Marvell Technology, Inc. (MRVL) Earnings Call Transcript & Summary

Good morning, ladies and gentlemen, and welcome to Marvell's conference call announcing the acquisition of Innovium. [Operator Instructions] Please note, this event is being recorded. I would now like to turn the conference over to Mr. Ashish Saran, Vice President of Investor Relations for Marvell. Please go ahead.

Good morning, everyone, and thank you for joining us, particularly on short notice. Earlier today, Marvell announced its proposed acquisition of Innovium in an all-stock transaction. Please note that this transaction is subject to the satisfaction of customary closing conditions, including applicable victory approvals, and we expect to close by the end of calendar 2021. To discuss this transaction, I'm joined on the call by Matt Murphy, President and CEO of Marvell; and Jean Hu, Marvell's CFO. Before I turn the call over to Matt, let me provide a quick update on Marvell's Second Quarter of Fiscal 2022. Based on preliminary financial information, we expect revenue in the range of $1.065 billion, plus or minus 1.5%, the same midpoint as we guided during our earnings call on June 7, 2021, but with a narrower range. Our second fiscal quarter financial results will be released on August 26, 2021. Please note that until Marvell reports its full financial results on August 26, these preliminary revenue expectations are estimates only and could be subject to change. Our press release on the transaction is available in the Investor Relations section of our website at www.marvell.com. This conference call is being webcast live and a recording will be available via telephone playback and also archived in the Investors section of our website. As a reminder, today's call will include forward-looking statements regarding our future business performance and revenue, stock repurchase plan, the expected timing and completion of the proposed transaction with Innovium as well as the financial impact of the transaction on Marvell. These statements include risks and uncertainties that could cause our actual results to differ materially from the statements made on this call. Please refer to our press release today and our recent filings with the SEC for information on specific risk factors. After Matt's prepared remarks, we will have a question-and-answer session. Please note that during Q&A, we will only be answering questions pertaining to this acquisition announcement. With that, I will now turn the call over to Matt for his comments on our acquisition of Innovium. Matt?

Thank you, Ashish. Good morning, everyone, and thank you for joining us today. Marvell has been on a multiyear journey to become a leader in semiconductor solutions for data infrastructure. We have rapidly grown our presence in the cloud end market both through organic investment as well as our transformational acquisition of Inphi. Today, we announced the acquisition of Innovium, which broadens Marvell's Ethernet switch platform and will further accelerate our growth from cloud data centers. Marvell is a leader in Ethernet switching silicon with a strong and growing position in the enterprise and carrier segments we service with a broad portfolio of feature-rich products. Given our growing momentum in the cloud, developing dedicated high-performance Ethernet switch silicon optimized for hyperscale data centers has become a strategic priority. These cloud optimized switches have a unique set of requirements and need a different architecture from our current set of feature-rich products, which we market under our Prestera brand. Innovium was founded in 2014 with a vision to deliver breakthrough innovation and choice for cloud data center customers. The company's TERALYNX switch architecture was designed from the ground up to deliver the ultra-low latency, optimized power, high performance and innovative telemetry that are of critical importance to cloud scale data centers. Over the last few years, the company has established itself as a strong supplier in the merchant cloud semiconductor switching market. They are shipping 12.8T products in volume and have announced availability of next-generation 25.6T products. The acquisition of Innovium allows Marvell to immediately participate in the fastest-growing segment of the merchant switch market and rounds out our platform to service all segments of this market. I'm very pleased to announce that Innovium has been selected to become a significant supplier at a Tier 1 cloud customer, and we expect this will drive a strong ramp in cloud switch revenue in calendar 2022. In addition, Innovium is currently engaged with several other leading cloud data center customers in building out their future network architectures. Over the next 5 years, we estimate the data center market for merchant Ethernet switch silicon will grow from approximately $1 billion this year to $2 billion by calendar 2026 at a 15% CAGR. Marvell's scale and leading technology platform will enable the Innovium team to accelerate their road map to 51.2T and beyond in this large and fast-growing market. With the broadest technology platform in the industry, Marvell is positioned to become a semiconductor leader in cloud data centers, including leading positions in high-speed electro-optics, OCTEON-based DPUs for security, offload and acceleration, custom Arm-based server CPUs, custom ASICs, flash and HDD storage and on completion of this acquisition, high-speed Ethernet switching. The addition of Innovium will provide Marvell with incremental engineering resources to focus on cloud optimized silicon through the TERALYNX platform while continuing to drive its Prestera platform for the enterprise and carrier market. Dedicated architectures will accelerate our ongoing growth in the 5G, cloud, enterprise and automotive end markets, which all leverage our Ethernet platform. We expect the acquisition of Innovium will add approximately $150 million in incremental revenue next fiscal year. As we have done in the past with our M&A consolidations, we will apply Marvell's business model and leverage our infrastructure, supply chain and operational excellence to drive value creation. We expect the transaction to be neutral to Marvell's non-GAAP earnings per share in the first full quarter after the transaction closes and accretive in its first full fiscal year thereafter. We are excited to welcome to Marvell, Puneet Agarwal, Innovium's Chief Technology Officer and Founder, who has more than 20 years of experience in defining and architecting groundbreaking products. Along with the talented Innovium development team, who have a strong track record in the industry for delivering multiple generations of highly successful products. Innovium's CEO, Rajiv Khemani, will provide advisory services to Marvell post transaction close on a transitional basis. Let me close by thanking Rajiv for his leadership in building Innovium into such a strong cloud switch contender and welcoming the entire Innovium team to the Marvell family. Operator, let's open the lines for questions.

