Ciena Corporation ($CIEN)

Hello, everyone. Thank you for joining us, and welcome to Ciena's Fiscal Q2 2026 financial results conference call. [Operator Instructions] I will now hand the conference over to Gregg Lampf, Vice President of Investor Relations. Please go ahead.

Thank you, Tracy. Good morning, and welcome to Ciena's 2026 Fiscal Second Quarter Conference Call. On the call today is Gary Smith, President and CEO; and Marc Graff, CFO. Scott McFeely, Executive Adviser, is also with us for Q&A. In addition to this call and the press release, we posted to the Investors section of our website, an accompanying investor presentation that reflects this discussion as well as certain highlighted items in the quarter. Our comments today speak to our recent performance, our view on current market dynamics and drivers of our business as well as discussions of our financial outlook. Today's discussion includes certain adjusted or non-GAAP measures of Ciena's results of operations. A reconciliation of these non-GAAP measures to our GAAP results is included in today's press release. Before turning the call over to Gary, I'll remind you that during this call, we'll be making certain forward-looking statements. Such statements, including our quarterly and annual guidance, commentary on market dynamics and the discussion of our opportunities and strategy, are based on current expectations, forecasts and assumptions regarding the company and its markets, which include risks and uncertainties that could cause actual results to differ materially from the statements discussed today. Assumptions relating to our outlook, whether mentioned on this call or included in the investor presentation that we posted earlier, are an important part of such forward-looking statements, and we encourage you to consider them. Our forward-looking statements should also be viewed in the context of risk factors detailed in our most recent 10-K and our forthcoming 10-Q. Ciena assumes no obligation to update the information discussed in this conference call whether as a result of new information, future events or otherwise. As always, we'll allow for as much Q&A as possible today, though we do ask that you limit yourself to one question and one follow-up. With that, I'll turn the call over to Gary.

