Ali


It does not include OCTEON III, this is just new set of products that we have introduced over the last, whatever, 12 months or so.


N. Quinn Bolton - Needham & Company, LLC, Research Division


Okay. So it sounds like it exits -- 10% for the full year exits at a higher rate?


Syed B. Ali


Yes. So what I said is kind of high single-digits to double-digit for the year and exiting the year at higher than 10%.


N. Quinn Bolton - Needham & Company, LLC, Research Division


Great. And then, second question, just Open Compute was being held this week, wondering if there are any announcements or any update sort of from a supply chain perspective for ARM servers that came out at a conference that you could give us an update on, whether it's software tools, operating systems, any noteworthy announcements coming out of Open Compute from your perspective?


Syed B. Ali


I think, there was very significant announcement by ARM along with their partners, of which we are one of them in this segment, about the standardization of the ARM server profiles. So that, essentially, there would be a lot of commonality in the firmware and how the firmware behaves between the different vendors. So that, I think, is going to be a very, very good thing for the entire ARM ecosystem. The second kind of major aspect that we've seen is that even companies like, I think, Microsoft had an announcement where they are adopting alternate architect -- where they would be adopting these alternate architectures. So overall, we are starting to see a fairly significant speed up in development of the entire ecosystem, whether it's in operating system side, whether it's in the software side, whether it's in the ODM food chain itself. So overall, I think, the pace is really starting to pick up.


Operator


Our next person comes from the line of Anil Doradla with William Blair & Company.


Anil K. Doradla - William Blair & Company L.L.C., Research Division


Sounds like from your tone and the line of questioning, there seems to be obviously some inflection point as you get into the second half. Now that's obviously driven by some of the new products that you talked about, but can you handicap the puts and takes given a continued -- in the likelihood of a continued macro and wireline softness, how does this second half or starting from Q2 play out in terms of some of the macro weakness versus your new products?


Syed B. Ali


Anil, I think, as I said in my prepared comments in terms of kind of summarizing for 2014, we have some pretty strong growth drivers there. We talked about the wireless infrastructure, we talked about Enterprise Data Center, we talked about new group of products, if you will. So we feel, at this point, pretty confident that we should continue to have strong growth performance in 2014.


Anil K. Doradla - William Blair & Company L.L.C., Research Division


So when you look out, at least in Q1, when you provided your guidance of 80 to 83 million, does that assume any form of improvement on the wireline segment or does it assume it continues to be as it is?


Syed B. Ali


No, it does not assume any improvement, in fact, probably a little bit even lower than Q4.


Anil K. Doradla - William Blair & Company L.L.C., Research Division


Great. And then, finally, Syed, you've talked about your 64-bit ARM solution being differentiated, I mean, there's been all kinds of mixed reviews on Calxeda's failure, some of your peers not doing so well. Can you give us some color on how your solution is better because, on paper, everything looks good. How is it you guys differentiate with respect to your competitors?


Syed B. Ali


Without getting into specific due to -- for competitive reasons, this is a product that we have kept the features and capabilities very close to our [indiscernible] all our customers now but outside of that, and the primary reason is that this SoC that we are bringing to market or the family of products we're bringing to market is completely very highly differentiated from anything that has been announced there. Like I said in my prepared comments, most of the existing announced guys or pretty much all the existing announced products are very, very similar to each other. Our product line is very different in terms of whether it's performance, whether it's features, and obviously, the amount of integration that we are bringing to market. So we feel very confident about this particular impact of this product family in the overall market, and our customers seem to be agreeing with that.


Anil K. Doradla - William Blair & Company L.L.C., Research Division


And finally, on NEURON, you talked about new Tier 1 customers, is that in addition to a couple of Tier 1 customers you already have or...


Syed B. Ali


That is correct, Anil.


Anil K. Doradla - William Blair & Company L.L.C., Research Division


So that brings the total Tier 1 customers for NEURON at about 4?


Syed B. Ali


That is correct.


Operator


Our next question comes from the line of Rick Schafer with Oppenheimer.


