威騰電子 (WDC) 2018 Q1 法說會逐字稿

Good afternoon, and thank you for standing by.

Welcome to the Western Digital First Quarter Fiscal 2018 Conference Call.

(Operator Instructions) As a reminder, this call is being recorded.

I would now turn the call over to Mr. Bob Blair.

You may begin.

Thank you, and good afternoon, everyone.

This call will contain forward-looking statements within the meaning of the federal securities laws, including statements concerning our expected future financial performance; our market positioning; expectations regarding growth opportunities; our financial and business strategies and execution; our acquisitions, integration activities and achievement of synergy goals; demand in market trends; our product portfolio, product development efforts and expansion into new data storage markets; deleveraging plans; our joint ventures with Toshiba; and the outcome of related legal proceedings, investments in Fab 6 and supply of NAND flash memory.

These forward-looking statements are based on management's current expectations and are subject to risks and uncertainties that could cause actual results to differ materially, including those listed in our annual report on Form 10-K filed with the SEC on August 29, 2017.

Any applicable forward-looking commentary is exclusive of onetime transactions and does not reflect the effect of any acquisitions, divestitures or other transactions that may be announced after October 26, 2017.

We undertake no obligation to update our forward-looking statements to reflect new information or events.

Further, references will be made during this call to non-GAAP financial measures.

Reconciliations of the differences between the non-GAAP measures we provide during this call to the comparable GAAP financial measures will be posted in the Investor Relations section of our website.

We have not fully reconciled our non-GAAP financial measure guidance to the most directly comparable GAAP measures because material items that impact these measures are not in our control and/or cannot be reasonably predicted.

Accordingly, a full reconciliation of the non-GAAP financial measure guidance to the corresponding GAAP measures is not available without unreasonable effort.

(Operator Instructions) I thank you in advance for your cooperation.

And now, I will turn the call over to our Chief Executive Officer, Steve Milligan.

Good afternoon, and thank you for joining us.

With me today are Mike Cordano, President and Chief Operating Officer; and Mark Long, Chief Financial Officer.

After my opening remarks, Mike will provide a summary of recent business highlights, and Mark will cover the fiscal first quarter and wrap up with our guidance.

We will then take your questions.

We reported continued strong financial performance in the September quarter, demonstrating the power of our platform and underscoring the differentiated value we can deliver as a comprehensive data storage solutions leader.

For the September quarter, we reported revenue of $5.2 billion, non-GAAP gross margin of 42% and non-GAAP earnings per share of $3.56.

We generated strong operating cash flow, reflecting continued healthy demand in our end markets, most notably in our flash-based businesses.

With unabated growth in data creation leading to new challenges and opportunities for our customers, our transformation continues to resonate in the marketplace.

I would now like to spend a few minutes reviewing where we are with Toshiba.

There has been significant media coverage around this situation, and a lot of it is speculative.

So as a starting point, I want to emphasize that the JVs continue to operate efficiently and productively, which we intend to maintain.

The JVs benefit from the best talent in the industry, and we are deeply appreciative of the professionalism, focus and dedication exhibited by the teams from Toshiba and SanDisk.

From the beginning, our #1 priority has been ensuring the longevity and continued success of the joint ventures.

That is why we have invested so much time and energy into finding a resolution, one that respects our partnership and resolves this matter so we can move forward in the spirit of collaboration and innovation that has been the hallmark of this 17-year relationship.

Throughout the course of the negotiations, we made numerous allowances to meet the needs of Toshiba and other stakeholders such as lenders, customers, suppliers and government agencies.

Most notably, we withdrew from the INCJ/KKR consortium.

This eliminated our participation in TMC equity ownership, thus minimizing regulatory risk and directly addressing key concerns of TMC's management.

It also would have meant that TMC would remain under full control of Japanese stakeholders.

While we believe we provided the best potential solution to Toshiba and their stakeholders, Toshiba announced the transaction with a consortium led by SK Hynix and Bain Capital.

We have made our concerns regarding their consortium clear.

It continues to be our position that the transaction is not permitted without our consent.

That leads to where we are today.

There are 2 potential paths for resolution: The stakeholders will either engage in constructive dialogue in the near future; or this matter will be resolved through the objective arbitration process.

With respect to arbitration, we are moving forward with strong momentum following our successful track record in the California courts.

On October 5, the International Court of Arbitration confirmed the 3-member arbitration panel.