Thanks, Gregg, and good morning, everyone. Our Q2 performance was once again very strong, reflecting our continued technology leadership, our deep customer relationships and the strength of our business model. In the quarter, we grew the business 40% year-on-year, with revenues of $1.57 billion. We expanded adjusted gross margin to 44.9%, and we nearly quadrupled the year ago adjusted earnings per share to $1.64. And I'd remind everybody that we delivered these results while navigating unprecedented demand and a constrained supply environment. With the combination of a strong and growing backlog driving strong visibility fueled by AI-led demand from both cloud and service providers, coupled with our leading technology portfolio, we are well positioned to gain share and deliver long-term value to our customers and our owners. The breadth and depth of our portfolio positions us to intersect this market growth, as AI drives new opportunities across the WAN and in and around the data center. Our portfolio spans systems, interconnect software, and of course, services. Specifically, systems includes our optical systems as well as our routing and switching platforms. Interconnect is comprised of modules for inter and intra data center connectivity, inclusive of our WaveLogic modems and pluggables as well as co-packaged optics and critical technology components that serve as foundational network building blocks. And of course, our software and services, which helps customers install, automate, operate and optimize their networks at scale. Taken together, Ciena's portfolio delivers our customers innovative products to meet a wide variety of high-speed connectivity solutions in the WAN and in and around the data center, with an unmatched competitive offering. So since we spoke to you last in March, the largest hyperscalers have increased their 2026 capital expenditures, with indications of continued expansion into 2027 and beyond. Given the priority to monetize somewhat constrained compute investments, we expect an increasingly larger proportion of that spend will be directed towards network infrastructure. Importantly, service providers are also reinvesting in network infrastructure after several years. This is creating net new opportunities with service providers across long-haul metro and managed optical fiber networks or MOFN. In fact, they were up 28% for us year-on-year. Simply put, all customers are prioritizing high capacity, low latency and high-speed connectivity, underpinned by the need to transport data for AI, including model training, data ingestion and inference. To that end, our latest view is that the addressable market will approximately double over the next several years to roughly $50 billion by 2029. And to be clear, this significant market growth includes our traditional WAN markets and the high-growth markets in and around the data center, both of which we've been strategically investing into in recent years. With that, let me add some color on how these dynamics are driving the demand for our line system specifically. The first generation of our intelligent line system, our RLS platform, set the standard for high speed, low latency and power efficient connectivity. The large global installed base of RLS have given us years of deep insights into technology requirements, as well as significant operational expertise and integration experience across both cloud and service provider environments. The resulting collaboration from these experiences has directly informed the development of our next-generation intelligent line system, the RLS hyper-rail. This is a multi-rail solution developed for both the leading hyperscalers and service providers to specifically address the growing capacity and efficiency demands for data center interconnect, scale across architectures and inferencing. Co-created with multiple hyperscalers and built on an innovative photonic design, RLS hyper-rail supports multiple fiber pairs in parallel over hundreds of kilometers using advanced amplification. The result is significantly higher density with materially improved space and power efficiency, which is particularly important at intermediary amplifier sites where space and power is limited. Notably this morning, I am pleased to announce that we've been awarded the industry's first multi-rail order from a leading hyperscaler, validating early market demand for our RLS hyper-rail platform and cementing Ciena's position as the industry standard. We are also engaged in discussions with multiple additional hyperscalers, neoscalers and service providers, both domestically and internationally, who continue to lean in and show a level of interest which is exceeding our expectations. I also want to touch on our data center out-of-band management solution or DCOM, which combines products from our routing and switching portfolio with our industry-leading PON technology. DCOM, as you can see, is ramping extremely well, which contributed to the 88% year-on-year revenue growth in our Routing and Switching segment. We are also expanding the customer base. In addition to Meta, we've received initial orders from a second hyperscaler customer, and lab qualifications are progressing well with a third hyperscaler customer. As AI is driving demand for our systems, it is also creating momentum across our interconnects portfolio. And we are pleased to share with you that we recently secured a new win with a major hyperscaler for our high-performance Coherent modules. These will be deployed in scale across in both metro and long-haul DCI network, supporting both WAN and in and around the data center applications. This technology solution was developed in close collaboration with the customer, and marks a competitive takeaway win. I also believe it is evidence of our strategy to leverage our systems capability into modules and component forms, and addresses the broad range of consumption models that our technology can address. We also continue to see strong demand from hyperscalers for our 400 gig and 800 gig pluggables, and we remain on track to more than double our pluggable revenue from 2025. Additionally, we have another first win, with a major switch OEM to use our market-leading WaveLogic 5 and 6 nano plugs. This, again, is a demonstrable further proof point of extending go-to-market and consumption models for our technology. We will continue expanding our interconnect portfolio with strong momentum behind our Nubis assets across both scale-up and scale-out use cases. First, Nitro, our linear redriver. We received the final chip back, and it is performing extremely well. And therefore, we are on track for general availability this summer. Turning to Vesta 200 6.4T, the optical engine for CPO use cases. Over the past 90 days, we've seen increased industry momentum and demand for open ecosystems, reinforcing our hypothesis and the strategic value of the Nubis acquisition. Given the breadth and depth of our portfolio, Ciena is uniquely positioned as the only focused supplier of high-speed connectivity solutions, enabling our customers to deploy across multiple use cases with best-in-class technology, software and services, from complete systems to modules, to components. Before I close, I want to touch briefly on customer co-creation, which we've talked about in the last few quarters and referenced several times today. It is a meaningful differentiator for Ciena, exemplifying the trust that customers place in both our innovation leadership and our ability to execute complex programs across multiple technology generations. Customers bring us in early on new requirements and architectures, and they trust us for engineering systems insights and to innovate new approaches that help evolve their networks. The resulting solutions are fit for purpose and deployable both at scale and upon pace on launch. For Ciena, it sharpens our road map decisions, increases win rates, gives us visibility into demand and build highly differentiated expertise, people, processes and capabilities, and that's reflected in our growing market momentum. The bottom line is our deep customer relationships and our sustainable technology leadership support our confidence in continued share gains, durable growth and increasing profitability over the next several years. With that, I'll turn it over to Marc to walk through the quarter's financial results and our outlook. Marc?