Richard E. Schafer - Oppenheimer & Co. Inc., Research Division


Just a quick follow-up on the first quarter guidance, obviously, you see a little more growth there. And I know we talked about wireline being a little softer in 1Q. Can you break down maybe how much was attributable to wireline versus how much was attributable to sort of your exit or your deemphasis on the broadband and consumer business, how much of it is attributable to that? And part of that question would be how much longer do we have headwinds from that broadband and consumer business and the top line?


Syed B. Ali


Yes, Rick. So overall, we expect, in Q1, our broadband and consumer to be down in the double-digits again. But the good news of that is the total amount of revenues in that segment kind of become pretty small. So I would say, maybe this quarter or maximum next quarter, after that, we have nothing more to reduce. So we feel that, that will be behind us in the first half of 2014. Now regarding the core business itself, the core business as a whole is increasing a few percentage points from Q4 to Q1. And offsetting that with a double-digit decrease in broadband and consumer, you can see that's where the numbers came out to be slightly higher -- I'm sorry, slight sequential growth compared to Q1. But again, with the consumer business becoming relatively small right now, this should all be behind us pretty quickly.


Richard E. Schafer - Oppenheimer & Co. Inc., Research Division


Great. That's great color. And then, talking about your largest customer, I mean, it sounded like, obviously, they were pretty strong for you in the fourth quarter, have you seen any evidence of lumpiness or softness or any change in those order patterns as we move into the first quarter here this year?


Syed B. Ali


Actually, it may be surprising to many, but they are our most predictable customer at this point. Right through Q4, month 1, month 2, month 3, and then January is already over, they are just about as linear as they can be. So we're very, very happy with that.


Richard E. Schafer - Oppenheimer & Co. Inc., Research Division


And Syed, is that a function of just all the new design wins that are starting to ramp for you guys there, are you guys sort of offsetting that, offsetting any kind of softness they may be seeing as a company or any color you can give there?


Syed B. Ali


Yes, Rick, that is absolutely correct. Obviously, we have a couple of larger design wins, which are in the mid stages of their ramp, products like the CAT3k, and the lower-end CAT3k, which was recently introduced, I think, about a quarter ago. But beyond that, there's a lot of other smaller design wins that are finally going to production for us. And the overall individual design win dollars for design is small, but when they add up, they give us a nice broad forte into their newer products. So we are extremely pleased with that.


Richard E. Schafer - Oppenheimer & Co. Inc., Research Division


And just last question on that line, I think, is there any change to the -- your thought that Cisco could exit 2014 at sort of a rough $100 million kind of run rate for you guys?


Syed B. Ali


It's difficult to say exactly because there's a whole year ahead of us, it's difficult enough to forecast the next quarter. But we will definitely see good growth from Cisco year-over-year. So we continue to expect that as new designs go to production, that revenues will keep ticking up.


Operator


And we have a question from the line of Suji De Silva with Topeka.


Sujeeva De Silva - Topeka Capital Markets Inc., Research Division


Understating it's harder to forecast beyond a quarter but are you getting any sign on your wireline segment that the downturn here would be short-lived, meaning 1 quarter in terms of customer discussions or order patterns or any sense there?


Syed B. Ali


Yes, the indication that we've gotten from our largest wireline customers was -- the primary thing was if you take a look at 2012 to 2013, they had very good growth. But towards the end of the year, there was some kind of inventory work downs that they were going through, which kind of affected it. So overall, looking at it from our viewpoint, the type of run rates we are at in Q1, we believe, are fairly minimal. And in fact, for 1 out of the largest 3 customers of ours in this sector, we are actually seeing booking patterns improved very nicely in January. Now whether the other customers follow the same trend or not, we don't know at this point. But overall, I think, Q1 in our expectations right now should be kind of the bottom-up and we expect Q2 to get better from here.


Sujeeva De Silva - Topeka Capital Markets Inc., Research Division


That's very helpful. And then, similarly, for wireless looking ahead beyond 1 quarter, what do you think the shape of the ramp in 4Q, is it front-end-loaded or is it linear in wireless infrastructure, any thoughts there?