Shortly thereafter, we informed the panel of our intention to seek injunctive relief to prevent Toshiba from transferring its JV interests to the SK Hynix-Bain consortium without SanDisk's consent.

The panel will set a hearing schedule soon, at which point, we will officially file our motion.

SanDisk consent rights are clear and explicit, and we therefore feel confident in our request for injunctive relief.

We expect a ruling in the first part of 2018, in advance of Toshiba's announced time frame to close the proposed transaction.

Just to be clear, we do not take -- we do not undertake litigation lightly.

We are not litigious, and it should only be a last resort, especially in the context of this joint venture relationship.

With respect to Fab 6, you may recall that over the summer, during negotiations for the first investment tranche, Toshiba announced that it would unilaterally invest in Fab 6. This was a surprise to us.

In fact, when Toshiba made its announcement, we had more meetings scheduled to further discuss our joint investment.

This was the first time that SanDisk was prevented from participating in a new fab investment.

To remind you, Toshiba's actions associated with the first tranche are already the subject of arbitration.

The negotiations regarding the second investment tranche for Fab 6 are ongoing.

Just as we did during the negotiations for the first investment tranche, we intend to jointly participate in the investment, and we are agreeing to all good faith, commercially-reasonable terms proposed by Toshiba.

However, we will not agree to terms such as SanDisk unilaterally waiving or negating its consent rights as a condition to participate, which is what Toshiba has proposed.

Consequently, at this time, we are not confident that an agreement will be reached on this next investment tranche either.

As we have noted, Toshiba's planned initial investment in Fab 6 is solely for its directly owned capacity that is outside of the JVs.

If Toshiba proceeds unilaterally with the second investment tranche, it would also be for Toshiba's directly owned capacity, not joint venture capacity.

It is also important to remember that the JVs are obligated to provide us our entitled share of flash supply through 2029.

Based on the JV agreements, we remain confident in our planned supply bit growth rate of 35% to 45% for calendar 2018 and calendar 2019, irrespective of these initial investments in Fab 6.

In closing, I want to emphasize the following: First, our board and management are focused on resolving our differences with Toshiba, whether that is through a negotiated agreement or the arbitration process; second, we are steadfast in our commitment to protect our interests and those of Western Digital's stakeholders.

We are confident in our fact-based legal positions and our right to injunctive relief; third, as I mentioned earlier, there has been a great deal of misinformation provided into the marketplace through various channels.

We expect this activity to persist, contributing to potential confusion about this situation and our legal rights.

Western Digital will continue to communicate consistently and transparently, as we did with the recently filed FAQ and through public forums like this call.

And finally, I want to reiterate that our #1 priority has been to ensure the longevity and continued success of the joint ventures.

From day 1, we have not changed this priority or our commitment to acting in the best interest of our stakeholders.

I also want to thank the outstanding Western Digital team.

You have maintained your focus and continued innovating and delivering for our customers in spectacular fashion.

With that, I will ask Mike and Mark to share the highlights of our quarter.

Thank you, Steve.

Good afternoon, everyone.

Our September quarter results were better than expected, with demand for our products remaining strong.

The joint venture fab operations in Yokkaichi continued as planned, and we made further progress in the ongoing conversion to 3D NAND technology.

In Client Devices, revenue grew nicely from the year-ago quarter, driven by increased demand for our embedded flash products and client SSDs.

Our embedded flash products, such as iNAND, gained further adoption within the mobile OEM ecosystem, and our design win pipeline for these solutions deepened further for both current and emerging growth applications.

Demand for our client SSDs grew due to increased adoption within our OEMs' product portfolios, coinciding with the expansion of our product offering.

We began ramping our 64-layer-based client SSDs for OEMs in the September quarter.

And in the December quarter, we expect to launch our 64-layer eMMC solutions for the mobile and compute markets.

Our Client Solutions revenue grew strongly from the prior year, driven by the diversity of our portfolio of HDD and flash-based products.

During the quarter, we launched 2 new mobile lifestyle products, iXpand Base and My Cloud Home, as well as the world's highest capacity microSD card at 400 gigabytes.

The strength and appeal of our retail brands, along with the broadening product portfolio, is enabling us to continue to deliver differentiated value to the global consumer marketplace.

