Good day, ladies and gentlemen, and welcome to the Second Quarter 2014 Akamai Technologies Earnings Conference Call. My name is Derrick, and I'll be your operator for today. [Operator Instructions] As a reminder, this conference is being recorded for replay purposes. I would now like to turn the conference over to Mr. Tom Barth, Head of Investor Relations. Please proceed.

Thank you, Derrick, and good afternoon, and thank you all for joining Akamai's second quarter 2014 earnings conference call. Speaking today will be Tom Leighton, Akamai's Chief Executive Officer; and Jim Benson, Akamai's Chief Financial Officer. But before we get started, please note that today's comments include forward-looking statements, including statements regarding revenue and earnings guidance. These forward-looking statements are subject to risks and uncertainties and involve a number of factors that could cause actual events to differ materially from those expressed or implied by such statements. Additional information concerning these factors is contained in Akamai's filings with the SEC, including our annual report on Form 10-K and quarterly reports on Form 10-Q. The forward-looking statements included in this call represent the company's view on July 30, 2014. Akamai disclaims any obligation to update these statements to reflect future events or circumstances. As a reminder, we will be referring to some non-GAAP financial metrics during today's call. A detailed reconciliation of GAAP and non-GAAP metrics can be found under the financial portion of the Investor Relations section of our website. With that, let me turn the call over to Tom.

Thanks, Tom, and thank you all for joining us today. Q2 was another excellent quarter for Akamai on both the top and bottom lines. Revenue in the second quarter was a record $476 million, up 26% over Q2 of 2013 and coming in near the high end of our guidance range. We had solid growth in all of our geographies and in all of our major product lines, with especially strong growth for our security and media products. Non-GAAP net income for the second quarter was $106 million or $0.58 per diluted share. That's also up 26% over Q2 of 2013. This strong result exceeded the high end of our guidance range due to strong operational execution and a favorable tax rate. I'll be back in a few minutes to talk more about our achievements in the second quarter. But first, let me turn the call over to Jim to review our financial results in detail and to provide the outlook for Q3. Jim?

Thank you, Tom, and good afternoon, everyone. Akamai had a great second quarter and a very strong first half of the year. As Tom outlined, Q2 revenue came in near the high end of our guidance range at $476 million, up 5% sequentially and up 26% year-over-year with strong growth across the entire business. Looking at our performance by solution category. Our media delivery solutions revenue was $216 million in the quarter, up 1% sequentially and up 20% year-over-year. As Tom mentioned, we are very pleased with the growth in media coming off a very strong first quarter. Traffic and revenue growth accelerated across most of our customer base, with particularly strong growth coming from our gaming, video and social media customers. Again this quarter, we delivered some high-profile live events, most notably the recent World Cup matches, which set peak traffic records and contributed modestly to the media revenue achievement. More important than the media -- the revenue contributions, however, these live events highlight our unique ability to deliver high-quality live streaming content at very large scale. Turning to our performance and security solutions. Revenue was $217 million in the quarter, up 10% sequentially and up 30% year-over-year. And for the first time, we had more revenue in a quarter from performance and security solutions than Media Delivery, which speaks to the growing diversification of our revenue base. Within this solution category, we saw solid growth in our web performance solutions. And as Tom mentioned, we saw very strong growth in our cloud security solutions. Net new signings were especially strong for our security offerings, and for Prolexic, in particular, another proof point of early traction with our Prolexic integration and also increased customer demand for our overall cloud security portfolio. Finally, revenue from our services and support solutions…

Thanks, Jim. It's great to see Akamai building on our strong start to the year. I believe that our excellent financial results are evidence of the sound fundamentals across our business and the pivotal role that we play in the growth of the Internet. We are continuing to focus on solving 4 grand challenges for our customers: delivering video over the Internet with unparalleled quality, scale and affordability; providing near-instant performance for websites and apps on any device, anywhere; securing websites and data centers from attackers that seek to disrupt their operations, corrupt their content or steal their information; and driving speed and affordability into branch office connectivity for major enterprises. The complexity and challenges of delivering high-quality video over the Internet at scale are often underestimated, and the problem is getting even more complicated as online video consumption increases, further raising the technical and financial hurdles faced by both content owners and service providers. As a result, some experts have questioned whether the Internet will be able to scale to support the delivery of high-quality video to large audiences in the future. One of Akamai's strategic goals is to enable the Internet to achieve such scale, and our recent success in streaming the 2014 World Cup to a global audience has proved once again that Akamai is capable of overcoming the bottlenecks inherent in the Internet so that millions of viewers can watch high-quality video events over IP. Akamai achieved several milestones during the recent World Cup. We supported our largest audience ever for a single event, delivering nearly 7 terabits per second of traffic to over 5 million concurrent viewers of the semifinal match between Argentina and the Netherlands on July 9. We set a new total Akamai traffic record of 23 terabits per second during the Brazil-Germany…