Thank you, Gary, and good morning, everybody. Thanks for joining us this morning. As Gary noted, Q2 was another testament to strong execution meeting robust demand. Consistent with industry views that demand will remain strong for at least the next few years, we are focusing our resources to secure supply and manufacturing capacity to deliver for both our customers and our owners. Our Q2 results demonstrate the progress made against our financial priorities, while simultaneously growing the business. First, to gross margin. We achieved an adjusted gross margin of 44.9% in Q2 due to focused efforts on engineering cost reductions, mix and price optimizations. We continue to see a path towards margin expansion based in part on our technology leadership in hyper-rail and DCOM, the ramp of our interconnects and components business and value exchange opportunities. Our discipline in managing working capital has also borne fruit. Most notably, our cash conversion cycle has improved by 20 days since Q1 on faster inventory turns and better payables execution. This helped contribute to free cash flow of $219 million or 13.9% of revenue and a cash balance of $1.4 billion. Finally, we are deliberately and responsibly allocating our owners' capital. We continue to invest in the business organically to capture new opportunities, such as developing RLS hyper-rail. We're also extending our product line to different customer use cases, such as data center out-of-band management with our PON technology. Lastly, we are making key investments, both with capital and operating expenses, to secure supply for the future demand. We remain on track to spend $250 million to $275 million of CapEx. All this while we continue to return capital to our shareholders through our stock buyback, returning $83 million in Q2 at an average price of $371 per share. Going into more details on our Q2 results. As Gary noted at the top of the call, revenue reached $1.57 billion, up 40% year-on-year and $71 million over our guidance, setting another quarterly record. Our optical networking business grew 42% over Q2 '25, driven by strong demand for our RLS and Waveserver product lines, both up over 55% year-on-year. Our routing and switching business grew 88%, primarily due to the ramp of DCOM, as Gary mentioned earlier, and deployment continues at pace. And our direct cloud customer revenue grew 70% over the year ago period, with service providers growing 28%. Of particular note, our India service provider revenue more than doubled year-on-year, reflecting strong demand for MOFN deployments. We had 2 customers, both cloud providers, contribute more than 10% of our revenue. As I noted earlier, Q2 adjusted gross margin was 44.9%, exceeding our guidance by 90 basis points, and up 4 full percentage points from a year ago period. Q2 adjusted operating expense was $398 million, driving an adjusted operating margin of 19.5%, exceeding the midpoint of our guide by over 100 basis points. OpEx was elevated in Q2 on higher variable compensation due to strong year-to-date performance on revenues and orders. Otherwise, OpEx is meeting our expectations. Adjusted EPS was $1.64, nearly 4x the year ago figure, demonstrating the strong profit generating capability of our business model. Before I provide our updated full year and Q3 2026 guidance, I'd like to take a moment to comment on the demand and supply environment. In Q2, our backlog increased more than $600 million sequentially to $7.7 billion, reflecting strong demand for our products and our leadership in the market, and we expect to exit the year with even higher backlog. The combination of customer collaboration, a growing services business, robust order flows and high-quality backlog provides us with excellent visibility into 2027. Against this demand backdrop, which has been noted across the industry, we continue to see an imbalance of supply not keeping pace with demand. We are navigating these challenges as well as evidenced by our strong performance and ability to increase our outlook over the past couple of quarters. We are working with both our supply partners and customers to ensure we can serve our large and growing backlog. Specifically, we are driving to achieve a greater balance as we make the investments with our suppliers needed to ensure supply security, while focusing on economic optimization opportunities with our customers. Now on to guidance. In Q3 2026, we expect to deliver revenue of approximately $1.625 billion, plus or minus $50 million, and an adjusted gross margin of 45%, plus or minus 50 basis points, and adjusted operating expenses of approximately $410 million, plus or minus $10 million, resulting in an operating margin of 19% to 20%. Based on our first half performance and continued ability to manage through a supply-constrained environment, we are once again in a position to raise our guidance for fiscal '26. We now expect to deliver revenue for the fiscal year of $6.3 billion, plus or minus $100 million, raising our midpoint growth to 32% year-on-year. We expect our fiscal 2026 gross margin to be between 44.5% and 45%. And we expect fiscal 2026 operating expenses of approximately $1.61 billion, plus or minus $20 million, due to higher variable compensation and additional investments in supply security. And we now expect '26 operating margin of 19%, plus or minus 50 basis points. In conclusion, we had another record quarter as market dynamics continue in Ciena's favor. Visibility remains strong, well beyond our historical norms, supported by significant backlog, multiphase customer programs and our co-creation activities, and we are confident that strength will continue into 2027. We see durable and robust demand underscored by the network's critical role in enabling AI and supported by the doubling of our TAM by 2029. We have proven success, translating increased demand into higher earnings per share on the back of a strong business model and operational execution. We are confident in our ability to bring value to our customers and drive EPS growth for our owners over the next several quarters. And before I turn it over to questions, I'd like to take a moment to recognize a meaningful milestone. This call marks Gary's 100th earnings call and 25th year as CEO. With passion, a relentless customer focus and a deep belief in Ciena's team members, he's guided Ciena through significant industry transformations. As a result, today, we find ourselves in the enviable position of leading the industry in high-speed connectivity at exactly the moment the world needs it most. On behalf of Ciena's employees around the globe, congratulations, Gary, and thank you for your continued leadership. With that, we'll take questions from our sell-side analysts.

[Operator Instructions] Your first question comes from the line of Samik Chatterjee with JPMorgan.