In Datacenter Devices and Solutions, our September quarter revenue was similar to the year-ago quarter, as combined revenue growth in enterprise SSDs and capacity enterprise hard drives was offset by the expected secular decline in performance enterprise hard drives.

In capacity enterprise, as we previously commented, demand continues to be muted due to the ongoing industry-wide shortage of key components, principally DRAM, resulting in slower industry petabyte growth in calendar 2017.

As we look into calendar 2018, which is increasingly informed by our joint planning with hyperscale customers, we expect the petabyte growth will reaccelerate to the 40% annual rate.

This will be driven by planned market migration to higher capacities, such as our 12-terabyte offering, to handle the growth of varied workloads and as the component shortages ease.

In the September quarter, we saw a continued shift in capacity enterprise to 10-terabyte drives, gaining further traction with our third-generation Helium offering.

Customer qualification activities on our fourth-generation Helium offering, our 12-terabyte drive, remain on track, and we expect its commercial ramp will accelerate in the December quarter.

We are pleased to note that we have shipped more than 20 million Helium drives since our introduction of this platform in 2012.

Just over 2 weeks ago, we hosted a very successful technology event to formally announce important breakthroughs we have achieved to make microwave-assisted magnetic recording, also known as MAMR, a commercial reality.

We have been the leader in capacity enterprise with innovative solutions, the most recent being HelioSeal technology.

The commercial availability of MAMR technology is a significant next step in maintaining this leadership.

MAMR substantially leverages past investments in perpendicular magnetic recording technology, making the justification for choosing this next-generation drive technology even more compelling.

Further, an additional advantage of MAMR is that it requires almost no ecosystem changes in both our internal manufacturing processes or in customer infrastructure.

As we have previously stated, we are planning to take MAMR into production in calendar 2019.

We believe the areal density advantages that MAMR can deliver will enable us to provide Helium drives at 40 terabytes and higher capacities for this growing market in years ahead.

In the September quarter, we completed the acquisition of Tegile Systems, a leading provider of flash storage systems for the enterprise data center application.

With Tegile's IntelliFlash products focusing on fast data and our ActiveScale products addressing big data, the combined company is in a stronger position to fully address diverse customer workloads.

The Tegile acquisition also -- will also enable us to accelerate our efforts to move up the stack, provide increasingly differentiated value for our customers.

In our flash joint ventures during the September quarter, we achieved bit output crossover for 3D NAND versus 2D NAND.

Manufacturing yields of BiCS3, our 64-layer 3D NAND, continued to improve, and we met our ramp objectives of this industry-leading technology.

We are also pleased to report that our entire retail portfolio has been enabled on BiCS3 already, and we are well underway in expanding the use of this technology into our OEM offerings, positioning us to deliver the industry's richest mix of 64-layer-based products in calendar 2017.

From a flash industry standpoint, our estimate for bit growth for calendar 2017 remains at the low end of our long-term industry outlook of 35% to 45%.

For calendar 2018, we expect overall industry bit growth to continue to be in that long-term range.

Given that the secular growth drivers for flash remain strong, we continue to believe that the favorable industry conditions will persist through the first half of calendar 2018.

Furthermore, as we've previously indicated, our bit supply requirements for calendar 2018 are secure, and we are confident based on the JV agreements and our ability to achieve bit growth in calendar 2019 within our long-term range.

In closing, our strong September quarter results continue to demonstrate the power of our platform.

The various ingredients that make up this platform, including technologies, products, go-to-market capabilities and our team, are helping us better serve our diversified customer base and manage our business to the best strategic and financial outcomes.

The combination of our strong competitive position, with our capacity enterprise Helium drives and the continued ramp of our BiCS3 products, should provide a solid base to drive year-over-year revenue growth in our current fiscal year.

I will now turn the call over to Mark for the financial discussion.

Thank you, Mike, and good afternoon, everyone.

I'm very pleased with our financial performance in the September quarter.

Our team executed well across our broad array of markets, as we capitalized on our diversified product portfolio, increased gross margins and achieved cost and expense targets, all of which resulted in significant earnings growth.

We also finished the September quarter with an improved liquidity position as a result of our continued robust cash flow generation.

It is important to note that our fiscal 2017 financial performance included SanDisk's operating results for the full fiscal year.

So any year-over-year comparisons I reference today will be a direct comparison.

Our revenue for the September quarter was $5.2 billion, an increase of 10% year-over-year, driven by strong performance in each of our end markets.