[Operator Instructions] And our first question will be from the line of Heather Bellini, Goldman Sachs.

This is Nicole [ph] in for Heather. I was just wondering about the sales productivity in the quarter. I know there are more tenured reps now, especially in the performance and security business. Could you just give more color on that?

Yes, so I -- we continue to ramp well with the sales hiring that we've done. We continue to do well on kind of making the reps productive. I think it's important to -- you commented on performance and security. The reps that we hire sell more than performance and security. They sell the entire portfolio. So we have reps that are also selling media solutions as well. So they're not just selling performance and security, but certainly, there's an emphasis on that part of the portfolio. But I'd say, in general, we track the reps by tenure class, and they're kind of tracking to historical levels.

Great. And in the media business, did you see any changes in the competitive landscape there? Any pricing pressure?

I would say that's pretty similar to how it's been for many years. As we've often talked about, there's a lot of competitors. And the very biggest companies also do it themselves, or try to, in some cases. And I anticipate that continuing as we move forward.

Your next question will be from the line of Michael Turits, Raymond James.

It's James Wesman sitting in for Michael. Jim, just to look at the 3Q revenue guidance for a second. I know you guys did post a fairly seasonal strong Q2. The 3Q guide, though, was a point or 2 below what you guys usually do seasonally in the third quarter. Is that a result of the repricing from your largest customer? Any color around there will be helpful.

Yes. I think it's important that actually, it's in line with kind of what we usually see seasonally kind of Q2 to Q3. So it's not kind of below that. But to give you a little bit of color on kind of the guidance range, kind of at the midpoint we're expecting another solid quarter. I think we've mentioned in the past, the summer months from a traffic perspective are lighter than other quarters of the year. So certainly, we are expecting some of that. There are a couple of large software releases that are planned towards the back half of the quarter. And effectively, what we're guiding here is that kind of at the midpoint of the range, those software releases, they happen within the quarter and they're kind of in line with the scale that we expect. On a lower end of the range. Maybe some of those software releases push into Q4. But we're pretty confident with kind of the range that we've outlined. So hopefully that gives you a little bit of perspective on kind of the guide.

And then, just a follow-up for Tom. Tom, just some insight into the security business. I know Verizon and EdgeCast had rolled out Web Application Firewall offering. Are you guys seeing that at all in your bake-offs to customers?

Not really, no. We've seen the press there. Our relationship with Verizon continues as it has been with Verizon Enterprise Services. And we don't see a lot of EdgeCast in the marketplace and certainly don't see them in the security area.

Your next question is from the line of Sterling Auty, JPMorgan. Sterling P. Auty - JP Morgan Chase & Co, Research Division: So the investment that you're making in the second half, I want to make sure that I'm clear. How much of the investment is really in the security and performance side versus some of the buildout to support media delivery? And traditionally, you guys have done a good job of kind of building in anticipation of demand. What are the things that you're thinking about specifically that would fill up that capacity investment that you're making?

