Good afternoon, and welcome to Equinix conference call. [Operator Instructions] Also, today's conference is being recorded. If anyone has any objections, please disconnect at this time. Now I'd like to turn the call over to Katrina Rymill, Vice President of Investor Relations. You may begin.

Thank you. Good afternoon, and welcome to today's conference call. Before we get started, I'd like to remind everyone that some of the statements that we'll be making today are forward-looking in nature and involve risks and uncertainties. Actual results may vary significantly from those statements and may be affected by the risks we identified in today's press release and those identified with our filings with the SEC, including our most recent Form 10-K filed on February 26, 2013 and our most recent Form 10-Q filed on April 26, 2013. Equinix has no obligation and does not intend to update or comment on forward-looking statements made on this call. In addition, in light of Regulation Fair Disclosure, it is Equinix's policy not to comment on its financial guidance during the quarter unless it is done through an exclusive public disclosure. In addition, we'll provide non-GAAP measures on today's conference call. We provide a reconciliation of those measures to the most directly comparable GAAP measures and a list of the reasons why the company uses these measures in today's press release on the Equinix Investor Relations page at www.equinix.com. We'd also like to remind you that we post important information about Equinix on the Investor Relations page of our website. We encourage you to check our website regularly for the most current available information. With us today are Steve Smith, Equinix's CEO and President; Keith Taylor, Chief Financial Officer; and Charles Meyers, President of the Americas. Following our prepared remarks, we will be taking questions from sell-side analysts. In the interest of wrapping this call up in an hour, we would like to ask these analysts to limit any follow-on questions to just one. At this time, I'll turn the call over to Steve.

Okay. Thank you, Katrina. Good afternoon, and welcome to our second quarter earnings call. I'm pleased to report that Equinix achieved another strong quarter of financial results, and we delivered against our key operating goals in the first half of 2013. This quarter, we saw strong performance across our verticals, with particular strength in cloud and IT services and continued uptake of our global offer, with strategic wins across each of our regions. Before I provide further detail on these Q2 highlights, I'd like to provide color on our outlook for the remainder of the year as outlined in our press release. As I believe everyone on the call is aware, our prior guidance contemplated a steeper slope in our growth for the second half of 2013. We continue to expect strong operating performance and an acceleration of growth in the second half. However, based on current visibility, we're now moderating our guidance for the full year. And bridging to our revised outlook, there are 3 primary factors, 2 of which are nonoperating in nature, included in this adjustment which are depicted on Slide 3. First, we continue to operate in a volatile currency environment and are building into our forecast an additional $11 million in negative impact from currency fluctuation. This is in addition to the $21 million in currency headwinds from Q1, which we already absorbed into our full year guidance when we maintained our initial forecast for the year. Second, our revised guidance includes a $16 million noncash decrease in revenues and adjusted EBITDA related to a change in accounting estimate that extends the amortization period for nonrecurring installation revenues. This is reflective of a focused effort to move strategic customers to longer-term contracts. The majority of these customers are critical to the strength and vibrancy of…

Thanks, Steve, and good afternoon to everyone on the call. So before I turn to the Q2 results, I'd like to take this opportunity to review our progress against 5 of our critical objectives that we established for 2013 and beyond. First, regarding our interconnection strategy, our net cross connects showed nice growth this quarter, and interconnection revenues as a percent of our total recurring revenues increased to record levels in both the Americas and Asia-Pacific regions. Europe's making steady progress against this objective, with particular strength in the U.K. These efforts present themselves in our MRR per cabinet yield metric in each of our regions and over the longer term will increase the level of customer renewal and retention, thereby reducing our future MRR churn risk while also increasing our operating margin and return on invested capital. Second, our operating margins continued to improve, increasing the level of cash we generate from operations after adjusting for the REIT-related cash costs and taxes. We continue to see a path to adjusted EBITDA margins of 50%, and we'll continue to balance our growth and profitability as we scale our business. Third, we're making nice progress against some of our key strategic initiatives, including the planned REIT conversion, and other tax optimization strategies. These initiatives will decrease our overall global tax rate whilst increasing our discretionary free cash flow and AFFO, both prior to and after the REIT conversion on January 1 -- the planned REIT conversion on January 1, 2015. Fourth, as we discussed, we've been working hard to extend the term of our customer contract, particularly those key strategic customers that drive our ecosystems. For example, in North America, we lengthened new customer contracts from 1 to 2 years to 3 years or more, which we believe will extend the…

