瞻博網絡 (JNPR) 2013 Q2 法說會逐字稿

Greetings and welcome to the Juniper Networks second-quarter 2013 financial results conference call. (Operator Instructions). As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Ms. Kathleen Nemeth, Vice President of Investor Relations. Thank you, Ms. Nemeth. You may begin.

Thank you, Operator. Good afternoon and thank you, everyone, for joining us today.

Here on the call are Kevin Johnson, Chief Executive Officer; Robyn Denholm, Chief Financial Officer; and Bob Muglia, Executive Vice President, Software Solutions division. In addition, Rami Rahim, Executive Vice President, Platform Systems division, is on the call and will be available for Q&A.

Please remember when listening to today's call that statements concerning Juniper's business outlooks, economic and market outlook, strategy, future financial operating results, and overall future prospects are forward-looking statements that involve a number of risks and certainties. Actual results could differ materially from those anticipated in these forward-looking statements as a result of certain factors, including economic conditions generally or within the networking industry; changes in overall technology spending and spending by communication service providers and major customers; the network capacity requirements of service providers; the timing of orders and shipments; manufacturing and supply chain constraints; variation of the mix of products sold; customer perception and acceptance of our products; rapid technological and market change; litigation and other factors listed in our most recent 10-Q filed with the SEC.

All statements made during this call are made only as of today. Juniper undertakes no obligation to update the information in this conference call in the event facts or circumstances change after the date of this call.

In discussing the financial results, Robyn will first present results on a GAAP basis, and for purposes of today's discussion, we will also review non-GAAP results. For important commentary on why the management team considers non-GAAP information a useful view of the Company's financial results, please consult the press release furnished with our 8-K filed with the SEC today. For the detailed reconciliation between GAAP and non-GAAP results, please see the Investor Relations section of our website.

On today's call, Robyn will also be providing forward-looking guidance. As a reminder, guidance is provided on a non-GAAP basis. Other than that, with respect to revenue and share count, all guidance is forward-looking and actual results may vary for the reasons I noted earlier. GAAP guidance measures are not available on a forward-looking basis, due to the high variability and low visibility with respect to certain charges, which are excluded from non-GAAP guidance estimates.

Please note that today's call is scheduled to last for one hour and please limit your questions to one per firm. With that, I will turn the call over to Kevin.

Thanks, Kathleen, and welcome to all of you joining us on today's conference call.

Before we review our earnings, I'd like to take a minute to comment on my decision, which was announced today, to retire as CEO of Juniper once a successor is named and an orderly transition is accomplished. After 32 years in a line operating role, with the past five years here at Juniper, my family and I have decided it was time to take a break. The Company is healthy and growing. We have a strong management team and world-class talent in the domain of networking.

The Board has formed a search committee and we are working closely together to ensure we manage the process in a thoughtful way. I am committed to working closely with the Board and my successor to ensure a smooth transition for all of our stakeholders. Until then, I continue to serve in my current capacity.

I will also remain focused on fulfilling my Board commitments through this process. It is an absolute privilege to serve as Juniper's CEO and I have enjoyed the past five years. Together, we have grown our business and strengthened our position in the networking industry, as evidenced by this quarter's results. The business is healthy and momentum is building.

So let's spend some time to talk about the quarterly results. We delivered strong second-quarter results, reflecting good execution on our strategy, as well as the strength in our product portfolio. I am pleased with the performance of this quarter in what remains a dynamic global economic environment.

To put these results into context, you may recall that for the past three quarters, we have been talking about demand improving in service provider. We saw good strength in service provider this quarter, both in terms of revenue and certainly in terms of the bookings environment. I will speak more about the service provider segments -- carrier, cable and content -- in a moment.

In enterprise, while there continues to be areas of weakness, we are seeing some early signs of stabilization and improving demand in the US and Europe. In service provider, we focus on three customer segments -- carrier, cable and content, also referred to as Web 2.0 providers.

Carrier includes both wireline and wireless. In the second quarter, we experienced good overall demand in wireline, while demand also was strong in wireless in the Americas. In EMEA, we are seeing emerging signs of growth in wireless. We performed well in cable and we are seeing great traction with major cable providers.

In content, our good traction in core routing and switching are translating into improved performance as CapEx strengthens in this segment. For the routing market, we continue to see very strong demand in the Edge. Core networks seem to be running hotter, which is a positive indicator, as we expect investment in core routing to begin to pick up in the coming quarters. Overall, the trend we identified three quarters ago of a routing CapEx cycle appears to be picking up.

On the enterprise side, our switching and data center business performed well in the second quarter. Highlights included strong demand from our newly introduced EX9200, as well as our QFabric family of products, which had a record quarter. Customer response to the EX9200 has been good as it satisfies our customer's need for programmability and core switches that power both campus and data centers. This also aligns with our SDN strategy.

In security, we are seeing positive activity in bookings. Bob will provide more detail on our strategy and progress in a few minutes.

As a Company, we continue to drive innovation and differentiation through our systems and our software. We are focused on delivering world-class scale, performance, and reliability in our systems, which is the heritage of our Company. At the same time, we are complementing the value delivered through our systems with a software strategy that enables our customers to be more agile. This is delivered with our network function virtualization capability in a way that helps customers reduce cycle times and be more efficient. And we expect to enhance this capability with future product introductions that are part of our SDN strategy.

We believe the combination of scale and performance in our systems, plus the agility and efficiency in our software, provides unique value that is resonating with our customers.