Yes. So I think, in general, the investments that we're making are not just weighted to security. We've been making investments across the portfolio, across security, across our kind of cloud networking products, across our web performance products and media. So the investments have been pretty broad from an R&D perspective and also from a go-to-market perspective. So I just want to make sure that -- and those investments we're expecting to continue into the second half of the year. I think to your point about security, we did comment on the last couple calls that we're building out more network expansion, in particular, to support our Prolexic offerings, and we're expecting to do more of that in the second half of the year. And as we've said before, from a CapEx perspective, on the network side, we continue to build out in support of what we believe is the demand that's there. And we know that we can grow into it with traffic. So, effectively, the way to think about it is we build CapEx to support what is an aggressive traffic outlook because we certainly don't want to be turning customers away. And if in fact the traffic is a little bit lighter than we expect in the quarter, you can always grow into it in the following quarter. Sterling P. Auty - JP Morgan Chase & Co, Research Division: Okay. And then if memory serves me right, in the first 2 quarters of the year, you had some kind of more one-off-ish items, I think, like contracts for IP performance or some specific things that you were doing. Was there anything in that kind of one-off nature that helped -- I'm sorry, in the first quarter, I think it was -- maybe the fourth quarter, but was there anything here in the June quarter that fell into that kind of bucket?

Not really. We had very good and balanced performance across media and pretty much across all of our solution categories. I think probably the one area that maybe tends to be a little bit more lumpy is our custom government projects. And we didn't have as many of them in Q2, but that's probably the only area. I think, outside of that, we had very strong growth across the portfolio.

Your next question is from the line of Colby Synesael, Cowen.

Great. This, I think, is the second quarter where you've noted that you've pushed out hiring. And I'm trying to get a sense of what's behind that. Is it simply you're having a hard time finding these people? Is there something else there? And then when you think about the momentum you're seeing in the business and the guidance that you laid out for 3Q, is it favoring an acceleration in any particular business, whether it's -- if you just break out the way you look at your revenues, the media business versus performance and security. Just trying to get a sense of what that's going to look like between the 2 businesses.

Sure. So on the hiring front, I mean, we set some pretty aggressive hiring expectations for ourselves for the year. So it's certainly not a matter of having difficulty hiring people because Akamai is a very attractive company for people to come to. So we're not having an issue whatsoever with attracting talent. We just put together a very aggressive hiring plan. Sometimes, hiring shifts from quarter-to-quarter. We want to make sure, obviously, we're hiring the best people. And so I think that we're quite diligent to make sure that the caliber of people that we're bringing on board are the caliber that we want. Relative to Q3, we're not going to -- I'm not going to guide by business unit. I can tell you that we expect strong growth again across the portfolio. So it isn't weighted to any one area. I'd say you should expect that the growth will be pretty balanced across the portfolio again.

Your next question is from the line of Mark Keller, D. A. Davidson. Mark Kelleher - D.A. Davidson & Co., Research Division: My first one is just a little bit of follow-up to that one. The buildouts shifted as well, was that connected to the hiring?

No. It's not really connected. I mean, again, we end up -- you end up having to -- for many of these network deployments, they take 3 to 6 months as far as planning. And sometimes, you have vendors and sometimes the vendors are maybe supposed to ship you goods at the end of a quarter and they slip by maybe a week or 2. So you really -- it's really only notable things like that. And then some of the facility expansion that I mentioned, sometimes in certain states and in certain countries, the buildouts take longer based on kind of just challenges with getting permits and things of that nature. There's nothing noteworthy, other than just things that are kind of normal that tend to move kind of from quarter-to-quarter and month-to-month. Mark Kelleher - D.A. Davidson & Co., Research Division: Okay. And on the security side, you mentioned Verizon, you're not seeing them there as a competitor. What do you think the growth rate there is in the DDoS security market? Are you maintaining share? Are you taking share? And what's the competitive environment right there?

I would say our solution, now that we're combined with Prolexic, is very compelling. And we're seeing high growth rates and a lot of traction there. There are a lot of competitors. Our approach is different. It's the new way of doing things, in the cloud. And that's very compelling when you have these large attacks where you really can't defend yourself by buying a box and sticking it in your data center and even your carrier really can't effectively defend you against these very large attacks. So we're seeing excellent traction. And over time, I think you're seeing us take share from the traditional way of doing things and add to the overall growth of the market with a new way of doing things. So the business is growing very rapidly.

Your next question is from the line of Jennifer Lowe, Morgan Stanley.

Maybe even following up a little bit on the last set of questions. About -- I think yesterday there was an announcement out in terms of a partnership you all had entered into with Microsoft around investing in security businesses in Israel. And first time I think I've seen you all working with Microsoft, so it was sort of interesting in that regard. And then, obviously, security is a big focus area as well. Can you talk a little bit about that project specifically? What the opportunities you see there are? And how that sort of came about in concert with Microsoft?