Thanks, Keith. Let me now shift gears and cover our outlook for 2013 on Slide 16. For the third quarter of 2013, we expect revenues to be in the range of $538 million to $542 million and include a negative foreign currency headwind of $4 million versus our prior guidance rates. Q3 guidance includes a $6 million decrease to both revenue and adjusted EBITDA due to the change in accounting estimate. Cash gross margins are expected to approximate 68%. Cash SG&A expenses are expected to range between $126 million and $130 million. Adjusted EBITDA is expected to be between $236 million and $240 million, which includes $11 million in professional fees related to the REIT conversion and negative foreign currency headwinds of $2 million. Capital expenditures are expected to be $180 million to $200 million, including $50 million of ongoing capital expenditures. For the full year of 2013, we expect revenue to range between $2.135 billion to $2.145 billion. As a reminder, this guidance includes a $16 million decrease to both revenue and adjusted EBITDA due to the change in accounting estimate. This is a noncash change only and a result of a longer estimated life for customer installations. Full year guidance is also adjusted for $11 million of negative foreign currency headwinds from our prior guidance range. On an FX-neutral basis and normalized for the accounting change, we expect full year revenue growth of 15.5%. Full year cash gross margins are expected to approximate 68%. Cash SG&A expenses are expected to range between $465 million and $475 million. Adjusted EBITDA for the year is expected to range between $985 million and $990 million, which includes $26 million in professional fees related to our REIT conversion and adjusted for $5 million of negative currency headwinds from prior guidance. We are guiding to a slightly higher adjusted EBITDA margin for the year as we continue to manage our discretionary and incremental spending programs. We are tightening our 2013 capital expenditure range to be between $575 million and $625 million, including $165 million of ongoing capital expenditures. We are investing in our business at a very attractive risk return profile, and we do continue to achieve our targeted returns. So in closing, we delivered a solid first half of 2013 with strong interconnection growth. We see continued momentum, and we'll execute with discipline while balancing top and bottom line growth. We are winning the right deals that enhance our vertical ecosystems and are committed to delivering firm MRR per cabinet yield, strong margins, superior returns on capital and profitable long-term growth. So let me stop there and turn it back over to you, Heather, for some questions.

Our first question comes from Jonathan Schildkraut with Evercore Partners.

I guess I'd like to get an update on what's going on with the sales force. I know that you guys were still ramping in the quarter. Maybe where the headcount is and what's going on with productivity. And then maybe if you can relay that kind of productivity commentary relative to the bookings activity in the quarter.

[indiscernible] So yes, we are continuing to ramp force. I think we are planning to add some initial heads in the second half of the year. I think that will probably take us on a worldwide basis somewhere in the 220, 230 range. In the Americas, we have added some as we continue to manage underperformers out, and we're trying to sort of keep a buffer in the system so that we can continue to do that and try to ramp the productivity. As Steve commented in the script, there are a couple of factors that are impacting productivity a bit. I think the median is still pretty strong. It dipped slightly this quarter. Averages were down, and that is attributable, really, to the deal size mix. As we've said before, when you take -- averages can be a bit deceiving because when you have large deals in the pipe, when you close those, they can spike averages. So averages were down a bit. But again, I think there's a general optimism that we're catching our -- we've got a lot of traction, we've got a record quarter in cloud and in content and digital media, solid performance in financial services and network. Network continues to deliver very strong price points, actually, strong pricing across the board. Really, some softness though in enterprise. And we talked about that. Even though we're short of our targets, I think there's some real reasons for optimism in enterprise. We're seeing solid funnel growth. Very simply, we're not converting the funnel at the same rate at our more mature verticals. And we're seeing deals slip as decision cycles are protracted. And they're more -- it's less losing deals as it is deals slipping into a subsequent quarter. And in terms of enterprise, I think it's really a drive to opportunity in many respects, but they're really the -- the enterprise are really the buyers or the buy side of the ecosystem. And they gain value from the platform by leveraging the critical mass that we have in both network and increasingly are seeing in cloud. So we're seeing a pretty irreversible trend towards hybrid cloud architectures, but it's just taking some time for that to shake out. And so our guidance really reflects a pragmatic view of the bookings productivity for that particular segment through the remainder of the year as that plays out. So that's a bit of color. Again, I think there's -- we're making good traction as we are definitely seeing bookings continue to increase from an overall trend line perspective, particularly in sort of small to midsize sweet spot for our business and consistent with our strategy.