Congrats on 25 years. Maybe just for the first one, can you talk about the multi-rail win that you are announcing with the first hyperscaler customer? What are you seeing in terms of deal size or deployment intent from them? Is that tracking relative to your -- how is that tracking relative to your expectations? And are the other discussions or engagements you have of a similar nature related to the first hyperscaler? Maybe anything you can share in terms of when -- how should we think about maturity of revenue in the near term as well on that front? And I have a follow-up.

Thank you, Samik. I appreciate that. Let me take the first part of that. I think getting an early win on this, which with the co-collaboration that we had with these hyperscalers, it's a strategic decision for them given the nature of the deployment. It really enables specifically, very high-intensity training across much greater distances with greater amplification and density. So therefore, it is a very strategic decision on their part to standardize on hyper-rail. And that will begin to be rolled out as we go through '27. And obviously, in terms of sizing, they vary. But given the nature of them, they are all hundreds of millions over multiple years. And we are engaged with most of the major players in hyperscalers on discussions for this, progressing extremely well, and we're a little ahead of where we thought we'd be from an adoption point of view.

Got it. Got it. And a quick one for Marc. Marc, maybe from an OpEx perspective, operating expense perspective, you did start the year thinking it's more flat with some of the actions that you have taken last year. This increase in the operating expense outlook for the year, which is going hand-in-hand with the increase with your revenue outlook as well, can you just parse through that? How much of that is variable compensation versus maybe investments of some sort? And how does that change? How you're thinking about a long-term OpEx trajectory for the company as well?

Yes. Thanks, Samik. Yes, I would say 90% of that increase is really driven by the higher performance than we were expecting at the beginning of the year of both orders and revenue. So as that scales, and I would expect variable compensation to scale as well. All that said, the strength of our model continues to deliver that operating leverage. So if you kind of compare where we were at the beginning of the year to where we are now, we're increasing that operating leverage even with the increase in variable comp. The rest of that delta is really around making sure that we're increasing the supply security for the demand that we're seeing. It's -- I would say that's 10-ish percent of that spend. Moving forward, I would expect us to continue to generate operating leverage and grow revenue significantly faster than OpEx, and you'll see that in further strengthening of EPS over time.

Your next question comes from the line of Simon Leopold with Raymond James.

So I mean maybe I'm mute? Can you hear me?

We can hear you now.

Can you hear me?

Yes, we do.

Okay. Sorry, I wasn't on mute. I swear. Just a quick one. If you could just give us the double click on the 10% customers. And the question I wanted to ask is a little bit of elaboration on your pricing strategy in that you've got substantial backlog. I'm wondering, one, are you able to raise price in the backlog? And how are you juggling your rising input costs versus your ability to raise product price? And how does your price hikes factor into your growth? Anything -- any insight we could get there in this kind of inflationary environment, I'd appreciate that.

Yes. No, I appreciate it, Simon. So first, in your question around the 2% or the 2 10% customers. As I said in the remarks, both cloud providers. Together, they are about 1/3 of the revenue for Q2. So we can't get into the names, but I'll leave you guys to hypothesize on who those are. On the gross margin and the pricing piece, we purposely are using this term value exchange because as we think through the supply-demand dynamics, we're trying to balance a couple of things. One is, obviously, how do we continue to expand our margin, how we make sure we're driving the right investment into securing that supply. But I also want to make sure that the other working capital activities are being taken care of, and you saw the improvement in cash conversion. So as we think about that, nothing is off the table. So we are having conversations with all of our customers around how do we balance the supply chain risk because we are making additional commitments with our suppliers to ensure that's secure supply. We're working with them to optimize which products they take out of our portfolio and making sure we've got good fill rates and we're driving what they require. And yes, we're looking at pricing opportunities across all of our products. I think the team has done a phenomenal job of taking what we see as some inflationary inputs and being able to, through really good engineering work, cost to reduce those to kind of mitigate the impacts. And you're seeing that. And I think this is our third guide for the year. We've raised gross margin each of the 3 quarters. And so I think you're seeing the fruits of that. I expect, as we continue to go through time, we have increasing confidence in being able to expand our margin, both at the gross level and at the operating level, for the reasons I talked about in the script.

Your next question comes from the line of Amit Daryanani with Evercore ISI.

Perfect. Two as well. I guess, first one, Gary, I think you spoke about in your opening comments about how the TAM could double essentially by fiscal '29 or get to at least $50 billion, I think. I think that would imply a 25%, 26% CAGR from where we are right now roughly. Could you just maybe break out and talk about how much of that TAM expansion is in and around the data center versus traditional [ van ]? And then how do we think about Ciena's ability to pick up share in that scenario?