Sure. That's an investment in very early stage startups. Of course, Akamai has its roots in the 50k competition at MIT. And Israel is a great place for hiring security talent. It's a great place for startups and security startups. And so that's a really good place for us to partner with Microsoft in investing and supporting early-stage security companies. You hope for a long-term payoff there. We have a large and growing presence in Israel where we base a lot of our security development. So it's a perfect location for us. We have a great partnership with Microsoft, and it just makes a lot of sense all the way around.

Great. And then, just shifting gears quickly, one last one for me. As you -- I think in the past you've talked about the potential for acceleration throughout this year as all of the hiring from last year starts to reap productivity and there's a life factor there, but this seems like that this might be the point where we start to see that take hold a little bit. Is that still the expectation as you think about -- and I know the first question got into this a little bit -- but as you see the sales people start to hit the road and start to bring in business, has that trended in line with maybe where you thought that would be 6 months ago? Better or worse? What's sort of the color on that?

Yes. And I think, in general, there's a lot of things that obviously affect revenue growth, right? The sales productivity is but one aspect. Obviously, traffic volumes, pricing volumes or pricing impacts. There's a lot of things that affect growth, but we are very pleased with the hiring that we made on sales. We are very pleased with kind of the overall productivity that we're seeing across the business. I'm not going to guide beyond the current quarter. You can obviously see from the Q3 guide that we're guiding to a pretty strong guide again, a continuation of what was strong growth in Q1, strong growth in Q2, expected strong growth in Q3. And certainly, the sales reps that we are hiring are selling into the installed base more of our solutions, as well as selling off the installed base. So, in general, I think -- we track it by cohort class. I can't tell you that every single rep is tracking to our expectations because that would -- that's not true, but you wouldn't have expected that anyway. And I think, in general, the reps by cohort class are tracking pretty well, and I think the expectations that we set earlier, we're still comfortable with that and the investments that we're making and the return that we're going to get.

Your next question is from the line of Sameet Sinha, B. Riley.

Couple of questions. Can you tell us what the contribution was from Prolexic during the quarter? And secondly, you mentioned number of security customers at 1,500. This is up from about 1,250 that you had at the end of Q1. Do you think that is the kind of pace that we should be expecting throughout the year? Or is there anything specific that helped you increase the number of customers by that much this quarter?

Sure. I think we told you in the -- after the last quarter's earnings, we're not going to break Prolexic out separately. But what I can tell you is what I shared last time is Prolexic's run rate was about $5 million a month. So call it $15 million a quarter, so you have a frame of reference of roughly the size of Prolexic. And relative to the customer ads on the security front, you're right, we were very pleased. This is kind of the first full quarter of having Prolexic and our broader security portfolio. We have not added that many of kind of net new customers in the security portfolio really in -- I don't think ever. So we're quite pleased with that traction, and we expect that the demand for our security portfolio is very strong.

Your next question is from the line of Richard Fetyko, ABR Investment Strategy.

Just backing into the organic growth rate in the performance and security solutions segment, excluding Prolexic in the first quarter and the second quarter's estimate. It looks like the organic PSS growth slightly decelerated from perhaps 23% in the first to 20% year-over-year in the third. Just curious if we should see any acceleration in the organic business or if there were any specific reasons why that deceleration happened.

Yes. I mean, I think we had an extraordinarily strong performance in security growth rate in Q1. If you take kind of the 30% growth rate, even if you do the math that you just suggested, you're talking about a 20% rough growth rate for Q2. I mentioned that the custom government business was lighter, so projects in Q2 were lighter. That is in the performance and security portfolio. That's one element of the portfolio that tends to be a little bit lumpier based on projects that you have with the federal government. But we were quite pleased with the growth rates that we saw in performance and security. And there are going to be things, quarter-to-quarter, like custom projects. I recall, last Q2 we had a very strong quarter last Q2. We talked about -- we had a very strong custom government quarter last Q2. And we also talked about the fact that we had a very, very large upgrade of our IP accelerator solution with one particularly large customer. So kind of having the wraparound effect of that. But organically, across kind of the performance and security portfolio, we're quite pleased with the growth rates.

Your next question is from the line of Ed Maguire, CLSA.