Revenue in Datacenter Devices and Solutions was $1.4 billion, Client Devices was $2.7 billion and Client Solutions was $1.1 billion.

Our Datacenter business continues to be fueled largely by cloud-related storage demand.

Our September quarter revenue for Datacenter Devices and Solutions was flat year-over-year.

We saw a sustained strength from capacity enterprise hard drives and enterprise SSDs, offset by an expected decline in performance enterprise hard drives.

Client Devices revenue for the September quarter increased 13% year-over-year, primarily driven by significant growth in mobility and client SSDs.

Client Solutions revenue for the September quarter increased 16% year-over-year, mostly as a result of the strength of our valuable global retail brands in removable and other flash-based products.

Our non-GAAP gross margin was 42.3%, up 840 basis points year-over-year.

This gross margin expansion resulted from a favorable supply-demand environment for flash-based products, product cost improvements, a higher mix of flash-based revenue and the strength of our capacity enterprise HDD lineup.

Turning to operating expenses.

Our non-GAAP OpEx totaled $819 million.

This included ongoing investments in product development, go-to-market capabilities, IT transformation projects and operating expenses related to our recently acquired companies.

We continue to make progress toward our integration synergy targets while also making investments in our future capabilities.

Our non-GAAP interest and other expense for the September quarter was $200 million, inclusive of $205 million of interest expense.

Our interest expense decreased $31 million year-over-year, primarily from the repricing savings, which were partially offset by LIBOR increases.

Our non-GAAP effective tax rate for the September quarter was approximately 7%.

On a non-GAAP basis, net income in the September quarter was $1.1 billion or $3.56 per share.

On a GAAP basis, we had net income of $681 million or $2.23 per share.

The GAAP income for the period includes intangible amortization, charges related to integration activities and stock-based compensation.

Therefore, the net difference between our GAAP and non-GAAP net income is primarily a result of noncash charges.

In the September quarter, we generated $1.1 billion in operating cash flow, an increase of 158% year-over-year.

We continue to reinvest in our businesses, with $286 million spent on capital investments, resulting in free cash flow of $847 million.

We also had good working capital performance contributing to our significant operating cash flows in the quarter.

We paid the previously declared cash dividend totaling $147 million during the quarter and also declared a dividend in the amount of $0.50 per share.

We closed the quarter with cash, cash equivalents and available-for-sale securities totaling $7 billion, resulting in approximately $8 billion of liquidity available to us, including our $1 billion of undrawn revolver capacity.

Since the beginning of the fiscal year, our net debt has decreased approximately $600 million to $6.3 billion at the end of the September quarter, mostly driven by cash flow generated by the business.

We remain committed to our deleveraging plans, and we'll continue to optimize our cost of capital and capital structure while retaining sufficient liquidity and flexibility.

We remain on track to achieve our planned synergy targets from both the HGST and SanDisk integrations in the time commitments we previously established.

The combined savings of the programs have contributed to our strong financial results and validated our strategy for the acquisitions.

I will now provide our guidance for the second quarter of fiscal 2018 on a non-GAAP basis.

We expect revenue to be between $5.2 billion and $5.3 billion.

We expect gross margin to be slightly higher than the prior quarter.

Turning to operating expenses, we expect those to be approximately $830 million.

Interest and other expense is expected to be approximately $205 million.

We expect an effective tax rate in the 6% to 8% range.

Our diluted shares are expected to be approximately 309 million.

As a result, we expect non-GAAP earnings per share between $3.60 and $3.70, resulting in EPS for the calendar quarter -- or calendar year 2017 of approximately $12.50.

We believe our integrated product and technology platform is a key differentiator that will enable strong, long-term growth and profitability.

While we expect to see our normal seasonal decline in the second half of fiscal 2018, we see the opportunity to achieve revenue growth at the high end of our long-term model of 4% to 8% for fiscal 2018.

Also, based on our current business outlook and capital structure, we expect our non-GAAP earnings per share will be approximately $13 for fiscal 2018.

I will now turn the call over to the operator to begin the Q&A session.

Operator?

(Operator Instructions) Our first question comes from Mark Moskowitz with Barclays.

My only question here is really around the guidance here for 2018.

Can you give us a sense here in terms of how you expect the mix of SSD versus HDD to play out?

Will there be any sort of outsized change there?

And do you actually get a benefit as NAND pricing does come down?