Yes. Thank you, Amit. Yes, if you look at the sort of doubling of it, you're right about the overall growth rate, and you think about that as being -- scale across is driving a lot of that. And probably by the time we get to '29, we expect that to be about an $8 billion to $10 billion market, and that would be part of an overall long-haul metro optical transport WAN-type market of probably in excess of about $20 billion. So you can see that is a large part of it. And then you've got things like the interconnect market, which obviously is opening up in our various offerings into there. So it's obviously a confluence of those. And again, I would stress that [ TAM ], and that's evolving very quickly as we're all seeing. We obviously think in our areas, particularly like across, we can continue to take share with things like hyper-rail and the technologies we have from a modem point of view and then deploying them inside and around the data center. Obviously, as we go into the -- inside the data center, you've got some large markets there where we are the new entrants, but we believe that we can take share with our technology, and we're seeing evidence of that.

Got it. Perfect. And then maybe just a follow-up on the DCOM side. I mean if I think about this 88% growth you folks had in routing and switching, is that really all DCOM driven? Or maybe there's a way to think about DCOM versus what the baseline routing and switching business did? And then really when you think of the DCOM opportunity, do you think it's a durable multiyear attached business? Or is it more of a onetime out-of-band refers that you're benefiting from? Just kind of maybe frame of how big it is and how durable this could be.

I would say that DCOM is a large part of the growth in routing and switching. But even if you took that out, routing and switching had pretty good growth as well outside of the DCOM piece. We think that DCOM is a multiyear, multifaceted application within the hyperscalers and potentially, outside of the hyperscalers as well. We're obviously seeing that with our anchor customer in Meta. It's proving to be much larger and much wider, broader expansion than we'd anticipated. We're actually engaged with a couple of other hyperscalers. We got orders from another hyperscaler. And they all have slightly different applications for it, but it's certainly not one and done. It's going to be an evolving application that we see growing and we think -- again, it's very early days, but maybe $1 billion to $2 billion to $3 billion by the time we get to '29 is the total TAM. So this is a very important part of the inside the data center strategy for Ciena.

Perfect. Congrats on the nice numbers here.

Your next question comes from the line of Sean O'Loughlin with TD Cowen.

First time, long time. I wanted to ask one1 on the competitive environment for both scale across and maybe the more traditional land. I think your share position is well established, but do you see any of these tech transitions coming up, whether it's the more true GPU to GPU connections on the scale across or whether it's hyper-rail? Does any of that change the competitive landscape with either the traditional competitors in that market or from the sort of bottoms up side where component vendors are potentially moving up in the systems?

Yes. First of all, Sean, welcome to the party. Nice to hear your voice. Scott here. In the short term, looking back in the rearview mirror, we've certainly seen consolidation on the system -- system competitor landscape, and that's been successfully happening at a pace slower than we would have liked, but it certainly has been happening consistently over the last decade, and it's really gotten down to a few folks that have the capability and scale to address the needs of these large customers. In terms of the other side of the coin, which is sort of folks coming into -- or trying to come into the systems space from the component vendor, there are certain folks that have done that and tried to do that in the past. It's a very difficult journey, in our opinion. I mean, the moat that's there from a system vendor perspective, people think of technology components, but it's a lot deeper than that. It's how do you put those individual components together into an end-to-end system that spans thousands of kilometers and make it sing in an economic way for our customers. It's the software that goes around the control of that. It's the integration of the back office systems. It's the services to be able to service these networks 24/7, 365 days a year around the world. It's a big step function. And we've been doing that for decades, and it's -- we think it's a significant competitive advantage against that segment.

Great. And then just a quick follow-up on the gross margin side. The full year guide, 44.5% to 45%, to get to that in my model, at least, I have to downtick a little bit in the fourth quarter. Is there a message there? Or is that -- am I reading the tea leaves too closely there?

Yes. Sean, it's Marc. Yes, I think you might be reading a little too much into that. As we think through Q4, there's still 90 days to go before we get there, and we're really paying attention to a couple of things in the supply chain, right? Obviously, one is mix. The other is, as the industry continues to be constrained, we're trying to be prudent about some inflationary pressures. I wouldn't read too much into the implied Q4 yet. Let's get through Q3 and then we can talk more specifically about that here in 90 days.

Yes. Loud and clear. And congrats, Gary.

Your next question comes from the line of George Notter with Wolfe Research.

It's Taran Katta on for George. I was just curious about any traction you're seeing on the coherent light side? Maybe any incremental use cases? I would love to hear anything there.