I was wondering if you could comment on the cross-selling success so far with performance and security and the media business. I know you've mentioned the number of common customers between Prolexic and the Kona Security, but was curious as to how that -- the focus on security is progressing more broadly across the customer base.

Yes, I mean, I would say -- one proof point is kind of the question I was asked earlier, the kind of the -- call it the net security kind of customers add of over 250, I think tells you that those were sold to the existing installed base. Those were not off the -- most of those were not off the installed base. Most of those were kind of embedded into the installed base. So I think we're doing early signs of traction with Prolexic, we're getting good penetration. I think we're getting decent penetration across the board. I can tell you, though, a lot of room for growth beyond that, that we still have very low penetration rates of kind of security with kind of our broader customer base. So we think this is a huge opportunity to kind of sell more security and sell more of our broader offerings to our customers, as well as off the installed base. So I think there's a lot of opportunity for incremental growth.

Great. And just a pulse check on the wireless, the Aqua Ion. How are -- how is the penetration proceeding among your existing customers?

I would say the penetration is small and growing. We see significant uplift in pricing, and there's significant interest, of course, in the product with the situational awareness, the optimization for cellular communication and making the website be really fast for a variety of situations with the end user, a variety of devices and a variety of locations. So Ion is a strong product, financially, with plenty of room for growth. So the overall penetration is still low, as with security, but achieving good success in terms of the upsell and, of course, in terms of the performance capabilities.

Your next question is from the line of James Breen, William Blair. James D. Breen - William Blair & Company L.L.C., Research Division: Just one more question on the hiring and then one on the technology side. On the hiring side, you hired, I think, about 100 salespeople last year and you'd hinted that you would hire somewhere around that number this year. Is that part of what you're pushing out or has that been pretty consistent throughout the year? And then, secondly, just on the technology side, as you look at the traffic from the World Cup relative to maybe the Sochi Olympics and the London Olympics. Are you seeing any trends there in terms of moving more towards the wireless side relative to wireline and regionally?

So I'll take the hiring kind of question. So I'd say again, we're not going to provide guidance of how many sales hires. We did that last year, I think, because we were trying to make a point that the company was hiring more sales reps last year, I think, than we had in the last 5 years combined. We're going to hire a similar -- roughly similar amount this year, whether it's 100, whether it's 80 remains to be seen. And it's not so much pushing out hiring. I just want to make sure that I'm clear that we're not pushing out hiring from Q2 to Q3. This was hiring that we had aggressive hiring plans for the first half of the year. Some of that hiring has moved into Q3, but it's not like this conscious push to Q3. And from a sales perspective, that I'd say our sales hiring is tracking pretty well.

Yes, to the second question. We're seeing across the board a very rapid growth in the penetration of mobile devices. On aggregate, you can think of roughly half of transactions being done with a mobile device now. Now traffic measured by bits is lower, a lower percentage, because when you're doing video and so forth, you'll tend to get lower bit rates to those devices. And most of the wireless device traffic and transactions are on Wi-Fi. So it's not cellular coming out of a tower. Cellular is still a relatively small percentage overall, but also growing.

Your next question is from the line of Gray Powell, Wells Fargo Security.

Great. I just had a couple, if I may. So first, I just want to make sure that I'm thinking of Q3 guidance correctly. Specifically, the midpoint of guidance implies $13 million of revenue in Q3, whereas the past 2 years, the guide has been more like 6 or 7 -- I'm sorry, $6 million or $8 million in sequential revenue growth. So I guess, my question is where is the strength coming from? Is it more of the CDN side or performance and security side?

Again, I'm not going to provide guidance by business, but I can tell you that our expectation is you're going to see strong growth across the entire portfolio. And I think you correctly outlined it that we're pretty bullish, that's why the guidance is pretty strong here at the midpoint. We're expecting a continuation of kind of strong performance in media, strong performance in performance and security and strong performance in service and support.

Got it. Okay. And then just one more, if I may. Can you talk about your partnership with Open DNS on the security side? And how do you feel about the security portfolio today? Do you see an opportunity to add new technologies beyond DDoS and Web Application Firewall? And specifically, can we see Akamai doing something more like cloud-based URL filtering or email security or maybe something else down the road?