[Operator Instructions] The first question is from Tore Svanberg at Stifel.

Congratulations on the acquisition. Matt, could you maybe talk a little bit more about the $150 million in incremental revenue for next fiscal year as -- should we assume that, that's primarily cloud revenue? Or is there other types of revenues in there as well?

Yes. Tore, yes, it's 100% cloud revenue.

The next question is from Srinivas Pajjuri of SMBC Nikko.

Congrats, Matt. On the same topic of the revenue, I recall Innovium was on the record saying that they have about 20% to 25% share of the [ 51.2T share ] and is based on the market sizing, it seems like the opportunity could be much larger. Just curious as to -- it does seem a little bit low to me, the $150 million. I'm just wondering if that's because some of these products have -- are yet to ramp? Or how should we think about, I guess, as you look into the next few years, the larger SAM and the growth potential for this business?

Sure. Yes, actually, we're pretty excited about it. I think the market today is about $1 billion roughly for this type of product and growing pretty rapidly. It will be $2 billion just in a few years. So there's a lot of headroom to grow. And remember, the -- I think the figures you're referring to are really for 12.8T share. And so as the company, the combined company goes forward and we address broader opportunities with new technology transitions, I think that's just going to open up the opportunity even wider. So the way I would look at it is we've got $1 billion opportunity going to $2 billion, and that's going to create an exciting opportunity for the combined company. It's going to be quite massive.

And if I could have a quick follow-up. Could you put your current revenue into perspective? You mentioned Prestera for enterprise and carrier. If you could maybe talk about how big that is? And once you close the acquisition, how big this whole business is going to be, I think that would be very helpful.

Yes, we historically haven't broken out our businesses at the sort of product line level. But what I would say is we have a very substantial total overall Marvell revenue in our Ethernet platforms. And let me just give you an example. In the switch area, we've got very strong technology and market share leadership that we've grown over time. An enterprise carrier with 5G moving to Ethernet-based switching has also been a great growth area and opportunity for us. And then finally, in automotive, actually, all of those use Marvell's high-performance switches that are architected, slightly differently for each of those applications. But this has been a major line of business that we've talked about, has been consistently growing well over 10% really kind of in the teens sort of range over the last few years as an organic Marvell business. And then on top of that, in those other markets, remember, we also have a very strong position in physical layer products in copper. So that also adds to our overall Ethernet platform. So Ethernet is very strategic for us, Srini. It's a very large business for us overall. And the nice thing is when we add in Innovium, it's just -- it's a really nice fit into the current company strategy and our Switch BU, which again is managing those other developments I mentioned, that the Innovium team is going to plug right in and be able to take advantage of the broader scale. So I think I'd leave it at that. It's a significant business that's been growing very -- in a very healthy way. And I think getting into the segment of the switch area, which is the cloud segment, which is the fastest-growing subsegment of the overall Ethernet switch market, really gives us a complete portfolio to address the market with.

Next question is from Vivek Arya of Bank of America.

Another kind of follow-up on the market size and share gain. Gaining share and high-end switching sustainably has been very tough in the past, given Broadcom's lead and their views on how important the software layer is in terms of creating the stickiness of the platform. So I'm wondering beyond this first year, the $150 million that you suggested, what is the sustainable share gain road map for Innovium? Are there any inflections in the cloud? Is there anything else that's opening that space to more competition? Or are there synergies in terms of the size and the portfolio you kind of complement their products with? So talk to us about what a 3- to 5-year view of Innovium kind of looks like? And how that competitive landscape will -- might play out in those years?