Our next question comes from Mike Rollins with Citi Investment Research.

I was wondering if you could talk about 2 things, the first on, I believe it's Slide 15. It talks about how the older IBXs have slowed down about 4% year-over-year revenue growth rate. And I was wondering if you could talk about how investors should think about the growth of the relatively more mature data centers you have and just your latest thoughts on that. And then secondly, just going back to the sales productivity. Can you give us some more details of what went wrong relative to your expectations, particularly in Germany and the average deal size? Was it competition where you just chose to focus more on smaller deals? Or was it just the size of the funnel and being able then to take that size of the funnel and convert it was the issue?

Charles, do you want to take the first -- the second question first?

Sure. Mike, let me take the back half of the question, and then I'll give it back to Keith to take the first part. I think the direct answer to the question in terms of what happens, one is, is what I just talked about relative to the enterprise opportunity. And it's just maturing more slowly. As I said there, the funnel is strong, and I think that's actually a fairly consistent theme across the verticals. But it's not -- that funnel is not converting at the same rate, and we're seeing deals -- I'm sorry, deal cycles, sales cycles a bit more protracted. Relative to the deal mix, as we referred to, we saw a reduction in average deal size and that we were talking about that as a contributing factor. And as I've talked about a lot in previous calls, we manage the business to a certain deal mix range that we believe is going to provide us with solid growth and deliver blended returns that fit the business model. As we review the results for Q2, we saw it really was a shift towards the smaller deal sizes consistent with our ecosystem strategy. We're still seeing and winning some portion of our large deal flow, particularly for certain applications or magnet-type customers. But there's no doubt that market clearing prices for some larger deals are dropping below what we see as acceptable, particularly when we consider the opportunity cost of selling our premium capacity at lower price points. So it's really a matter of rather than chasing growth for growth's sake, we're really maintaining the deal discipline, stay focused on long-term value creation and we're just taking a pass on deals when they don't meet our hurdles. In terms of how that affects the bookings trajectory and the results, you have to remember that even a large deal portion of our bookings is fairly small, it can take 10 to 15 smaller deals to deliver the bookings of a large deal. So if we -- one drops out because we determine it's not consistent with our model, then it can take 10 or 15 smaller deals to deliver that same set of bookings. And so it takes some time for our marketing and sales execution capability to scale up to that challenge. But again, the other side of the coin on that is that this deal that's disciplined in execution is really driving the solid revenue per cap, strong new deal pricing and help the interconnection, all of which are evident in our Q2 results.

And on the middle part of that question -- Keith, you could take the last piece of that on Slide 15. Mike, on the German situation, first of all, the overall business in Europe is continuing to perform well, but it's important to keep that in context. But the forecast in the second half in Germany did weaken as we exited the quarter. A little bit tied to the German market were some softness in the German market in our primary verticals there. But as Keith mentioned in his script, we did have some sales execution challenges. We are addressing those now with sales leadership and individual performers. The enterprise slowdown was felt in that market a little bit, too, so there were some slowdown in sales cycle converting that Charles talked about in the North American market. And a little bit of softness in the financial services could have been tied to the hangover from the New York Stock Exchange-Deutsche Börse failed merger. I think that stymied the market from connecting into the activity we have in Frankfurt. So those are kind of the factors that affected the German market.