Yes. We have a rich roadmap ahead in security. The next area for us that you'll start to see products is on client reputation database, where we're looking at a particular client and trying to decide are they a legitimate user or is it an attacker of some kind or some kind of a bot, which might be a friendly bot or an unfriendly bot. And providing that information to our customers as a request comes in so that they can decide do they want to accept the request, alert the request or block the request. Even if the request itself doesn't look malicious, maybe that's because it's a new form of attack that hasn't been seen before. Or maybe it's some kind of a bot that they decide that they don't want to admit to the site. So that's the next product that we're working on and that we hope to have a beta version available later this year. And there's a rich roadmap from there that yes we might well be doing things like URL filtering and preventing man-in-the-middle attacks. I don't know about email filtering, that's not coming up anytime soon, that we get into the email business per se. But I think there'll be a very rich roadmap for security, especially as we do more and more with enterprises and their traffic.

Your next question is from the line of Jeff Van Rhee, Craig-Hallum.

Several quick ones for you. First, just in terms of renewals. While you have big renewals in any given quarter, any notable renewal patterns we should be thinking about the next few quarters? And then second question, as it relates to the landscape around security, could you just touch on the competitive landscape: when you are competing, just clarify for us who is the toe-to-toe typical if you're in there competing against somebody.

So on the renewal front, I'll let take that, I'll let Tom comment on the security landscape. But I think you kind of answered the question yourself, which is that we have large customers that renew every quarter, and we're going to continue to see large customers renew every quarter. There's nothing more notable in that. Obviously, I shared a few quarters ago one specific customer because they were our largest customer, and they were a customer that was pretty darn close to a 10% revenue customer so -- but I think, outside of that customer, it's just standard renewal that happen every quarter and a handful of large customers.

In terms of the competitive landscape for the security products, I would say the dominant players out there today where most of the revenue goes are the folks that build devices that you would deploy in your data center. Now the big problem with that as we move forward is that the attacks are so large in many cases that the connectivity at the data center becomes overwhelmed. So even if the device is doing a good job, it never gets a chance really, and then the site is DoS-ed or brought down. Carriers, many of them provide security solutions. Many of them also we sell Akamai. The carriers will try to provide some kind of a clean pipe. And again, you have the capacity challenges. The carrier solutions often are not able to inspect the packets that really limits what they can do. They take a while to be turned on, which can be a problem, they're sort of reactive and slow to react. And they can hurt the performance of the site. So even when they can defend the site, the site is sort of compromised in terms of its ability to do its job performance-wise. I think you're starting to see -- there's a lot of start-ups out there claiming to have various kind of security products. You see the CDNs, some of them trying to have a security offering. I think for the state-of-the-art attacks out there, really, our solution, combined now with Prolexic, provides really something that's special, and that you just can't get other places that really works. And Akamai is a performance company and so we've made sure that when we're defending the site and filtering the traffic, there's not a big performance degradation. The site still is going to work really fast, and it could be on in a proactive manner. We have enormous scale compared to any of the competitors because our servers are inside over 1,200 networks. As we talked about earlier, we'll deliver a World Cup game at 7 terabits a second, and the big attacks out there that overwhelm a lot of the competing services are a tiny fraction of that. So we have tremendous scale to bring to bear to defend websites, and I think that's why you see the business growing so fast with a lot of the marquee brands now turning to Akamai for the next generation of security defenses.

Your next question will be from the line of Ben Rose, Battle Road.

Yes. Question for Tom. Looking out beyond the third quarter, can you provide an update on your ongoing R&D relationship with Cisco and your expectation for the hybrid cloud optimization market?

Yes, sure. Now, tomorrow, Cisco is making their new router that has what's called Akamai Connect or Akamai software on board generally available. So that hits the marketplace tomorrow. Obviously, we're very interested to see that product become successful in the market. The beta test results are excellent. The product provides significant offload for the WAN and the branch office connectivity. So it allows that infrastructure to do a lot more and it provides significant speedup for branch office applications, Internet access, the basic kinds of things you want to be doing in the branch office. And so, we're very optimistic. That's a product that will be sold by Cisco in the Cisco channel. We're obviously supporting that as best we can, but that's a Cisco-led effort with Akamai software on board. There is a rich roadmap beyond that first product to do more, and we have a very good relationship with Cisco in R&D and are looking forward to the release tomorrow.