Do you actually see the potential for pent-up demand for more NAND-based or SSD-based product for WD?

So Mark, let me give you some color on that.

So I think as we see ourself progressing through this fiscal year, we feel fully comfortable with placement of our bit output in the various market segments.

So the demand is there, it's sufficient to consume our bit output.

Relative to the mix and penetration of SSD, I think that has somewhat slowed based upon NAND constraints at an industry basis.

We would expect, though, the long-term penetration to kind of continue as planned, nothing sort of untoward.

Our next question will come from the line of Amit Daryanani with RBC Capital Markets.

I guess, the first question maybe to start with, on the Toshiba dynamic, if they do go forward with the transaction that they proposed, I'm curious, and you don't need to give me the exact detail, but do you think you have a viable plan B, C and D to secure bit capacity beyond the calendar '19 you've talked about?

And broadly, the cross-licensing agreement that you guys both have mentioned, the SanDisk IP, does that apply for Toshiba, even if they decide to ramp up production unilaterally without Western Digital?

Or does that cross-licensing only get held up while it's done together?

Yes, Amit, this is Steve.

I'll take the second question first.

I would prefer not to comment on that in terms of how the technology licensing arrangements work.

And then the second thing is, as I indicated in my prepared remarks, the joint venture agreements require bit output to be provided to us over a period of time out to 2029.

Got it.

I guess, if I could just maybe follow up on one thing.

Intel on their call was just talking about signing a long-term NAND agreement and getting $2 billion of prepayment for that in '17 and '18.

I'm curious, given the fact that NAND demand keeps outpacing the supply, based on even what you guys are talking about, is there a potential for you guys to enter into strategic agreements with some of your bigger customers?

And are you perhaps already doing that?

Yes, I think those sort of long-term arrangements can be a part of our ongoing practice.

So yes, we do those sorts of things.

And from time to time, prepayments would be one of the elements of those agreements.

Our next question will come from the line of Mehdi Hosseini with SIG.

Two follow-ups.

In terms of NAND availability, how are you dealing with the allocation?

Right now, we are going through a ramp-up with some of the -- by some of the smartphone OEM customers that you have.

And in that context, given the tightness, how should we think about allocation mix between enterprise SSD and embedded application?

Do any of these customers have a long-term agreement which would require you to fulfill in?

Or should we think about economics of each driving the allocation?

And I have a follow-up.

Yes, so our allocation philosophy has 2 dimensions.

Obviously, one of them you referenced, which is economic.

The other one is strategic.

So we're looking at our goal to sort of optimizing between our current and midterm financial results and our long-term objectives of growing and diversifying our customer base across multiple product segments.

So it's a complex formula, and it really is based on those 2 things.

Obviously, based upon our results, we are growing in basically all of those segments, and you can see the results of some of that activity.

Yes, I think to -- just to add to that, the diversity from a market, product and customer perspective, that we have, oh, by the way, is by design.

And the other thing is, is that it allows us -- it provides more resilience from a financial perspective to our model.

So I think that diversification concept is something that is very central to the way that we run and manage the business.

Got it.

And then looking longer term, especially with the nearline and higher-density products, when should we expect MAMR samples to be out there?

And I'm assuming that both HAMR and MAMR products are going to be out there.

And for us from outside, we're trying to figure out, when is the earliest that we could look into the market and figure out the differences and the cost benefit for each technology?

So we said in our announcement that we would have samples out late in 2018 and production in 2019, which I just commented on.

Relative to sort of our reasons for picking it, I stated in my comments, obviously, our view is the economic or commercial competitiveness of the technology made it a pretty obvious choice for us.

Obviously, we needed to get some technology breakthroughs to get confidence around our ability to achieve the areal density results required.

But once that was done, it was a clear choice for us.

One of the things I'll point you to is really our internal manufacturing processes and the ability for us to leverage the current perpendicular, our PMR-based manufacturing technologies and tooling that we have in place.

So by late '18, we should see field data that would help us compare the difference between the 2?

No, you won't -- we will ship customer samples in 2018.

We will ramp production in '19.

Our next question will come from the line of Rob Cihra with Guggenheim.

Mike, you talked about the high capacity growth reaccelerating or picking back up in 2018 after being muted in part by DRAM in 2017.

I mean, is that a -- just a sort of a DRAM availability thing?

Or is there also an indication in terms of you talking to customers, cloud builders, service providers?