Syed B. Ali


We think that it will be fairly -- there will always be some lumpiness on a per customer basis, right? Because it depends upon their deployment schedules of their end customers that they have won. But it should be a fairly -- fairly, I would say, linear growth, if you will, quarter-over-quarter right through the year. Obviously, 1 customer for example could be high in a quarter, could be little bit lower in the next quarter, the other customer could go vice versa. So it's very difficult to predict on a quarter-over-quarter basis, but over the next few quarters, just from all the feedback that we got into our 2014 revenue planning process, customers seem very bullish on year-over-year growth.


Sujeeva De Silva - Topeka Capital Markets Inc., Research Division


Last quick question. On LiquidIO, is the design cycle typical with the overall product or is it shorter or longer design cycle for LiquidIO?


Syed B. Ali


Typical design cycles there, Suji, are lower, much lower. I would say it is probably in kind of the three-quarter range. So it's less than a year, which is about the fastest that we can go into production from initial sampling.


Operator


Our next question comes from the line of Sanjay Chaurasia with Nomura Securities.


Sanjay Chaurasia - Nomura Securities Co. Ltd., Research Division


Syed, I was wondering if you could rank order your highest growth opportunity, highest growth drivers between -- among LiquidIO, NEURON and Fusion for 2014?


Syed B. Ali


For 2014, obviously, LiquidIO will be the highest in terms of revenue, followed by small cells and NEURON kind of in the small proximal, so actually, like I said, we've seen an uptick, and if those deployments really start in the back half, could be a little bit better than NEURON.


Sanjay Chaurasia - Nomura Securities Co. Ltd., Research Division


And the next question is for Art. Art, you obviously given non-GAAP OpEx much lower than the revenue growth. Could you give any rough guidance for 2014 in terms of do we maintain this relationship or how do we think of OpEx growth versus revenue growth?


Arthur D. Chadwick


Sure. So I think, for the year in total, our objective is to maintain OpEx growth to something less than half of our top line growth. It won't be perfectly linear through the year, but for the year in total, our goal would be to keep it to about half of the top line growth. And that obviously allows us to expand our operating margin as we progress through the year, which is one of our financial goals.


Operator


Our next question comes from the line of Kevin Cassidy with Stifel, Nicolaus.


Dean Grumlose - Stifel, Nicolaus & Co., Inc., Research Division


This is Dean Grumlose, I'm calling in for Kevin. I was hoping you could clarify or provide some additional color on LiquidIO's potential customer base and applications, is this expanding your customer base or geographies or is it fitting in very closely to existing reach?


Syed B. Ali


A large portion of this customer base, at least the 1 that is in deployment, are customers that we have not sold to in the past, right, because we do not have a big presence in the data center before. So these are new customers. And secondly, also, customers where we are in evaluation are people who are providing their functionality in the cloud, for example. So it's kind of a new class of customers than we are generally selling to. But also, it is being evaluated and checked out at our existing customer base also. But the initial ramp is in the newer customer base.


Dean Grumlose - Stifel, Nicolaus & Co., Inc., Research Division


Okay. And as a follow-up. To the extent you can, is the SG&A and investment in LiquidIO something that's going to be tapering off or is it increasing over time or how should we view your ongoing investment in this area?


Syed B. Ali


I think, from an overall viewpoint, if we have the product, it's in production now, that will obviously be a next-generation product sometime in the later part of the year. But all of those development costs are factored in when Art says that, overall, on the OpEx side, that we will be at or below, half our top line growth.


Operator


And we have a question from the line of Hans Mosesmann from Raymond James.


Hans C. Mosesmann - Raymond James & Associates, Inc., Research Division


Syed, the core counts of OCTEON III, in terms of the design wins that you had so far, how do they compare to OCTEON II?


Syed B. Ali


The core counts on the lower-end products, obviously, are the same. In the higher-end products, for example, our highest-end OCTEON II was 32 cores, and our highest-end OCTEON III is 48 cores. Also the frequency on a per core basis in the OCTEON II was 1.5 gigahertz stop frequency, whereas in the OCTEON III, it's 2.5. So it's a fairly significant jump in terms of performance improvement.


Hans C. Mosesmann - Raymond James & Associates, Inc., Research Division


And then, as a follow-up. In terms of Project Thunder, can you give us a sense of when that product will tape out or when we can see sampling the general public?