Okay, great. And then a question for Jim. I know the company has obviously done a really outstanding job over the last several quarters of wringing out some of the inefficiencies in the network and improving gross margins. Do you think that there's further room, going forward, to achieve the kinds of efficiencies that you've achieved in the last year?

I mean -- I think we do it every quarter, we do it every year. You're right. Notably, over the last 3 years, it's been a pretty impressive improvement in gross margins for the company. And I have to kind of really give the credit to our network and media team, they've just done a fantastic job, and you can expect that we're going to continue to have ongoing platform efficiency initiatives that can kind of maintain the level of gross margins that we have. I'm not calling, obviously, in the guidance that it's going to improve, but I think we can maintain the margins at the levels that we're currently at.

Your next question is from the line of Philip Winslow, Crédit Suisse. Sitikantha Panigrahi - Crédit Suisse AG, Research Division: This is Siti Panigrahi for Phil Winslow. So one more question on the gross margin side. Could you give us some color on like what you are seeing in terms of co-location and bandwidth costs? And maybe what you have seen through first half and what do you expect in the second half?

Yes. Again, we continue to make progress on driving kind of the bandwidth per kind of megabit per second down. So our bandwidth costs continue to trend down nicely, not on an actual dollar basis, but kind of on a per-bit basis. And the same is true for co-location costs. So we have a whole playbook that is focused on areas that are going to be able to reduce bandwidth costs. Areas that can help us reduce co-location costs, areas that can help us build kind of reduced costs and optimized costs associated with kind of network broader buildout. So you can expect that we're going to continue to kind of initiate all those initiatives, not just this quarter, but the second half of the year and beyond.

Your next question is from the line of Kevin Smithen, Macquarie.

First of all, nice job with the guidance, especially with the customer repricing and all the investments. Wanted to ask you, as we see more high-def and 4K sporting events and online gaming more pervasive -- online content more pervasive, do you think that we're going to see a new normal in terms of growth rates for the media business over the next couple of years? And how do you sort of think about the trade-off with some of the large media customers between sort of these -- this extra traffic and margins? And do you continue to expect to give volume discounts to those big customers or will you try and diversify the customer base more than, say, 2 to 3 years ago?

Obviously, as -- if and when 4K becomes more widespread, that provides dramatic increase in traffic for every hour of video that's watched online. 4K has obviously not penetrated the market yet. We do support it on our platform, but it's not widely adopted. Traffic is growing at a rapid rate. There is the potential over a period of several to many years for several orders of magnitude of growth in the online traffic. Of course, pricing is going to need to come down, and we've talked in the past about the many things we're doing to make video be more affordable online. But on balance, when you multiply the traffic growth rates with the price reductions, we believe that revenues have a lot of room for growth going forward. And we're very bullish about our media business, as I've talked about before. Of course, as customers get larger and push more traffic, they do get traffic discounts, but that's already factored into our estimates as we look forward.

Your next question is from the line of David Dixon, FBR Capital Markets. David Michael Dixon - FBR Capital Markets & Co., Research Division: Yes. So I wondered if, first of all, you could just give us a general sense of the momentum that you're seeing, or to the extent you're seeing momentum in the mobile space, which is a key area of growth for you? And then, secondly, as you were talking about your advantages in the marketplace and you centered in on the scalability of the platform. One of the other things that I think seems to be an advantage is the customer responsiveness relative to the telcos. And in talking to some the major enterprises that we speak who checks the -- they don't really care if it's in the cloud or in their own enterprise, as long as that stack can be evaluated for attacks and analysis. So I wanted to actually ask you a question around the network compute capability and whether you see the need to overhaul or significantly augment that compute capability as you see Google stepping forth and investing very aggressively, very heavily in, perhaps, hundreds of data centers here with that compute capability.