I mean, are you seeing their demand looking better irrespective of component availability?

Yes, let me back up a little bit.

I think the demand has been there all along, right?

So what's happening in 2017 is not a reflection of the true demand environment.

It's certainly been muted by external or temporary factors.

And yes, our view of 2018 is based upon close planning work with our customer base, both in hyperscale and outside of that.

And do you find from your OEM customers, I mean, are they more traditional OEM?

I mean, are they taking enough high cap to offset declines in mission-critical?

Or is it really now, at this stage, it depends on what the cloud service guys do?

Well, yes.

So for us, certainly, the big consumers of capacity enterprise are the cloud providers, our hyperscalers.

But the traditional storage OEMs still buy at a substantial rate.

Relative to its impact on performance enterprise hard drives, it's really not capacity enterprise eroding that, it's SSDs and the new architectures and better cost performance you can get displacing a new system designs and new architectures, that performance tier, with flash-based product.

Our next question comes from the line of Aaron Rakers with Wells Fargo.

Just curious, as you roll up your forecast for fiscal 2018, and you -- I'm sorry, looking into calendar 2018, I should say.

I'm just curious, what assumptions -- I know you talked about long-term supply growth in that 35% to 45% range -- what assumption are you making in terms of your own capacity, ability to supply capacity growth?

And in that context, what are you assuming as far as your costs, the price erosion and your ability to take costs down as you ramp 3D, 64-layer and beyond?

Yes, Mike can take the first part, and then Mark can take the second part.

Yes, Aaron, I think we talked about our expectation for bit growth rate, which is 35% to 45%.

We would expect we will be growing in that same range.

And we expect to be able to maintain the expected cost reductions that we've talked about, which are in that longer-range model is 15% to 25%.

As we continue to convert from 2D to 3D, it's sort of at the 15% to 20% at an industry level.

Okay.

And kind of in that same context, as you look at the overall industry supply-demand dynamics, I know you talked about more or less a reiteration of favorable trends through the first half of the year, have you seen any indications at all, either positive or negative, as far as that potentially even extending further or shortening?

Yes, I think at this point, it's really too early to tell.

Certainly, we don't have any indication strongly to one side or the other at this point.

Our next question comes from the line of Wamsi Mohan with Bank of America Merrill Lynch.

Steve, it sounds like you're pretty confident in supply here in '18, '19 and beyond.

But can you give us some sense of confidence that these JVs will keep up with future NAND development, so future output at -- as the technology sort of progresses?

And will you still have the supply if arbitration proceedings potentially go against you?

And I have a follow-up.

Yes, it's in both of our best interest that we continue to advance the capabilities of the joint ventures from a technology and from a manufacturing perspective.

And so we're confident that ourselves and our joint venture partner will continue to invest appropriately in that capability going forward.

And that's independent, frankly speaking, of how the arbitration proceedings turn out.

Okay.

And can you comment on expectation of sort of NAND pricing?

How much was that up relative to your expectations of flat in September quarter?

And what is sort of embedded in expectations in your December quarter guide?

Yes, I think we talked about last quarter that ASPs had flattened.

That was our expectation coming into this quarter, and that's really our expectation as we sit today.

So we don't see continued movement in a substantial way up, but it's being sustained at this -- at the current level.

Our next question comes from the line of Joe Wittine with Longbow Research.

Just one for Mark.

On gross margin over the mid-term, so obviously, you expect supply-demand fundamentals that may, favorable through the first half, but can you just address even qualitatively how you expect the GM line to trend after that point?

And I ask mostly because The Street is still north of 40%, which is obviously solidly above the high end of your range.

So any forward-looking GM commentary would be helpful.

Sure.

And I think, certainly, as we look more near-term, we are getting some benefit from the ramp of our BiCS3 technology.

So that will give us some gross margin benefit, and that will continue into the first part of calendar '18, which is a positive margin benefit.

As we look out through and into 2018, that is our seasonally slower period in the first half for HDD.

So we're going to have some offsetting declines in gross margin there.

And I think net-net, we would expect still healthy gross margin through the rest of the fiscal year.

And -- but we wouldn't expect it to continue at the exact same level due to those seasonal factors and the mix with HDD.

Our next question will come from the line of Karl Ackerman with Cowen.

I wanted to circle back on the NAND technology road map questions that were addressed earlier today.