Yes. So first of all, the coherent light is an opportunity we still see in the future. As the need for high-bandwidth communications increases, the distances of the current technology shrink and shrink and shrink. So coherent is going to find its way closer to the 4 walls of the data center. And our belief system actually is ultimately inside the data center. The vehicle for doing that would be a lighter version of coherent with the industry as you said, coherent light. We still see that as an opportunity that's going to intersect the marketplace, we believe, sort of late 2027 into '28 at data rates of 1.6 and 3.2 terabits. And our portfolio road map reflects that as shaped by our conversations with the lead cloud providers, and we can talk more about the technology in the future.

Great. And then what are you seeing on the NEO cloud side in terms of demand? How should we think about size opportunity? It sounds like you guys are starting to see more traction there as well.

Yes. Let me take that. I mean the neoscaler piece covers a very wide range of different kinds of organizations and customers. We are leaning into that. We're seeing great opportunities around them leaning into the network piece. Clearly, the neoscalers are looking to invest in the network piece, and they're doing that through either shared networks, MOFN. Some of them they're driving themselves. We have significant wins in most of the major neoscalers, so we are rolling out networks for them. Difficult to kind of size that right now, but we think that's a particularly good growth for us over the next 2 to 3 years, for sure.

One of the interesting dynamics and the engagements that we've had with them also is that they're trying to go fast, and they're trying to go fast, not necessarily staffed up some of the larger folks that have gone before them. So we're seeing great services opportunities there as well, which is a nice business for us.

Your next question comes from the line of Ruben Roy with Stifel.

Ruben? Ruben? You may want to move on and come back to Ruben.

Sorry, can you hear me?

Okay, there we go.

Yes. Now we can, yes.

Okay. Gary, I got a bunch of questions on hyper-rail. I wanted to come back to something you said about hundreds of millions of dollars over multiple years. I think you're saying that you're a bit ahead of adoption expectations. But can you help us understand the deployment pace, especially sort of how I think about sort of co-development with some of the hyperscalers, et cetera? Is this sort of lumpy project-based type of revenue rec? Or do you think it will be sort of more linear as you think about the next couple of years on multi-rail or hyper-rail, specifically?

On multi-rail, specifically, yes. I mean, we think it starts in '27, and we think it will be linear. And we think by that point, we'll be doing it with multiple hyper-railers and also service providers, too. There's some very large service providers, particularly those that are exposed to the wholesale MOFN-type market, where this kind of technology is absolutely transformative for them. So we think starting in '27, that will provide nice linear growth for us for the next few years, frankly. And given the scale that you're talking about on -- and this is not just training, that is obviously the sort of killer app for this because what it enables is greater density over greater distances and it enables different training models. And when you think about the constraint that is on compute right now, generally speaking, the networks have to go to the compute. Therefore, you're seeing -- in addition to the demand that we've seen so far, I think we're beginning to see even an increased demand because the networks got to go to compute. And when you think about that in the training context, the timing of us coming out with this hyper-rail co-created with the hyperscalers, it could not be better. And then in addition to the training, you're just talking about very long distance, high-density, low-latency, super intelligent line systems for cloud, for connecting data centers. And we've obviously talked about the training piece. You're seeing the beginning of all the inference and agentic AI, which will drive essentially cloud growth as well. And that infrastructure right now is based on RLS. That is the industry standard. So we're leveraging our expertise and technology there into hyper-rail.

Yes. Maybe just to add on, Ruben, this is Marc. Obviously, we're at $0 of revenue for hyper-rail until we introduce that later this year. But you'll see a meaningful uptick in revenue in '27 as a result of hyper-rail.

Great. Maybe as a quick follow-up, Marc, on the commentary on OpEx growth, and you're already at 19.5% adjusted operating margin on the revenue guide. Can you maybe talk about longer-term signposts? Maybe how you're thinking about potential upside from there on operating margins as you continue to see some of this new revenue come into the model starting next year?

Yes. I'm not going to really get into 2027 or maybe at this point, Ruben. But how I dimension this is we've got pretty strong backlog. We're leaving Q2 at [ 7 7 ]. I said we're going to increase that throughout the rest of this year going into 2027. We've got a pretty good map in terms of how we're going to expand gross margins in front of us. And I think what you'll see is continued operating leverage as we go into 2027 and beyond. So I think when you put all that together, the power of our business model, you're going to see a pretty meaningful EPS acceleration as we go through the next several years. And as we get closer to the end of the year, we'll give you a little more dimensioning on what we think that OpEx is going to be. But again, I would count on more operating leverage.