Sure, Vivek. Yes, I think you've characterized that the market historically in the right way. I think what we're excited about, and we always look at this, right, whether it's in our own business in terms of organic investment or M&A. We look for market transitions. We look for inflection points. And I think if you look at the technology transitions that are happening in this particular area, it's going to accelerate. We are on this sort of steady 1.6 terabit per second, 3.2, 6.4 kind of cadence. And when you start looking at the future, there's obviously -- we've got now a 25.6% node, but even I think more exciting and will be probably larger will be 51.2. At that point, you also have the potential intersection, [ the back of ] co-packaged optics and kind of unique architectures that are going to be required for these data centers. And then there's a road map to 100T and beyond. So we see a tremendous amount of activity and development effort required to really address this inflection point. And I think that's where -- when you have inflections, that's where market share changes can happen, that's where you can drive innovation and breakthroughs. So I think it's actually a perfect timing for us with. And also with Inphi coming in, obviously, now we have very strong electro-optic capability. There's the ability to now actually provide solutions that close the link between the SerDes coming off of the switches and the SerDes inside the optics. So I think with all that in mind, there couldn't be a better time to do something like this. And as far as the stickiness factor, we've -- Innovium has done a great job there. That's why they've been able to generate the wins that they have and have the line of sight to the revenue that you're talking about next year. But that certainly only -- [ opportunally ] only grows and they have proven very strong software capability and stickiness as well as the customers' investment, which shouldn't be overlooked that they've put in, which also makes the solution sticky. So we see a very durable, long-term opportunity here, Vivek, and actually with the market sort of transitions happening in front of us, we think there's a big opportunity for Marvell plus Innovium to really innovate in this area and drive the market forward.

The next question is from Ross Seymore of Deutsche Bank.

Congratulations on the deal. I just want to talk a little bit about the $150 million in revenues and really the customer reach of Innovium. How much concentration is there in there? I know you're not going to mention who the one Tier 1 cloud customer is, but how much concentration is there in the $150 million? And perhaps more importantly, as you go forward, is the inclusion of Innovium into Marvell and the resulting scale that they have in reach, increase the odds of some revenue synergies in that regard?

Yes, Ross, the revenue for next year is fairly concentrated, with sort of -- as you normally see in these one strong lead customer pulling them through. But their pipeline is actually -- their opportunity pipeline is quite broad. And I think to the point you're making, under the Marvell umbrella, our access now and relationship and footprint with all the major cloud companies has really increased dramatically over the last few years with our own organic growth, the addition of Avera, now plus now with Inphi closing earlier this year. So those relationships now, I think, will provide an opportunity to really open up Innovium's total potential. At the end of the day, Ross, we're in a scale business, and in particular in the hyperscale and cloud area, this is just critical infrastructure for these companies and to be frank, for the world. And so the notion that I think small startups could really penetrate in a meaningful way on their own. Certainly, you can make progress, you can do a good job. But even if you look back to Cavium as an example, which was a public company, and it was much larger than Innovium. Even we found after we acquired it, that the combined opportunity set just got so much larger for Cavium products because of the combined scale, Marvell's reputation, our ability to invest. So I feel very excited about having the Innovium team plug into our footprint, whether that's access to all the IPs in Marvell, the manufacturing footprint and I think most importantly, the broad set of customer relationships we have. So yes, we expect that we can address the entire market going forward with this team and their capabilities and along with the rest of the Marvell total platform that we offer in the cloud.

The next question is from Harlan Sur of JPMorgan.

Congratulations on the acquisition, Matt. Help us understand on the Marvell side, I mean I know the Marvell team has talked about capturing some hyperscale switching ASIC wins. I actually think some of these are going to be ramping maybe potentially second half of this year and some in future years, right, using your SerDes technology and sort of the broad outline of the Prestera switching architecture. And so I guess the question is, you have all the building blocks to build a merchant hyperscale switching platform yourself. What IT or what about the Innovium TERALYNX architecture is different from some of the cloud ASIC switching programs that you've won or are developing for a few of your cloud customers?

Sure. Yes. As we -- as I think I mentioned to Ross earlier or Srini, we have a very strong Ethernet switch position today. And as you mentioned, it's based on what we call our Prestera platform. And Prestera has been highly successful in the markets I mentioned, including some level of traction in the data center area. But remember, those products are highly feature rich and really designed for -- not designed specifically for the requirements, which are much more speeds and feeds, optimized types of solutions you'd need for the cloud. And so it's fundamentally 2 different architectures, Harlan. And we've looked at over the years -- I mean, we've gotten this question for 5 years since I've been CEO, is why don't you go build a cloud switch and can't you guys do it? And why wouldn't you go there? And it's a pretty substantial investment to do it on your own. And we could -- we've looked at leveraging Prestera. We've looked at make or buy over time. What we -- excited us about Innovium is we've got -- we get an intact team that has been really the first company to actually get meaningful revenue and design wins in this area outside of the market lever. And they know what they're doing. They've been at this for 7 years. And I think with the roadmap acceleration we get where they're already sampling 25.6, 51.2 is well underway. So we get a roadmap acceleration. We get a team. And it allows our existing Prestera team to continue to focus and gain share in their areas. So that was -- had we gone off and done this on our own, which is maybe what you're suggesting -- we could have. It would have taken longer and I think it would have impacted our existing roadmap. So we decided to double down on this area and really buy the best asset that was out there. And we're thrilled with the talent of this team. The technical due diligence was really impressive. Puneet is a highly respected engineering leader in the entire semiconductor industry, and we're thrilled to have him and his team join us and really create a much bigger switch offering to compete for the long term.

The next question is from John Pitzer of Crédit Suisse.