The second quarter was a strong one for revenue, bookings, and backlog. In addition, we remained focused on prudent cost management and effective capital allocation, which Robyn will discuss later on the call.

To sum up, our 2013 results show that we are successfully executing on the five operating principles we established at the beginning of the year -- expect macroeconomic environment to remain uncertain; expect modest growth in the markets we serve; take share in routing and switching, and stabilize share in enterprise security; expand 2013 operating margins over 2012; and continue to generate solid cash flows and prudently allocate capital.

The strength that we are seeing in the service provider market and the stabilization we are seeing in the enterprise market reinforce our confidence that we are in the right markets with a great product portfolio. We continue to feel very good about the health of the business, and we will continue to stay focused on executing with agility and delivering value for all our stakeholders.

Now, I will turn it over to Bob to discuss the latest developments in security and SDN. Bob?

Thanks, Kevin. Today, I'd like to update you on our progress in the security business, as well as with SDN and our overall software strategy.

As we've previously outlined for security, we expect our recovery to occur over multiple quarters. With that background, Q2 was a tough quarter for our security business with revenue of $126 million, a decline of 20% year over year and 8% quarter over quarter.

However, the actions we are taking are showing some positive signs and results. Our security bookings grew 6% over Q1 2013 and we exited Q2 with a strong backlog in our security business. We are also seeing encouraging growth in our security sales pipeline.

Providing color on this, as previously outlined, you can divide the network security business into three distinct customer value propositions -- securing service provider customers, securing corporate end users within the campus and branch, and securing the data center from Internet hackers.

The primary usage of network security for service provider customers is enabling high-scale firewalls for mobile users. Juniper has benefited from the growth of LTE networks as the scalability and reliability of our SRX product is ideally suited to this application.

During 2013, we have seen this business slow in North America as the large service providers focus their capital spend on other areas, including Edge routing. Moving forward, we expect to see growth opportunities in EMEA, South America, and, to a lesser extent, in Asia as these carriers roll out LTE. We have a leadership position in service provider security and we intend to maintain that as more carriers around the world modernize their mobile networks with high-capacity LTE service.

For campus and branch, we continue to see the impact of the product transition from our older ScreenOS product line to the current SRX products. Our ScreenOS products continue to decline, but we have seen good year-over-year growth in our branch SRX products. That said, this is the area where our products fell behind in the transition from ScreenOS to the SRX, and we expect recovery in campus and branch security will occur over multiple quarters.

While we still have work to do, this transition is well underway. In Q2, ScreenOS products accounted for approximately 20% of overall security revenue.

For the data center in Q2, we shipped the highly innovative Spotlight service that works in conjunction with Junos WebApp Secure. Spotlight is the only cloud platform in the world that shares fingerprints of individual attackers' devices. This intelligence service protects our customers by blocking attackers before they have a chance to do damage.

The combination of WebApp Secure and Spotlight provide a foundation for differentiation of Juniper's overall data center security products. This strategy is working for us, generating considerable interest from major enterprise customers, including the Chicago Mercantile Exchange and Interactive Data, who have chosen Juniper to secure their data center environments.

In Q2, we also shipped Junos DDoS Secure, a unique application-centric service that provides protection against distributed denial of service attacks. DDoS Secure was chosen by First American Bank to secure and protect their financial systems, and this win further strengthens Juniper's relationship with this important customer.

We are seeing strong interest in using our security products to protect the enterprise data center, and to further capitalize on this, in Q2 we took steps to restructure and grow our enterprise security sales force. We expect those steps, together with the introduction of new and improved products, to bring stability to our security business in the second half of 2013 and enable a return to growth in 2014.

Switching to SDN and our broader software business, we have established a strong industry thought leadership position with our SDN strategy and products. That strategy has resonated with customers, industry analysts, and the press, as demonstrated by the positive feedback and [news] we've received, as well as our overall share of voice within the press, which far outstrips what might be expected from a company of Juniper's size.

In May, we announced that we had entered beta with our Contrail SDN controller and we pulled in the ship date of this product from 2014 to second-half 2013. Pulling in the ship date of a new product is an unusual event in this industry, and the progress we've since made with our beta customers further increases our confidence that we will meet our commitment.

In terms of those beta customers, they are relatively evenly divided between enterprise and service providers and include a strong representation of customers in all three global geographies. We see several value propositions emerging for SDN across both the enterprise and service providers.

Within the data center, as customers implement private and public cloud infrastructures, they must virtualize their network to enable elastic scaling of resources. Enterprises want to build hybrid clouds that connect their private cloud to commercially provided public clouds for disaster recovery and on-demand availability of resources. Service providers seek to improve service creation agility within their Edge network by implementing network function virtualization.

All of these capabilities are enabled by our Contrail controller, and with the beta trials that are underway, we've established strong mind share and experience in this important emerging area. Interestingly, many of these trials bring in other Juniper services, such as Firefly, our virtual firewall.

While we don't expect to see material SDN revenue in 2013, we are well positioned to benefit from the growth in this area in 2014 and beyond.

For our broader software business, we achieved record Q2 bookings, which is reflected in backlog, and this will be recognized over multiple quarters. We expect to see the trend of ratable revenue recognition continue as we move our software licensing to Juniper's Software Advantage, the networking industry's only comprehensive software licensing model.

Earlier this month, we moved the majority of our MX application software to Juniper's Software Advantage and we anticipate transitioning more of our security products in Q4 of this year.