Your next question comes from the line of Meta Marshall with Morgan Stanley.

Congratulations. Just in terms of -- a couple of questions for me. First is just on supply chain. I think in the past, you've had more availability on the pluggable side versus the system side. But if you could just kind of give any update into where you're seeing kind of the greatest tightness? And then just on the gross margin leverage, you obviously spoke to kind of the steps that are being taken to kind of work around designs. But just in terms of kind of the ramp of the pluggables business, just how much of that is contributing to kind of the gross margin pickup we saw?

Yes. Meta, it's Marc. On the supply chain piece, yes, I mean, again, the team is doing a wonderful job, I think, of using our engineering prowess to cost reduce some of that. As we think about those things that are probably most constrained, we do have some constraints on modem, so typically like CDMs, but what's really helping us on the modem side of the business is we're the most vertically integrated supplier in the industry. And so we're able to buffer some of those supply chain challenges on the modem side. On the system side, you've heard a lot of people talk about the laser pump or the pump lasers that go into our amplifiers and into the line systems. That's something that we continue to work kind of on a daily basis across our supply chain. And those are things, as we make those investments and we put capacity in place and we're thinking about the longer-term discussions we're having with those suppliers, that's a lot of where we're spending our time. And again, I would say the team is doing a wonderful job even on just getting more units into the door so that we can sell them. As soon as we get them, they're going out the door, which has allowed us to meet and beat the revenue expectations for the year. On the gross margin leverage, I think there's a couple of things that are going on here. One is you've heard me talk a lot about the engineering cost reductions, which I think, again, the team is doing a great job on. We are having conversations, both with the supply chain on how we optimize those costs as well as with our customers on how we do that value exchange that I talked about earlier. And it's across -- it's beyond just price, although obviously, that's on the table. We're also looking at how do we balance the supply risk with our customers and our suppliers. How do we make sure we get terms so that we can manage our working capital a little more tightly. So I think we're looking across the entire scope of not just price in terms of gross margin, but how do we improve the foundation of the entirety of the business.

Got it.

Oh, sorry, I totally forgot your plug question. Sorry about that. Sorry, I get so excited about gross margin. On the plug piece, as we said, we're driving -- we expect to double our total plug revenue. At this point in time, I don't think we're seeing plugs have either a hugely negative or a hugely positive impact on our gross margin. It's obviously a component as we think about mix, but I'm not too hung up right now on the impacts of our plugs, although I am pretty pleased with the trajectory of the business.

Your next question comes from the line of Adrienne Colby with Citi.

I was hoping you could provide some more color on the service provider side. You're clearly seeing some strength in the MOFN business. I'm interested what you're seeing outside of India? And in the past, you've had a greater than 10% customer that was a service provider. I'm interested if that just dropped off because of the strength in hyperscalers or if there's some other dynamics you'd call out with service provider?

Yes. Thank you, Adrienne. Yes, as we talked about, service providers overall, were about 28% up year-on-year. I think what you're seeing there is 2 dynamics. One is they basically have underinvested in their optical infrastructure for the last -- frankly, the last 5 years. They've been very preoccupied with 5G investments. That's obviously now tailing off. And they are looking at putting their optical infrastructure up to date, and that's a development that we're seeing across the service provider landscape across the globe. The second dynamic we're seeing is really the sort of MOFN piece, which is managed optical fiber networks built explicitly for hyperscalers and for cloud players in various countries. So we're seeing both of those dynamics come into play. We think it's multiyear and very durable because as omnipotent as some of the hyperscalers are, it's tough for them to be everywhere across the globe and also the last mile in various countries, and there's regulatory issues around that as well. So that ecosystem of partnership and opportunity for service providers, I think they are leaning into when we are the beneficiary of that. I mean when you step back from all of this, it's really about the network. And generally speaking, in the context of AI and what needs to happen, it's actually under invested in for the last few years. And so across the service provider landscape, you're seeing that cascade through as well.

Yes. Maybe just to add on to that, Adrienne. You mentioned the MOFN piece. I would say across all of our service providers at 28% growth is a pretty fair representation on average of what that group is doing, both the wholesale guys who are very exposed to MOFN and those that aren't necessarily exposed to MOFN. And then on your question about the 10% customer, it's just math, right? Those hyperscalers are just growing at such a huge rate that they kind of cross out the others that aren't growing at that pace.

Your next question comes from the line of Tim Long with Barclays.