Sure. First on the mobile question, as I mentioned before, we're seeing a rapid penetration of mobile devices that if you measure in terms of applications that we would transact, roughly half being on a mobile device. Now the large majority of that is on Wi-Fi so, ultimately, it's going across a landline connection. Still, a small fraction of mobile is over cellular. And also, if you measure not by transactions but by gigabytes delivered, mobile will have a lower penetration because you don't tend to get the really high-def video that takes a lot of traffic going to a mobile device that's -- usually you're going to send that to a much larger screen that's a fixed device. Yes, we do have tremendous advantages and scalability, but also in sophistication and capability of our defenses. Now in terms of the question on network compute, we do offer compute kinds of capabilities in the specific use case of your website and something that needs to be done for the website. So for example, say, you're an automotive site and you want to do a car configurator so that the user can design their car online, well that kind of thing can be supported by Akamai's platform. I don't think you're going to see us go into direct competition with EC2 or with Google in terms of building out gigantic data centers just for compute capabilities -- for generic compute capability. We are focused around making a website and a web app be fast, reliable and secure. And so we will offer capabilities that support that in terms of storage or compute and delivery, not generic compute for the sake of compute. David Michael Dixon - FBR Capital Markets & Co., Research Division: You're not seeing any competition from Google or signs of competition in the area that you're focused at this stage?

No. I wouldn't say that we see competition from Google. Google does a lot of delivering of content, of course, and they do a lot of that themselves, but that's not really directly competitive to Akamai. We worry about all of the large guys out there in terms of what their capabilities are, but we don't really see them in the marketplace competing for our business.

Your next question will be from the line of Chad Bartley, Pacific Crest.

Wanted to go back to the competition topic. As you look out medium to longer term, how much of a threat are networks either launching competitive CDN services or maybe large media companies connecting directly to them? And in terms of the outlook, I mean, based on your previous second half EBITDA margin guidance and your updated Q3 outlook, should we be thinking about Q4 EBITDA margin also in the 41% to 42% range?

All right. Let me take the first question on competition. And again, the competition really depends on the product area. It's different for video delivery or media delivery than it is from making a web application be super fast for mobile, and that's different than the competition for security. Now in terms of your specific questions about networks as competitors, I'd say there's been a lot of improvement there over the last few years. For example, AT&T used to be a large competitor for Akamai. And now, they're one of our largest partners. Orange in France, Telefonica in Spain, KT in Korea, Swisscom, Turk Telecom, a lot of the world's leading carriers have switched from a model of competing with us, either by buying CDN capabilities from a vendor or developing it themselves, to using Akamai's technology. So they're part of the Akamai family and they integrate very well with us and resell our products. So I'd say that, yes, there's some carriers out there that compete with us, for sure, but that I would say the trend is more that they are moving to work cooperatively with Akamai. And that's because I think we offer the most compelling solutions when it comes to making a website or an app be fast, reliable and secure. And we can do that at a better cost and a better performance than they've been able to do on their own. It is certainly true that any large media company, to your other -- your point, will be looking at do-it-yourself. I think we've said probably for 15 years now, that do-it-yourself is our largest competitor, and you can certainly find examples of that. That said, a lot of the folks that do that still use us for a lot of their business, and they do that again because we do really a very good job of accelerating their content and defending their websites at a very good price point. And there's some pretty high-profile examples of major media companies that did try to build it themselves, spent years doing it and ultimately gave up and returned to Akamai. So I think do-it-yourself is always going to be a factor in the marketplace for us with the very largest media companies. I don't think direct peering makes sense or you'll actually see that for any but a very small number, maybe a handful of the big media guys. The complexity of trying to manage direct peering with 12 dozens or hundreds or thousands of the ISPs out there is really complicated, and the expense in doing that is quite significant. And so it just doesn't make sense to even be tried for any but a handful of the big media guys. So that really doesn't worry me very much.

And on the EBITDA question. I'm not going to provide guidance obviously beyond Q3, but I think I said kind of in my opening comments that we expect to operate in the near term in the low-40s EBITDA. So I think you kind of have what we're expecting to do in Q3 and I think you can expect that we're going to continue to operate kind of in that range, kind of in the near term.

At this time, there are no further questions in queue. I would like to turn the call back over to Mr. Tom Barth for any closing remarks.

Okay. We'd just like to thank you again for joining us today on the earnings call. We would look forward to speaking to you further about our performance and our future opportunities. So please feel free to call us, and have a wonderful summer. Thank you.

Thank you.

Thank you.

Gentlemen, that concludes today's conference. We thank you for participation. You may now disconnect. Have a great day.