All of these steps -- rebuilding our security business, establishing leadership in SDN strategy and products, and moving to our new Juniper Software Advantage licensing model -- are meant to build a foundation for the future. It will take time, but we are on our way.

Now let me pass this on to Robyn.

Thank you, Bob, and good afternoon, everyone.

I am pleased to report that our second-quarter results reflect strong revenue and even stronger earnings growth. Revenue exceeded our expectations due to two main factors that were roughly of equal size, the first being increased service provider spending momentum and improved enterprise demand, and the second factor was the earlier-than-anticipated recognition of some US federal revenue.

Operating income and EPS both improved significantly, reflecting the leverage in our financial model.

Looking at our demand metrics, we had a good bookings quarter with strength across our customer base, resulting in book to bill greater than 1. Product backlog, which has already been at healthy levels, increased both sequentially and year over year. Product deferred revenue was down modestly quarter over quarter.

Total revenue was $1.151 billion, up 9% sequentially and 7% year over year. The sequential growth was driven largely by enterprise, whilst the year-over-year growth was balanced between service provider and enterprise. No single customer accounted for more than 10% of total revenue in the quarter.

For the second quarter, GAAP diluted earnings per share were $0.19. Non-GAAP diluted earnings per share were a strong $0.29, up $0.05 sequentially and $0.10 year over year.

Now let me provide some color on revenue by region, market, and business segment. Americas revenue was up 14% sequentially and 15% year over year. Enterprise had a very good quarter, driven by federal and financial services that resulted in double-digit sequential and year-over-year growth. Federal had a good quarter, even without the benefit of the deferred revenue recognition.

The year-over-year strength in the Americas was due to another quarter of double-digit service provider growth. This growth reflects the improving broad-based demand across cable, content and carriers.

EMEA revenue was up 4% sequentially and 1% year over year, driven by enterprise strength, whilst service provider revenue was flat sequentially. We did see improved demand in northern Europe and in Germany.

APAC revenue was down 1% sequentially and 7% year over year. We saw soft demand in enterprise across the region and Japan continues to be weak. We did see sequential growth in service provider, primarily in China, Taiwan, and South Korea.

Looking more closely at the markets we address, service provider revenue was $726 million, up 2% sequentially and 7% year over year. US SP was strong, especially in cable, content and pockets of strength in EMEA and APAC. Enterprise revenue was $425 million, up 23% sequentially and 8% year over year. Sequential growth was driven by federal and financial services in the US and broad-based growth in EMEA.

Now let me review our revenue by segment. Platform Systems division revenue was $916 million, up 13% both sequentially and year over year. Total router product revenue of $578 million was up 12% sequentially and 14% year over year. The MX product line had another record revenue quarter, growing 19% sequentially and 43% year over year. This reflects continued strong customer acceptance of our Universal Edge offering.

In addition, we recognized our first revenue on the recently released MX2020. Total switching product revenue was a record $160 million, up 22% sequentially and 16% year over year.

In the data center, the newly released EX9200 is off to a good start and QFabric continues to generate momentum. SSD revenue of $235 million was down 6% sequentially and 11% year over year. Total security product revenue was $126 million, down 8% sequentially and 20% year over year.

This area of our business remains challenged. However, as Bob mentioned, we are seeing good customer activity and response to our new initiatives with some early signs that the business is turning around. In the second quarter, both orders and revenue from new products increased sequentially, and we feel good about our ability to achieve our year-end quarterly run rate of $150 million.

Moving on to gross margin, and operating expenses, non-GAAP gross margins for the quarter were 63.7%, compared to 64.6% last quarter and 63.4% a year ago. Non-GAAP product growth margins were 63.8% and non-GAAP services growth margins were 63.3%. Growth margins for the quarter would have been roughly flat sequentially except for the negative impact of the federal deferred revenue recognition.

We are pleased with the progress that we have made in our supply chain and our services cost reductions. Non-GAAP operating expenses decreased $3 million sequentially and $5 million year over year to $515 million. OpEx was at the high end of our guidance, reflecting good progress in cost reduction, partially offset by higher variable compensation and legal expenses. OpEx as a percentage of revenue was 44.8%, versus 48.9% last quarter and 48.4% in Q2 of last year.

Looking at our headcount, we ended the quarter with 9,513 employees, a sequential increase of 191, primarily in lower-cost regions.

Non-GAAP operating margin for the quarter was 18.9%, compared to 15.7% last quarter and 15% last year. Operating income increased 31% sequentially and 35% year over year. This reflects our strong focus on achieving our long-term model by driving operating leverage with good topline growth and OpEx management.

The non-GAAP tax rate was 27.4%, compared to 19.8% last quarter and 30.7% a year ago. As a reminder, last quarter the tax rate reflected a one-time benefit of the 2012 R&D tax credit.

Looking at the balance sheet, we ended the quarter with $2.8 billion of net cash and investments. For the quarter, we generated strong operating cash flows of $284 million, in line with our historical pattern. DSO was 40 days in the quarter, down from 45 days last quarter, due primarily to better order and shipment linearity.

Capital expenditures were $71 million, down 1% sequentially. Our depreciation and amortization was $42 million. We have now largely completed the construction of our Sunnyvale campus and also consolidated some of our labs and office space in our existing leased facility.

We repurchased 6.4 million shares for $106 million. We have $333 million remaining on our current authorization. As announced, the Board has approved a new incremental authorization of $1 billion and we expect to continue calibrating our buybacks in future quarters with market conditions.