And Mike, the deal with -- the first question on the oldest assets growing 4% quarter-over-quarter -- sorry, year-over-year, I think it's important to note, number one, a lot of these assets are reaching sort of a critical level of occupancy. And so number one, a lot of that growth does come from pricing to the extent that there's some level of optimization in those older assets. It does come from new products there, some more cross connects. But one thing that I don't want to be lost on the people listening on our call here is some of these assets are highly valuable to us. And we'd rather let those assets like in LA1 -- we'd rather maintain and cordon off some of that space for our unique set of customers. And so instead of selling it to whomever, if it's a network-oriented IBX like they typically are, we're going to reserve that capacity for that future opportunity. And so sometimes that can retard the year-over-year growth. So it's fair to say that we're being disciplined about filling up those assets. Going forward, I think it's reasonable to expect a 3% to 5% year-over-year increase, and that's sort of how I think about those assets.

Our next question comes from Brett Feldman with Deutsche Bank.

You noted that you have really come out of the optimization project, which was very much concentrated in North America. And we saw a positive revenue churn response in the quarter, in line with what we had expected. But when I dig into the nonfinancial metrics for North America, we saw a slight decline in cabinets, even though revenue churn was lower across the business. And presumably, tenant quality has improved as a result of this, and yet the MRR per cabinet has been sort of flat. It didn't grow even though we've seen some more interconnection. And so I was hoping maybe you could just help us make a bit more sense of this trend and maybe give us an idea as to what appropriate expectations are for your North American business during the balance of the year.

Sure, Brett, this is Charles. Yes, I think the key thing to remember is that the -- it takes time generally when we're -- when you -- as we do some of these optimizations and customers are moving to multi-tier architectures, for example, then when you replace that revenue inherently, implementations mature over time and the yield per cabinet come back and sort of mature along a normal path. And so I think what you see is -- I actually am quite satisfied that we see -- and we've always talked about that the cabinets billing is a fairly timing of customer installations, assuming a big cab adds in Q1. It's just a more erratic number. And I think that as we look at this quarter, I'm comfortable with where it's at. I'm very pleased to see that we're keeping the yield per cabinet strong and very much towards the high end of what we've talked about in the past. And again, I think it's just a matter -- and if you look at interconnect, what we're seeing now is, I think, both a waning of the network consolidation activity that was providing pressure there and the effects of the optimization where, typically, we're trying to optimize out less interconnected business, replace that with small to midsize deals that we will later going to ramp up from an interconnection perspective. So as I look at the nonoperating metrics, I have a very high level of confidence that those are trending in the direction that we would expect and are reflective of the long-term strategies. So again, I think that's just us -- really a matter of us doing the right thing, continuing to have the deal disciplined, and I think it will take shape in the metrics over time.

So just to follow up on that. We sort of drive our models here, and I'm trying to think about the growth variables in the North American region. Or should we be increasingly, for example, be putting more weight on, say, adoption of interconnections as the reason why you grow revenues, say, versus cabinet additions? I just want to true to understand what the key growth model is in North America right now.

I don't necessarily think so. I think that our ability to add cabinets continues to be strong. Our bookings and our sweet spot are -- in fact, our bookings are growing nicely. I think there is this pressure created from us needing to perhaps replace some large deal volume with a larger number of smaller cabinet deals. And I think that's one of the reasons why as we continue to ramp up on that. But it's growing. We're definitely selling new cabinets in addition to hopefully growing the interconnection. So I definitely wouldn't see a change in that trajectory. Again, it is going to be a little bit lumpy, but I think we're going to continue to do both of those things.

Our next question comes from David Barden with Bank of America.