Syed B. Ali


Yes, I think, we publicly said maybe in the last call or the call before that, that we will have the product in the market in the second half of this year. And once again, for competitive reasons, since we have so many different differentiations in this product compared to pretty much everything that is being announced out there, that we don't want to give our competitors -- we want to give them as shorter time to respond as possible. If you take a look at our history, for example, the OCTEON products, we used to announce them kind of way ahead of time. But on this one, since it is a very, very differentiated architecture, we believe that announcing it very close to silicon is the right strategy for us.


Operator


Our next question comes from the line of Ruben Roy with Mizuho Securities.


Ruben Roy - Mizuho Securities USA Inc., Research Division


Syed, I just want to follow-up on OCTEON III. A couple of things. One is it sounded like the design cycle time is a little bit shorter than your previous OCTEON products, and so I was wondering when we should start to expect some reasonable revenue? And then, also, you had said something about it adds a new serviceable TAM, and I'm wondering if the OCTEON III TAM, at least for the low-end products that you have now, is a completely new incremental TAM for Cavium.


Syed B. Ali


Regarding revenues for this product line, the lower end of the product line 70, 71XX, the time to revenue obviously is maybe 1 year or maybe a little bit over 1 year, so it's much shorter than a standard enterprise service provider where it could be 2 to 3 years. So we should start seeing revenues from the initial product ramp, somewhere in the second half, probably in the Q4 timeframe. I'm talking about meaningful contributions. And regarding the TAM itself, we did have lower-end products, but essentially, in OCTEON I and OCTEON II, we really took our high-end architecture and just did a very dumb cut down to address the lower end of the market. So it was not a very efficient implementation. And as such, we really didn't address probably maybe 90%, 95% of the TAM. In the 28-nanometer, we took a very different tact, where we actually designed the product to be completely optimized in terms of cost, power and features for the target market segment. So this is a fairly large TAM, and I think, one of our incumbent competitors has pretty much a very large portion of their revenue in this segment. So this will be a very nice increase in addressable TAM for us, not only in just addressable TAM, but the rate at which we're winning designs, like said in my prepared comments, is the fastest. And we won 20 customers, 35 designs in just about over a quarter, which we've never seen before. So obviously, it seems like this product has set the right performance cost and power metrics, and we are very bullish that this will take a very nice market share in this segment.


Ruben Roy - Mizuho Securities USA Inc., Research Division


Great. And then, Art, just quickly, last quarter, I think, you had touched another customer, a second customer that was over 10% of revenues, did that continue into Q4?


Arthur D. Chadwick


Some of our customers were bumping up towards that 10% range, and so we really have not announced who they are or when that happened. So the answer is we really don't -- we're not disclosing that in Q4.


Syed B. Ali


Because generally, over the last 1 or 2 quarters, the guy -- I mean, you may have a guy that kind of goes over it, but he's not there, 10% for the year. So once it becomes a consistent number, we will go ahead and announce it. Obviously, it's one of the wireless infrastructure customer. So in any quarter, in one quarter, it was X, and in another quarter it may be Y, just based on their quarterly ordering patterns.


Operator


Our next question comes from the line of Betsy Van Hees with Wedbush Securities.


Betsy Van Hees - Wedbush Securities Inc., Research Division


Congratulations on a solid quarter and guidance with the headwinds you're facing in your end markets, and then, of course, with your consumer broadband business exit. I wanted to go back to the Q1 guidance and wireline and telecom from your prepared remarks, Syed. You talked about softness due to seasonality and then you had inventory work downs, can you help us better understand how much of that is seasonality and how much is due to inventory work downs? And then, for the inventory work downs, do you think that's going to be completed in Q1 or carried over into Q2, and what's your confidence level on that?


Syed B. Ali


Yes, it's a little bit difficult for us to exactly kind of breakout how much of that is due to seasonality and how much of that is due to inventory work downs. But like I said, the overall numbers that we have planned for in Q1 are at very, very reasonable levels there. So overall, again, one of the top 3 customers, we had all 3 customers kind of trending down. One of them seems to have recovered and has started to give healthy ordering pattern in Q1. Hopefully, that's a sign that the others will also do that, but we are not betting on it.