Now I will review our outlook. As a reminder, these metrics are provided on a non-GAAP basis, except for revenue and share counts. Our outlook for the September quarter reflects our expectations of continued good service provider demand in the US and modest improvement in EMEA. We expect routing and switching to remain strong, while the security business begins to stabilize.

For the third quarter, we expect revenues to range from $1.140 billion to $1.180 billion. Gross margins are expected to be in the range of 64.5%, plus or minus 0.5%.

Operating expenses are expected to be $525 million, plus or minus $5 million. As we have demonstrated, we have focused on driving leverage in our financial model and continue to bring our cost structure in line with our long-term operating model. As an update to our cost-reduction activities, we have experienced high legal fees all year, and given the improved demand environment, variable expenses will be higher than we anticipated for the year. We also are making some incremental investments to maximize our topline growth.

With that in mind, we now see a net year-over-year OpEx reduction in 2013 of approximately $50 million. We are well ahead of our $50 million cost-reduction goal.

At the midpoint of guidance, operating margins are expected to be roughly 19.5%. This is expected to result in non-GAAP diluted EPS of between $0.29 and $0.32 per share, assuming a flat share count and a tax rate of 28%.

To summarize, we expect the demand environment to continue to improve. We are confident of our product portfolio in routing and switching and improvements in security. We are focused on executing to drive revenue growth, maintain healthy gross margins, and manage OpEx to achieve our long-term model. I would like to thank our employees for their continued dedication and commitment to successfully executing our strategies.

Now, let's open the floor for questions.

(Operator Instructions). Ehud Gelblum, Morgan Stanley.

Hey, guys. Thanks. Appreciate it. Wasn't expecting to come in that quickly. But I've got a bunch of questions for you, so I hope you are ready. Question one is on the switching business. That was the strongest I think we have ever seen it, $150 million. Can you give a breakout on -- you said QFabric was strong. How strong was QFabric? Is it still on the order, you know, small numbers or a substantial -- kind of just give us a sense as to how big it is?

And when you talk about QFabric, is it the top-of-rack switches, or are people buying the directors? And clearly, I assume it is more in the minis than the original QFabric. And if you can give us a sense, also, as to how much of your strength in switching was from data center versus campus LAN, and if you can tell us about Wi-Fi, just breaking out switching?

And then on security, when you look at your two businesses in security, the carrier side and the enterprise side, Bob mentioned that the carrier SRX was not doing as well this quarter because carriers are focusing more on routing. Does that mean that the two are kind of out of sync with each other and you can never get the two on the same page?

And then, finally, enterprise. I'll let someone else take care of that.

Ehud, what was your last question? Was it on enterprise security?

It would have just been how -- is it possible to separate that from the carrier security side, and sort of maximize that for cash as opposed -- run it for cash as opposed to for growth the way you are doing in carrier?

Okay. Got it. Thanks for your questions. Let me sort of kick off on the first one on the switching business. I think we had a very strong quarter on switching, and I think a part of that was the introduction of the EX9200, which replaced our EX8200 switch.

And the difference in the EX9200 is the programmability that that switch enables, and it's a great switch both for data centers, and we see some customers even using it for campus because of the programmability. And so, I think the product transition from the 8200 to the 9200 was one contributor to the strong performance in switching.

And as we noted, QFabric had a record quarter, and we talk about QFabric as the QFabric family of products, which does include the top-of-rack switches. It includes the smaller version of the Interconnect, as well as the larger version of the Interconnect. And this was a quarter that we had -- we certainly had continued sales and good performance of top of rack, but we also had very good performance of the Interconnects, both the lower end and the higher end.

And, Rami -- we don't break out the specific numbers, Ehud, but Rami, do you want to comment more on the switching part of the business?

I'd be happy to. On the question of the particular segments or subsegments of switching, I can tell you, for example, with the EX9200 the early traction that we saw was very nicely balanced between both campus and data center, and it actually fits really nicely with the strategy that we have been executing towards.

In fact, what we found is that many of our customers deploy our switches in scenarios where it's a common core that is used for both the data center and the campus. The software that we are introducing for our switches in the form of Network Director as part of our broader SDN strategy is very much about converging the management of the switching environment across these multiple layers, as well. So we are pleased with the performance we have seen certainly this quarter.

Okay, Ehud. Thanks for your question. With regards to the relationship within the SP environment between routing and security, as Kevin has described a number of times, we see different trends within service providers of areas of capital spending, and those change over a period of time.

And in the case of security, the area that is most strongly influenced is it follows the RAN, the buildout of the RAN of wireless carriers. As they increase the availability of -- they use their spectrum and increase the ability to move on to newer networks, such as LTE, that is what we see security being most closely correlated to. And what we've seen now is in North America, some of the major carriers have largely gone through their LTE buildouts and so we see opportunities in other geographies around the world.

Now the interesting thing here is that as the capacity increases of the radio network and that is being secured, obviously there is a need for Edge services and that is a routing thing, and as we've discussed over time, that makes the cores run hotter and so there is opportunities there. So the correlation is really based on how the carriers are doing their capital allocation.

And I will just comment on the last part of your question, which was if we just focused on enterprise security and didn't do carrier, would that be a way to run the business for cash, I think was the term, used in your question, Ehud.

And the fact is, the high-end SRX -- the same high-end SRX that we use to secure these LTE networks as we use to secure large data centers. So there may be some slight differences in the features -- software features that are used in that high-end SRX, but fundamentally, the incremental R&D required to be in security for the LTE mobile networks is not significant -- significantly different or more from what you would do in data centers. So we think it's the best return on investment to invest in that R&D and then be able to monetize that same product in both service provider and enterprise.