So Keith, I guess I want to ask the usual question. I'm trying to figure out the guidance. So if I just ignore the currencies and I ignore the accounting changes, the EBITDA -- the revenue this quarter was $532.4 million and the EBITDA was $250.8 million. So you had $13 million of sequential revenue growth, and you had $8 million of sequential EBITDA growth. If I look at the third quarter guidance, we're looking at $550 million of revenue, which is an $18 million revenue increase, and we're looking at $246 million of adjusted EBITDA, which is a negative $5 million, which implies that there's going to be $23 million of brand-new expenses showing up quarter-over-quarter, which if I look at the history of the company as far back as I go, I can't even find a quarter where you've had anything like that. So I'm trying to understand kind of how that expense profile changes so dramatically. Personally, I find it hard to believe that it's going to be that dramatic, but this is an opportunity to reset guidance and reset expectations, and you don't want to miss. So I wonder if you could kind of just walk us through that. And the second thing I would like to ask is just on the REIT stuff. If there's any inside baseball that your professional fees are buying you that gives you some color as to where the working group stands and what's next in this process and when.

Sure. Thanks for the questions, David. I think first and foremost, you sort of hit the nail on the head. And so what I will really do is just orient everybody on the call and certainly you, David. So when we started the year, we said we could do roughly $2.2 billion of revenue, we say greater than $2.2 billion. As you're aware, there's $34 million of currency and a $16 million accounting adjustment. That's a $50 million, if you will, noncontrollable movement, if you will, in our revenue line. That takes you to 21 50. We're guiding a midpoint 21 40. And the reason we're explaining the $31 million, as you know, is because based on the first quarter, the $21 million shortfall from FX, we said we could capture that. And Steve and Charles have done a good job of explaining that. So from that perspective, you then move over to EBITDA and you say, "Look, 15.5% revenue growth. Now let me look at EBITDA. What's going on there?" Well, we told you we could do 10 10. And so as you're aware, doing 10 10, if you take out the $16 million again from the accounting adjustment and you do -- and the $14 million, if you will, attributed to currency, that's $30 million. We're guiding you to 90 90 at the top end of our guidance range. If you add back those 2 items alone, you get to 10 20 And so the real shift that you're seeing here is we also told you we're going to spend $20 million in REIT costs. That number has now gone up to $26 million. So on an apples-to-apples basis, you're looking at 10 46. That's sort of EBITDA. And so how that manifests itself really in the Q3…

If I could just ask one quick follow-up. I mean, you threw this new Slide 14 into the deck to try to address, I guess, working cap -- or sorry, maintenance capital expenditures. Is that meant to be Equinix's official view of what you would bake into an AFFO calculation? And were you to provide one is kind of a 2% revenue number or are we just talking apples and oranges still?

Well, David, we certainly think we're moving down that path. Certainly, we're just trying to give you a sense of the things that we think about when it comes to maintenance CapEx and the level of reinvestment we need to make. I still think we have some work to do on the definition of AFFO. Clearly, we are working on it as you can appreciate, we're looking at our peer group. We're certainly looking at AMT, and we're looking at the Navy definition. Only when we get, if you will, the green light from the IRS do I think we'll come out and give you a very detailed and descriptive point of view on what that AFFO calculation is. Suffice it to say, though, when you look at the discretionary free cash flow, we continue to believe that's a really good surrogate right now. And overall, as some have written, there's some level of conservatism in there, given the fact that it includes all of our ongoing CapEx. So our general view is we're moving down the road, but we're not yet there -- we're not there yet to define what the AFFO metric will be. We use this more as just an interpretation. If economic life of the asset is 30 years or greater, the accounting life is different than effectively the economic life. Part of it is due to GAAP accounting. Part of it is due to interpretation. And this just gives you a pretty good idea of what our spend is going to look like.

Our next question comes from Jonathan Atkin with RBC Capital Markets.

I was wondering, given what you said about average deal size and also the price competition you're seeing per deals in larger end of the spectrum, can you put that in the context of your business suite software? And is that something that you're still expanding in Virginia and possibly into other markets? And then more broadly, I wondered if you could maybe discuss from a strategic point of view any impacts that you would expect from the efforts on the part of some of the existing data center leads to focus more on connectivity-type services.