I understand that the Bain-led consortium must fulfill your supply agreements in place through 2029.

But do the master agreements stipulate that Toshiba and the Bain-led consortium must invest in 3D NAND technology for your existing fabs?

Because if not, it would appear that your lack of a supply agreement with Fab 6 today, where I assume the current plan is to build 96-layer 3D NAND, limits your technology road map beyond the 3D NAND tools that are installed in Fabs 3, 4 and 5 today.

Yes, I mean, as I indicated, we provided a lot of details in terms of my prepared remarks, in addition to the 8-K, and as I noted, that our #1 priority is the continued longevity and success of the joint venture.

And some of the details that you're asking, I'd prefer not to get lost in the weeds in terms of some of those questions at this point.

Our next question comes from the line of Katy Huberty with Morgan Stanley.

When do you see Data Center Solutions returning to growth?

Could that happen as early as the calendar first quarter?

And then maybe, Mark, can you comment on where you see OpEx levels exiting the fiscal year?

Sure.

I'll take the OpEx question, and then Mike can talk about the Datacenter Devices and Solutions.

In terms of OpEx, one of the things we should keep in mind is that we tend to have or we annually have a small spike in calendar Q1.

That has to do with the reset of payroll taxes.

So we would otherwise expect OpEx to be relatively consistent through the rest of the fiscal year, other than this slight increase in calendar Q1.

And I think the key point is, we will be exiting the year roughly in a position that's consistent with our long-term model, plus or minus.

And I think we're currently operating within our long-term model, much earlier than we had anticipated, the long-term model being 14% to 16% of revenue.

And so I think our continued efficiencies that we're gaining from our achieving the synergy targets and from other transformational projects we have ongoing, should allow us to continue operating in a healthy range consistent with our long-term model.

And relative to the Datacenter Devices and Systems or Solutions growth, I think 2 things, 2 factors.

I talked about exabyte growth rate relative to capacity enterprise expected to reaccelerate, as well as our continued and planned progress around enterprise SSD product offerings, puts us in a good position as we progress through 2018 calendar year to have growth as we go.

Our next question will come from the line of Vijay Rakesh with Mizuho.

Just wanted to, on the hard disk drive side, I was wondering if you could give some color on 10-terabyte, how it grew sequentially and how do you see the December quarter and looking out?

Yes, the color we'll give you is we did see significant quarter-to-quarter growth on 10-terabyte, we also saw the initial ramp on 12, and we would expect that will accelerate into the current quarter.

So the continued movement of the market to the highest capacities, in this case, 10 and 12s, is progressing very well, and we're very pleased with how that's moving in, in our business.

Got it.

And on the NAND side, when you look at the 64-layer NAND, where do you expect the mix of 64-layer by bit exiting this year?

And let's say, if you take the middle of next year, how do you see that?

And then the Toshiba spat going on, do you see any disruption in that 64-layer ramp?

Yes, I don't think we provided any numbers for calendar 2018 at this point.

And -- but we have -- we are getting more BiCS3 output than the 2D NAND at this point.

So we've realized that crossover point, and we've seen no impact of the situation with Toshiba that's changed that.

No disruption to what our expectations and plans were previously.

And we do not expect that into calendar 2018, either.

Our next question comes from the line of Stanley Kovler with Citi Research.

I just wanted to continue the discussion about the outlook.

As we think about the beginning of next year, clearly, the PC market has held up a little bit better this year.

What's your thinking about the first half of the year?

You mentioned some seasonality in your business.

Is that the area where we should see the most seasonality?

And implicitly, with gross margins staying in low 40% range, does seasonality in the first half, presumably when volumes return into maybe the second half of '18, does that imply that we get some strength back in margins as well?

Yes, I think your perspective, one of the drivers of seasonality is PCs, so yes, you're on it.

Also, the consumer segments of our business also see a seasonal trend.

That would include gaming, as an example.

And yes, obviously, when we move into the second half of the calendar year, we do see some absorption benefits that help us on the margin.

Yes, from an HDD standpoint, we'll see that.

And then for the remainder of the calendar year, the flash margins will depend on the supply-demand environment.

All right.

Sorry.

You got a follow-up question, quickly, please?

I think we're done.

Okay, all right.

I want to thank everybody for joining us today, and we look forward to speaking with you going forward.

Have a great rest of the day.

This concludes today's conference call.

Thank you for joining.

You may now disconnect.