Okay. Great. And Kevin, thanks and best of luck.

Thank you.

Rod Hall, JPMorgan.

Thanks for giving me the question. Just had a couple of questions for you. First of all, I wonder just kind of housekeeping for Robyn. Could you give us any kind of quantification on that deferred revenue from federal and let us know if there's anything in guidance for further unwinded deferred revenue there?

And then, secondly, I just wondered on the security business, I mean it feels like we still saw a sequential decline in security revenues, but it feels like you've got a grip on that market now a little bit better. I wonder, could you give us some idea -- do you expect that market's revenues or that your revenues from security to be flat quarter on quarter in Q3 or maybe even up a little bit? Or are we still expecting a little bit of decline there?

And then, lastly, you guys seem more and more optimistic about the routing cycle. I wonder if there's any more color you can give us on that. Are you seeing particular regional movement? Is it mostly in North America? Do you see routing traction in Europe? Could you just give us some more on the routing cycle that you see coming the second part of the year?

Okay. Thank you, Rod. In terms of the deferred revenue, the way I'd characterize that the upside in the quarter, for Q2, was half related to the earlier-than-anticipated deferred revenue recognition and half from stronger demand from enterprise and service providers. So that's how I'd characterize it.

As you can see from the balance sheet, the deferred revenue was only modestly down quarter over quarter, and so you can see that we are adding to the deferred revenue pool, as well as recognizing deferred revenue in the quarter. And the deferred revenue balances are healthy. The backlog is healthy and the bookings were healthy in the quarter.

So our view is that in terms of the underlying strength of the business, even if you take out the deferred revenue recognition item, it was healthy.

And so, on the securities side, I will let Bob answer that question.

Yes, Rod, in terms of security and the expectations for the rest of the year, as we said, we expect the security business to stabilize in the second half of this year with a return to growth next year.

In terms of whether it will be flat or up or what over the next quarter, the variability here is perhaps associated with the split between the decline that we've had in the enterprise and the service provider. Q2 was just like Q1 in the sense that the decline was structured as 2/3 of service provider and 1/3 enterprise.

And we have seen -- we see some good traction on the enterprise side. In the service provider, which just the very nature of that business is fairly lumpy, so it's difficult for me to predict on a quarter-to-quarter basis exactly what we will see, but overall we do expect stability going forward for the rest of this year.

And on routing, Rami, would you like to comment?

Yes, I would be happy to. So just to provide a little bit of color, certainly we have been saying for the last quarter or two that the majority of the drivers in routing have been in the Edge.

That continues to remain true throughout Q2, and you have to understand that the drivers for investment in the Edge are just more diversified than that in the core, so people invest in the Edge in order to provide new services, to keep up with capacity, to provide connectivity to various different types of customers. So we had a very good Q2 in terms of Edge routing.

From a geo standpoint, North America is definitely sort of the strong area for us, but we are continuing to see good signs in parts of Europe, both from a standpoint of just the market in general, but also in terms of the design wins that we have gotten over the last several quarters that we are now executing towards delivering for our customers.

And then, just from a segment standpoint, we saw a good balance of investment in routing, both in our cable operators, our content providers, our Web 2.0, and actually also our telco -- traditional telco. So cable operators are dealing with an influx of video traffic that they need to support, telcos with the sort of trend of convergence and delivering on new services, and content providers for Data Center Interconnect for peering and that sort of activity. So a really nice balance of business across a large segment of our customers.

Great. Thanks, guys.

Tai Liani, Bank of America Merrill Lynch.

Hi. This is Eric Ghernati for Tai. Just a quick question on the enterprise again. I guess the [swint] came from federal and financials would settle being due to deferred revenue recognition. The verticals were excessively weak last quarter. I guess, how should we think about the sustainability into the second half of these two verticals? Are they strong or will they -- other verticals or geos will strengthen?

And then, just -- if I recall correctly, AT&T and Verizon were 10% customers. You said that you had no 10% customers. I assume that affected some degree of core routing business, but you voiced your confidence that core routers would rebound second half. Are you seeing that in your bookings? What are your customers telling you, specifically the US large telcos? Thank you.

So the first question you had was on enterprise and we posted an 8% growth in enterprise. Certainly, as Robyn mentioned, I think our product deferred revenue was down a net of $9 million sequentially. So certainly there was some benefit to the US enterprise business from the recognition of that deferred revenue in federal.

But even net of that deferred revenue that was recognized in federal, we still saw improving -- improvements in the performance of federal. And so, things seemed to be stabilizing a bit more and actually getting on more of an improved footing in federal, and we expect that to continue into the second half.

EMEA posted a good growth rate in enterprise as well, and the places I think we are underperforming a bit is in Asia, and I think we have certainly seen some challenges in certain countries, like China and certainly some others. But I think we have opportunity to improve our execution in Asia a bit.

But I think we saw a good, strong enterprise performance in the Americas and EMEA, and certainly I think the fact that switching was up 15% was a part of contributing to that enterprise performance.

On the second question, related to core routing, I commented that -- part of the analysis we do is we look at the total addressable market spend against core and Edge relative to the traffic growth that we are seeing on the networks. And from that, we can start to deduce how much of that was replacement equipment versus new capacity.