Sure, yes. This is Charles. Let me respond to both of those. I think that -- again, business suites, as we've talked about, is something that we're having good success with and something that we think is a continued important part of our product continuum as we evolve with customers. And so as we've said, some customers, as they move to a multi-tiered architecture, depending on the size and the performance characteristics of those things, do consider a range of options. And many of them, if they're extraordinarily large in size, go to a wholesale alternative as that sort of third tier in their architecture. Many, however, depending on, again, the size and performance characteristics and their desire to have a single source provider of very often, Equinix, they are saying, "We would like you to be able to provide us with a solution, a more economic solution for a large footprint." And so business suites has been that for us. We've seen and continue to see strong funnel. We're converting that funnel successfully. And so the answer is we are absolutely looking to expand that, not only in Ashburn where we have an expansion underway now, but also in other key markets where we believe, selectively, customers will want to have the full range of infrastructure with us. So that's a little bit perspective on the business suites. And in terms of what the -- relative to other providers that may be adjacent to us in the wholesale phase, for example, focusing on connectivity, that, of course, is completely unsurprising to me, given the economics of our business and the relative appeal of the economics of our business, vis-à-vis a more raw wholesale offering. But I think when you look at it, the real value, and in terms of the amount of time that it takes to build a global reach and scale and scope of network density that we have, is simply not something that happens quickly. And so I think there are ways that they might, vis-à-vis their own offers relative to other wholesale players, be able to differentiate themselves and improve their win rate in an otherwise highly competitive environment. I think that it may help some of those players with that. But I don't really view it as, in any way, a substitute for the level of performance in network density [indiscernible].

Jonathan, I'd add on the rest of world, the way to think about the business suite question is that we'll take the learning that Charles and his team have executed in the North American market and apply a hybrid model outside of North America to focus on those types of opportunities. But again, they'll be focused on magnetic ecosystem pull-through-type deployments that have pull-through effects throughout the ecosystem and the interconnection business.

And then just real quick. On Germany, if I understood the commentary and some of the challenges there, that all relate to the legacy operations, it's not related to ancotel?

That's correct. The core issues were with the sales execution just in the verticals across the core market. The ancotel business is doing fine. The asset continues to be a differentiating force in Europe. We're attracting networks and content because of the network density there. So it's playing exactly into the ecosystem strategy as we thought it would.

And then finally, Steve, I think during your prepared remarks, if I heard it right, you talked about public cloud adoption by SMBs being a bit slower, and I wonder if you can clarify what you meant by that or if I heard that correctly.

Yes. I think what we mean by that is we're seeing uptick in the hybrid cloud deployment as the public cloud players continue to deploy across Equinix. And the biggest activity that we all read about obviously is behind AWS and the other small to medium business models out there. So there's a lot of activity there. We're seeing some of it. Some of it goes to the cloud at their location. But I think as enterprises mature, the type of workload they want to put in to source and to data centers like colo models, they're getting more and more sophisticated about understanding the hybrid model, how they can take advantage of connecting to the networks and the public cloud nodes in our data centers, and that's taking time.

Our next question comes from Frank Louthan with Raymond James. Frank G. Louthan - Raymond James & Associates, Inc., Research Division: With the lengthening of the contract periods, what sort of implication does that have sort of for your pricing power? I mean, in the past, if you'd had shorter contract lengths to see if it effectively could be raising pricing a little more quickly, does this imply that things are getting more competitive or you're trying to lock in customers for longer periods? Any change in the pricing power in your model currently?