Betsy Van Hees - Wedbush Securities Inc., Research Division


Okay. So that's baked in, too. So that's upside if that happens into your Q1 guidance?


Syed B. Ali


That is correct.


Betsy Van Hees - Wedbush Securities Inc., Research Division


Okay. And then, Art, I'm sorry if I missed this, but looking at the inventory days, so they went up to 153 from 144, are we going to look at the -- and you talked about because of the new product launches that you have coming in Q1, are we going to look at that kind of being the run rate for the next couple of quarters or will we see a trend back down into the 140 to 145 days?


Arthur D. Chadwick


Yes. So in terms of inventory days, I do expect it to drop over the next couple of quarters. We did have some new product builds at the end of the year and we have some stuff that we added to our cusp for early 2014 shipment, which is why it went up. But to answer your question specifically, the inventory days, I would expect to drop in Q1 from where it was in Q2.


Operator


Our next question comes from the line of Gary Mobley with Benchmark.


Gary W. Mobley - The Benchmark Company, LLC, Research Division


Congratulations to a nice finish to the year. I'd like to focus on some questions that Rick was asking earlier about the broadband and consumer business. I know it was less than 15% of revenue in 2013, it sounds like it will be less than 10% in the first quarter. Should we expect that business to go down to 0 at some point in 2014 or '15? And Art, you mentioned a couple of factors influencing the gross margin in 2014, neither of which was the mix coming from the broadband business, I know that business is about 10 percentage points lower gross margin than the rest. So I'm just wondering how those different factors are going to play out in the year?


Syed B. Ali


Gary, this is Syed. That's a very good question and I'm glad you asked it because I'd really want to confirm that to clarify that. In the broadband and consumer, we have 2 pieces. One is broadband and one is consumer. So the consumer segment is the one that was a lot of the video stuff and set-top box stuff, that is the portion that is coming down and will trend to become very, very small in the first half of this year. The broadband business on the other hand will start growing to contributions from OCTEON III, to 70, 71XX. A decent portion of those designs are actually going to be allocated to that segment. So this is a 1-core, 2-core, whether it's a WiFi router, VPN router, these are kind of the lower-end products. So we expect this segment will stabilize and then start growing again as the OCTEON III products launch and go into production.


Operator


[Operator Instructions] Our next question comes from the line of Brian Modoff with Deutsch bank.


Vijay Bhagavath - Deutsche Bank AG, Research Division


Syed, Vijay on behalf of Brian. Two questions for you, 1 is I'd like to get your sense on the end markets, from your guidance, could we infer that the new chip product cycles for Fusion, NEURON, et cetera, would be more of second half opportunities or you see some of those looking to inflect from the second quarter onwards?


Syed B. Ali


LiquidIO will be through the year, but the Fusion and the NEURON are second half '14.


Vijay Bhagavath - Deutsche Bank AG, Research Division


And then, also, AMD announced a development platform, also sampling of their A1100, the Seattle chip, would you see Project Thunder competitive with AMD heads-on or are you targeting a separate use case in ARM core processors?


Syed B. Ali


I think, the best thing to say there is to stay tuned, Vijay, because pauses again -- once we announce, take the wraps off the product, you will see that is a very, very different product, but it is a whole family and it will address multiple market segments in very, very effective ways. So I just like to hold off on that, there's not too much time between now and when we will be ready to take the wraps off that product.


Vijay Bhagavath - Deutsche Bank AG, Research Division


Is it a calendar '15 opportunity for you, Syed, Project Thunder?


Syed B. Ali


In terms of revenues, yes. I'm certain, I think, in answering another question, that our product will be out in the second half of '14, and then our revenues -- meaningful revenues really will be 2015.


Operator


Ladies and gentlemen, this concludes the Cavium Inc. Fourth Quarter 2013 Earnings Conference Call. If you'd like to listen to a replay of today's conference, please dial 1 (800) 406-7325, and enter access code 4659011. ATT would like to thank you for your participation. You may now disconnect.



This entry passed through the Full-Text RSS service — if this is your content and you're reading it on someone else's site, please read the FAQ at http://ift.tt/jcXqJW.





Aucun commentaire:

Enregistrer un commentaire