And what we are seeing now on the Edge is it is now building out new capacity to handle what was networks that were running hotter and hotter. And we think that trend is going to continue for several quarters. We have seen -- the core of the network, in our analysis, has been running hotter and hotter, and we are just now starting to see the signs that investment will, in addition to being on the Edge, will start to kick in for the core. And we expect to see that start to unfold later this year. Rami, I'll let you comment further on the core routing side.

Sure. The drivers for core are really all around capacity. So customers will tend to defer that for as long as they can before they absolutely need to invest, and we are certainly seeing some of that.

I'll just add that in addition to what Kevin said, I will just add that I think that the strategy that we have been outlining, which is around the diversity of our products in addressing different types of core challenges, whether it be the T series as a very seamless upgrade path to an existing T series network or the MX, especially now with the MX2020 and the denser line cards, or the PTX where we have just introduced the PTX3000. We've announced the PTX3000 to be shipped in the second half of the year, is resonating with our customers, and as Kevin said, we are seeing early traction that the strategy is right on in terms of what our customers require.

Bill Choi, Janney Montgomery Scott.

Okay. First on the clarification part here, did I understand here after the deferred rev recognition on fed, you're still looking for a sequential improvement into the second half, meaning that you will benefit from the potential fed budget flush specifically in the September quarter?

And then, secondly, I want to understand this reinvestment in OpEx for growth opportunity. I guess just six month ago, you guys were trying to cut costs. Now you see some opportunities for additional growth, and just wanted to understand where that is coming from. I mean, part of this improving demand certainly is that you already had the footprint, already had the relationships of those customers come back to spend. Those don't require incremental OpEx investment to capitalize. Is there a lot more opportunities that you are seeing in this current environment for new footprint growth, new projects that require investments? Thanks.

I'm sorry. And then, how long it will take before you get some of the revenue momentum from the incremental investment. Thank you.

I think that last bit didn't make the buzzer, but anyway. In terms of the first part of the question, let me just clarify. What Kevin was talking about was in the second quarter, absent the federal revenue recognition item that we talked about, if you took that out of the numbers, we would have still been up in federal in the second quarter.

And we are expecting a continuation of that trend. In other words, sequential increase in terms of what we saw between Q1 and Q2 on the base demand as continuing for the second half.

So -- and I might add that we also saw good financial services improvements in the second quarter as well. And Kevin also mentioned EMEA enterprise was stronger in the second quarter. So I think that in terms of that underlying business trend, we see enterprise demand strengthening.

In terms of the investments, what I talked about in terms of OpEx, so for the full year we will be down year over year $50 million in terms of operating expenses, 2013 over 2012. We have been very focused on continuing to drive cost reductions across the board and we are still driving those.

Having said that, I have reduced the total year-over-year decline from roughly $100 million to $50 million, and there are two major factors in that. The first is the continuation of the legal and variable expense increases that we have seen versus our expectations at the beginning of the year. And then, the second is some modest investments that we are making in the second half to capture both the near-term and the longer-term revenue opportunities that we see. And so, we think that's a very prudent thing to do from a Company perspective to continue that revenue momentum as we go forward.

Having said that, our OpEx as a percentage of revenue has come down dramatically from last year and we are continuing to focus on that, in line with our long-term model that we put out there this time last year. And I'll remind you that our long-term model for OpEx as a percentage of revenue is somewhere between 39% and 42% of revenue. We still have some ways to go to get to 42%, being the upper end of that range. But we do expect to exit this year close to that upper end of the OpEx as a percentage of revenue range.

Simona Jankowski, Goldman Sachs.

So your service provider business in aggregate was up 2% sequentially, which was about $13 million, and I think you commented that MX Edge routing was up 19% sequentially, which is probably more than $50 million. So it seems like some of the other categories within service provider were down sequentially, and I just wanted to see if you can provide some detail on how that was split between core routing and SRX within the service provider vertical, and specifically if core routing declined?

And since I know you commented on networks getting hotter in the core for the last three quarters, just wanted to get some specificity there on either bookings or backlog or any kind of customer activity that would suggest that that is likely to improve.

So I'll start the answer to that question, Simona, and then I'll hand it over to Rami to talk about some of the trends.

So in terms of our SP business, it was up 2% sequentially, nicely up year over year. In terms of the trends there, yes, MX was up. As you know, we have some older Edge router products that are -- have been declining and they have continued to decline in the second quarter. The core revenue was down sequentially, but the bookings were strong in terms of core.

And as Bob mentioned in the security area, that was -- the SRX was also down sequentially in terms of the security product into SP. And so in terms of the overall trends, Rami can focus on that as well.

Yes, not much more to add. I'll just say that on the Edge, certainly the MX has been very strong. And the -- a component of the Universal Edge strategy that we are executing towards actually involves the other Edge products. Some of the older products that we were working on in the past will decline. So that's completely anticipated.

As far as Edge versus core, I think we have outlined what we are seeing in terms of the drivers in the industry. And these are things that play out over multiple quarters. I don't anticipate any major changes anytime soon.

That said, we did say that we are starting to see early signs that the core investment cycle might be starting, and bookings is one metric that we are using as a reflection of that.

And Rami, just as a follow-up to that, are there opportunities when you think you might be selling a PTX or an MX or MX2020 in place of what would have been a T4000 opportunity? And if that is the case, is that $0.50 or $0.80 on the dollar?

And then, along the same vein with the new routers in core from Cisco and Alcatel, what is your early take on how that is impacting the marketplace, either from a market share or a pricing perspective?