Frank, this is Charles. No, I would say, in fact, those multi-year deals, many of them, we really had a very focused effort on sort of long-term renewals, particularly with our core strategic customers. And many of those continue to include PIs, annual PIs, which is consistent with the model and something that, I think, people have -- they understand the reasons for that, and it's typically baked into those. So no, there is -- our overall pricing remains firm. I think we look at each of those renewals. And sometimes we do, for example, say we're willing to make some concessions to get even longer durability and term in the contract, and we're willing to make those trade-offs as appropriate. But no, I think that -- I definitely don't think there's been a shift in the pricing power there. Our objective is just to make sure that we have a long-term productive relationship with critical strategic customers that are not only fundamental to our business because of their inherent scale but equally or more importantly because they're central to our developing ecosystem. Frank G. Louthan - Raymond James & Associates, Inc., Research Division: Okay. And are there any meaningful wholesale leases that you have currently on the cost side that are coming in? What's your expectation for rent increases where you do have leased space relative to what it's been the last couple of years?

Yes, Frank. Certainly, we have a number of leases. Obviously, we have almost 100 data centers around the world in different sort of -- in different incarnations, as you can appreciate. So there's certainly some that will be forthcoming over the next few years. There's no sort of sea wave of renewals that are forthcoming, and our team, our real estate team, does a very good job of getting ahead of the negotiations, working with the landlord. And certainly, some of the ones we have done in the past, yes, there's been some level of increase that's anticipated, that we've anticipated and, certainly, the landlords anticipated. But that's just baked into our operating plan and our operating model. It goes back a little bit to the comment that Charles made. There's an understanding of PI growth from the customer side. But when you look at it from our side, there's an expectation that there'll be some level of PI in our lease arrangement, and that's just a given. So we continue to work very hard in renewing them. Again, recognizing our lease costs are a relatively a small piece of our overall cost bucket and recognizing most of our leases are on a very long-term lease arrangement with multiple renewal period, we feel very good about, if you will, lease position at this stage.

And our last question comes from Sterling Auty with JPMorgan. Sterling P. Auty - JP Morgan Chase & Co, Research Division: I just want to make sure I understand. In terms of the large deal dynamics that you're talking about, are we talking about a scarcity value in those large deals? Are we talking about a competitive dynamic within those types of deals? And then separately, when you talk about the sales cycles on some of the enterprise stuff, is it because of budgeting or because of their decisions on which way they want to take their compute architecture?

Sure. So I think relative to the large deals, I would say we actually continue to see -- one of the great things about being the market leader is you have visibility to the deal flow. And we tend to get visibility to virtually all of the deal flow. And so -- and there is still plenty of and perhaps increasingly, an increasing amount of large deal opportunity out there. And there are some of those where there is a particular requirement relative to performance, latency, global reach, et cetera, that we are still well-positioned and will pursue those deals and win rates that we have become accustomed to. However, I think there is also a portion of the deal flow that is in part due to the competitive intensity of the wholesale space and market clearing prices dropping to a level that we simply view as unattractive. And I think that if you go back several years, you would say, "Okay, there were probably a portion of those because the premium, if you will, was smaller, that we would have one that I think now you're saying we're unwilling to go to the price points there and are therefore not closing this." So I think that's the general dynamic. It's slightly -- and that's causing a slight shift towards our sort of sweet spot in the small to midsize deals. But that's generally these large deal dynamics. On the enterprise side, it is not really a budgetary issue. It is really more related to the fact that these customers are taking time to sort through what exactly hybrid architectures mean to them. And they are moving from trying to figure out how to get their infrastructure out of their own basements and figuring out what portion of that goes colo, what portion of that is well-positioned to move into a public cloud setting, perhaps due to the need for reversible workloads, and what does it mean to implement a hybrid cloud. And oftentimes, as they come to grips with that, they realize that the network density that Equinix provides is a very compelling reason to sort of center their hybrid cloud infrastructure around Equinix. But that is a nontrivial sort of assessment on the part of an enterprise CIO. And anybody who's holding an enterprise realizes that those sales cycles can be protracted. And they often start with smaller deal sizes. And that's really the dynamic that we're experiencing.

That concludes our Q2 call. Thank you for joining us today.

Thank you for participating in today's conference. Please disconnect at this time.