One question and then a follow-up. Maybe on the question -- Marc, do you want to handle this one. Backlog orders continue to be really strong. It sounds like they're going to be strong through the rest of the year. Maybe just talk a little bit about how you see that backlog playing out over the next few years in the context of -- we saw this post-COVID and it wasn't necessarily the most smooth downtick in backlog and recognition of revenues for a lot of the companies in the industry. So maybe if you can just touch on that? And then second, this definitely too, Marc. If you could touch on multi-rail. I think this is something in the past you've talked about being one of the really key areas for value exchange. So now that you're getting this to the contract level, can you talk a little bit about the profitability, gross margin profile? It sounds like this is adding a lot of value for the customers. So it might be one of the products that could have a bigger step function up in profitability for Ciena.

Tim, thanks for the questions. First on backlog, I think I would first say, it's a pretty different dynamic than we've seen after the COVID hangover. What we're seeing right now is a couple of things. One is because of the service engagements that we have and the co-collaboration that we have with each of these hyperscalers, we've got really good insight into what's going in the ground. And these things are going into the ground. They're not going into warehouses, which I think was part of the issue that we saw during the COVID era. So first and foremost, we know it's being deployed. I think the second thing I would tell you is, if we could deliver that backlog in 2026, they would take it, right? And so we sometimes get questions of, oh, hey, is this forward ordering? I actually think it's the exact opposite, right? It's actually pushing out. And if they could get that product in 2026 this year, they would, and they turned that into revenue. In fact, we get quite a bit of feedback from our customers, as you can imagine that. If we could deliver more networking capability, they'd be able to increase their revenue. So I think the dynamic is very different. And just to kind of dimension that 7.7, about [ 6 4 ] of that is hardware roughly, and we would expect about 80% of that to be delivered in the next 12 months, right? So just kind of thinking through our confidence in the delivery and how that turns into bookable revenue.

And Tim, just like looking in the rear view mirror and the lessons from the COVID experience. Signs that we were seeing during COVID of inventory buildup in our customers' warehouses, requests for pushing out their desired delivery dates or order cancellations, we are seeing absolutely none of that right now. In fact, quite the opposite.

Yes. And then on your second question, Tim, around multi-rail. Yes, a couple of things. If you start with RLS, the single rail product, over the last couple of years, we've made amazing strides in improving the margin accretion of RLS, the single rail. As we go into hyper-rail, it is a step function elevation in what we think the accretion is going to be from hyper-rail. And so you've kind of got 2 things that are happening. We think hyper-rail is going to be a bigger component of our revenue. And we think that component of revenue is going to be a much better margin for us. So I would expect as this thing ramps in '27 and beyond, that is going to provide good margin expansion at the company level, which is why we keep talking about that as part of our road map to get past that way point we've been talking about in terms of gross margin.

I'd say the early commercial conversations, including the orders that we've taken, have proven to be consistent with our expectations about that value exchange.

Your next question comes from the line of Karl Ackerman with BNP Paribas.

Two questions, if I may. First, to follow up on the multi-rail. As cloud providers transition to multi-rail, does that continue to tilt the tail towards 2/3 line systems versus pluggables, where you have a more entrenched position? And then second, is there a way to size the linear drive opportunity for you with Nubis as demand for active copper cables expands?

Yes. Let me take the first one, Karl. As you think about multi-rail, and again, I'll do the comparison with the single rail, RLS. Yes, if you think about single rail RLS in a scale across environment, that's about, call it, 100 kilometers. As you extend that, right, and you use hyper-rail that has the intermediary amplifier sites, I would expect the photonics component of that thousand kilometers to be 4 to 5x of the photonics component of that 100-kilometer radius. So I think we are going to see a shift in photonics mix as you get to longer distances for scale across an AI backbone and long haul. So yes, I think your intuition is right. On the Nubis on the redriver piece, we do think that there's some competitive advantages that being in the active copper cable gives us over AEC. We've talked about that in terms of [ 4x ] the distance are getting up to 4 meters and being about 90% of the power consumption. In terms of sizing that, for 2026, it's not going to be meaningful. As we get into '27 and into '28, you'll start to see that product ramp. We're not getting into exactly how big we think that is right now, but we expect that to be accretive going into '27 and '28.

And just a reminder on the Nitro piece, to that, it's really -- the analogy is it's a silicon model, it's a shift model. So the revenue is interesting, but what's more interesting actually is the margin levels on that in terms of the business model.

Thank you all. We appreciate you all joining us this morning. Look forward to catching up over the coming days and weeks. Thank you all, and we'll see you soon.

And this concludes our conference. Thank you for attending. You may now disconnect.