Yes, to your first question, I mean, the way I look at this is that we have a tool bag full of tools at our disposal that we can leverage when we have the conversation and we go after an opportunity with our customers.

So any core opportunity starts with a conversation between ourselves and our customers, trying to understand their business challenges, what they are trying to achieve. And we can go, based on that understanding, reach into our bag to determine whether we are going to be satisfying their requirements with the T series or an MX or a PTX.

And that is a very positive thing because, quite frankly, our competitors don't necessarily have all of those options available for them. So that diversity of offering that we have, the strategy that we are executing towards, I think is really a powerful one.

The second question that you asked is around --

Alcatel and Cisco.

Yes, so the competitive situation. I'll say sort of what I said the last earnings call, which is that I am not going to underestimate any of my competitors, Cisco or Alcatel.

Certainly, Alcatel is trying to get into the core space, but that said, I have great confidence in the strategy that we are executing towards. I think the success that we have seen with the early products -- for example with the PTX has a very dense next-generation transport device that really our competitors do not have. The new multi-chassis offering that we just introduced for the T series -- the T4000, I should say, the early traction there, again, gives me very good confidence.

So we won't underestimate our competitors, but I think that the strategy we're executing towards is a winning one.

Thank you.

Kevin Dennean, Citi.

Just going back to expectations for core routing, it's been asked a few different ways. You are obviously sounding more confident on the outlook. Rami or Kevin, I'm just wondering if you can talk about the expected duration of the core routing cycle. Is there any reason to believe that this is more than just a couple of quarter pickup or is this going to be something more durable along the lines of a multiyear cycle?

Yes, I guess I'll start, and then I'll let Rami complement with additional comments.

In our analysis, it has been several quarters that core and Edge routing have been running hotter. And that is partly why the addressable market for routing declined 3 percentage points in 2012. It wasn't just that one quarter networks are running hotter. It was several quarters. And that really started mid-2011.

There was probably a six-quarter period where, in our observation, core and Edge networks were running hotter. We started to see the pickup on Edge network investment in Q3 of last year. We saw the signs of it and we talked about that, and that led to the results that we delivered in Q4 and what we are seeing today. And we think that, as Rami pointed out, the Edge investments go beyond just incremental capacity. It goes to enable new services and other things that our customers are trying to do.

And so, we believe that Edge cycle is going to continue for quite some time. And what we are now seeing is early signs that the core investment cycle is going to pick up, and that's got to pick up just to address capacity issues. As you build out more Edge, it's funneling more traffic into core.

And certainly, things where a customer will build a metro packet core -- basically, a core within an metropolitan area to keep traffic within that area, we see projects in those areas that will -- you know, in some ways, they put in a core router in that metro -- in that metropolitan area.

So the net is, I think the core -- the signs we are seeing will lead to increased CapEx on the core. You can't exactly predict when it is going to occur, but the early signs, based on what we are seeing in bookings and our discussions with customers on projects, tell us that it is coming soon, and I would suspect it, too, will be multi-quarter, multi-year processes as they build out for what would be a year and a half to two years of running those networks hotter. It's going to take some capacity to catch up with it. Rami, I don't know if you have anything else to add.

Not much more. It's very difficult to predict the timing and the duration of the timing. I think the thing that is in our control is to just make sure that we are best equipped to capture the opportunity when it comes about, and that is exactly what I'm focused on. That's what my team is focused on, and I think that we are going to be prepared, whether it's a solution to converge packet optical, I think we have a fantastic one. Or a migration to it in a T series capacity, I think we've got a great story there as well. So I think that is the thing we are focused on right now.

But Rami, just to follow up, is there anything that you see structurally different in network architectures that would point towards a shorter cycle in core routing versus what we have seen historically?

You know, one can always have an architectural discussion around capacity in the metro versus capacity in the core. And things could evolve over time, but I'd say, at this point, no. There is nothing that points to a radical change in the architecture that is going to change the balance of metro versus core.

It might be the case that you will see some of that happening in some customers than others, but there is no sort of obvious trend that I would say at this point.

Okay. Thank you very much.

Ittai Kidron, Oppenheimer.

Thanks, and good numbers, gentlemen. I wanted to ask about -- going back to this CapEx routing cycle. In the past, you used to estimate the roughly the seasonal spending pattern as 45% in the first half of the year and 55% in the second. Is this something you subscribe to this year? And if this routing cycle does happen, is anything in that pattern going to be different from your standpoint?

Well, I think the routing cycle is happening, and it is more focused on Edge and the core will kick in, and I think that in 2013 we expect sort of the typical seasonal patterns on routing CapEx. So we expect to see an uptick in the second half.

Okay. And lastly, regarding the switching business, good results and numbers there. I guess the question is on the EX9200, when you look at the 8200 and the 9200 together, was there growth in that business or most of that business was really a replacement factor of the 8200?

And second, regarding your QFabric business, it's been -- although the meaning is relatively new, the portfolio itself is a little bit long in the tooth here, and with Cisco introducing a lot of new products just a month ago, when would be the right time to think about an upgrade of your high-end data center switching portfolio?

Yes. So we are pleased with our switching performance in Q2 overall. Certainly, as we introduced a new product, like EX9200, the transition is a very important thing for us to think about as we get our customers sort of thinking about what is the right product for them.

We are pleased with the way that that transition has been rolling out, especially in the Q2 timeframe. And then, the 9200 in and of itself is a very programmable device that is really ready for SDN. That message is really resonating with our customers. So I am optimistic about the growth potential of that platform.

And then on the QFabric, the QFabric remains. I mean, there are many that are talking about fabrics today, but the QFabric remains the single true converged fabric architecture for the data center. That provides us with thought leadership, it opens up doors for us all the time with our customers, and again, the introduction of some of the smaller versions of the QFabric and the new features have provided us with some nice momentum.

And last I'll say is when we think about the upgrade, and we are always at Juniper thinking about next-generation upgrades and roadmaps and technology, so it's all the time.

Very good. Good luck, guys.

Mark Sue, RBC Capital Markets.

Kevin, I hope we didn't wear you out over the past five years.

No, no, not at all, Mark.

Okay. So my question is at the peak of your historic routing cycles I've seen over the last 15, 20 years, the revenue growth can exceed 20%. And at the bottom of the cycle, the revenues can actually be flat to slightly negative. With that in mind, as we go forward, are there things Juniper can do differently to kind of manage the business because it is cyclical -- manage just kind of the mix and the cash flow and the margins?

And then, a question for you, Bob, on security. I guess Juniper has been trying to stabilize the business for some time. You have $2.8 billion in cash, which, incidentally, is the check that Cisco wrote this morning to acquire Sourcefire. So with that in mind, how do you accelerate change for the security division with the limited resources? And is stability in the business worth all the investments that you are making or is it better to kind of reallocate those resources and refocus on wireless core routers or PTX, which is your traditional areas of strength? Thank you.

Okay. So I think Kevin will start on the first question, and then I can add on that, and then we will all hand over to Bob.

Yes. In terms of -- Mark, in terms of your first question on managing the mix, one of the strategies that we think helps us do that is customer diversification. So even within service providers, the fact that we focus on not only just carriers, which is our traditional customer base of wireline and wireless, but also cable and content providers, that helps us manage the mix.

And then, the fact that we've got a strong enterprise business as well that has a wider base of customers and industry verticals, so if one industry vertical is maybe in a down cycle, there's another one that's in an up cycle.

So that diversification of customers across service provider and enterprise is really a core part of our strategy to diversify that mix. Likewise, then, focusing on three products or three business areas -- routing, switching and security -- is also a key part of what we do to try and manage that mix. And certainly, when you see cycles where we have good performance in two of those business areas and maybe weaker performance in the third, that helps balance things out.

Now certainly the challenge is, when you run R&D in the type of projects that Bob and Rami oversee, many of those are certainly multi-quarter, but in many cases multi-year projects. When we take decisions in terms of silicon roadmaps and major systems roadmaps and things we are doing, those become a longer cycle.

And so, the investment that you have to make has to endure both times when revenue is up and revenue is down. But at the core of it, we try and diversify the customer base, both in service provider and enterprise, to ensure that we don't have as much variability in the overall op margins or profitability of the Company.

Yes. And I would also add, you know, obviously, over this last cycle, we have not only learned a lot, but also ensured that we have taken the right actions from a cost perspective and made sure that we are continuing to focus on that side through even a spending cycle that we see with the service providers and also improved demand on the enterprise side as well.

So on the securities side, would you want to comment (multiple speakers)

Sure. I'll make a couple of comments, Mark. Thanks for the question. On the securities side, when we talk about accelerating change, when I think about that, first, we have been very open for some number of quarters that we've had some challenges in our security business, in particular some challenges as we executed the transition from the NetScreen product launch over to the SRX and some of the issues we have had with enterprise.

When I think about accelerating change, I sort of divide that into three sections. There is the set of organic things we can do inside our R&D organization. There is the inorganic things that one might do. And then, there is the go-to-market side. And we are really focused at some level on all three.

Most of our change over the last 18 months has been on our organic R&D where we have made major changes in our management team and our engineering team and have really, really reinvigorated the work we are doing there. We are attracting talent and are continuing to attract talent to the securities business, including in the last quarter where we have added some great new talent that we think will pay dividends in the long run.

On the inorganic side, we haven't done anything at the $2.8 billion level and we don't really anticipate anything like that, but we have done some very strategic, tuck-in acquisitions that have helped us to move into some new business areas. We saw that with Mykonos, you saw with the work that we've done on DDoS Secure. We think those will pay dividends in the long run.

And as I mentioned in my prepared scripts, we are doing some things in terms of our go-to-market efforts on the enterprise and in security to really help accelerate our ability to move forward in that business. So there are changes we are making there.

In terms of your comment about, is it worth it, we pretty strongly believe the answer to that is yes, and I will give you a couple of good reasons for that. One is that when we look at the sorts of businesses that are growing, security overall, we believe, will continue to be a growing business in the years to come. When we see the challenges that our customers are facing with hackers attacking their data center environments and stealing their intellectual property, those threats are going to continue to increase, so a sophistication of the ability to defend against that is also going to increase, and we believe we are very well positioned to actually capitalize on that and actually have a leadership position there within WebApp Secure and Spotlight.

The other thing I will point out is that we see great complements with our security business and our networking business. That is true on the systems side. It is also true as we move into the software side with SDN. And as I mentioned, quite a few of our trials of the SDN work we are doing brings together both our networking side, as well as our security side. We think that better positions us competitively in the market.

All right. That's helpful. Thank you, ladies and gentlemen.

Thank you. That is all the time we have today. We would like to thank everyone for your great questions and for your participation on today's call. Thank you. Bye-bye.

Thank you. Ladies and gentlemen, this does conclude today's teleconference. You may disconnect your lines at this time. Thank you for your participation.