 Live from the Moscone Center in San Francisco, California. It's theCUBE at AWS Summit 2015. Welcome back everyone. You're watching theCUBE live in San Francisco. This is our flagship program when we go out to the events and expect to see the noise. I'm John Furrier, the founder of Silicon Angle. I'm with Mark Farley, distinguished guests, author, blogger, evangelist, expert guru, and Nathan Pierce with F5 Technical, C Technical Marketing Manager, director. Welcome to theCUBE. Thank you, nice to be here. Okay, great to have you on. So what's going on with F5? Okay, you guys have obviously big play in networking. The box business has been around for a while, but been making some moves here all in Amazon. What's the story with Amazon, you guys? Are you guys all in or a little bit in? We are all in, all our functions and services are available via the Amazon Marketplace. And the driver, what's kind of taking us there is push for business agility, faster business. And like you said, boxes, we've been making them for a long time, definitely. We're actually a software company that also makes hardware for those higher-scaling kind of situations. And our customers have been looking to take advantage of cloud and these utility models for quite some time, but they're wanting to take that policy they defined on their F5 equipment to make sure their applications are always fast, highly available, and secure. They wanted to bring that to the cloud with them. So being available via Amazon's Marketplace, we're able to address that for our customers. How's that work on the Marketplace? Explain to the folks out there, because at Reinvent, this became apparent to a lot of the people saying, hey, this is cool, people want to buy that way. How does it work? And what do you guys, how do I engage with you as a customer on Marketplace? Sure, so on the Marketplace, they can go directly on there and select the big IP platform and actually purchase it. They can bring their own license or use utility licensing. So they can even have F5 services hourly or annually if they like, for however time frame that they actually require it. But it's all brought directly through the Marketplace. So it's very easy for them to uptake and it's the same software and platform which is important. So the people that have been using us for years to keep everything running, there's no new learning curve. Once they log in, it's the same administration. It's all very familiar for them. So they can bring that same ratified policy from their private data center into Amazon, into AWS. So what's your take on the whole cloud discussion here? Jassy laid out essentially a platform that's awesome. A lot of goodness coming from Amazon. For an average enterprise who wants to go to the cloud, the network's a big deal. I mean, it's like a lot of stuff going on in the network layer. What's that bridge between the DevOps, network application piece, because applications are driving a lot of the conversation and dictating policy if you will. Definitely, and there's a lot of overlap now between the application architect and the network architect because historically in private, more rigid data centers, people would architect a network and that would be it. It was just connectivity. There were little services in there providing extra value. And then the application architects would be in their own siloed area that was quite a clinical web environment developing their applications until it was ready to go live was effectively thrown over the fence to the network and often they'd pointed each other over who the performance blame was for accessing it. So now with more agile business, faster business, more quicker code development cycles, we've got continuous development that's being thrown out by a lot of the development of DevOps orchestration kind of systems that the network has to be just as agile, just as flexible, it needs to react daily. We've been talking to an organization a few minutes ago that's done 12 releases of their product this year already. So we can't have this total separation and isolation. The network needs to own some of that agility that's coming from the business. As to the application guys, they need to appreciate that the network needs to be aware of state from the application. What's the analytics? How's it performing? How should that affect the way the network works to ensure that there's this symbiosis between the two of we're both there to deliver an app and to create a happy customer. So there's overlap coming between them which I think is great. So what has this done to your own application development cycle? Has that changed in response to this? We've always been a very API focused organization. So we've got a CLI and we've got a GUI for our application delivery platform. However, we made sure that all of that rich functionality for performance availability and security would be available via the API as well. So the application developer can be sitting there and rolling an application out and say, you know what, I could do some things better to optimize this. The app itself can make a call to their F5 platform and it can use its application services, its app fluency to make it run faster. So we always have that philosophy of, make it disposable by an API so that we can keep up with the app development cycles. And Nathan, actually what I was getting to is this transition to being a cloud entity and a cloud business model. Has that had any impact on the software development cycles that you have internally? How has that changed? Because our software is the same whether it runs on our hardware platform or whether it's an instance, an image running out of Amazon's marketplace, it's actually the same. It's the same code maintained across both of them. So we continue the same philosophies before working very close with our customers and our technology alliance partners to ensure that we're addressing the key apps and services they need. But yeah, the difference is the same from our perspective of what we deliver. Okay. Has the partnership model changed at all as a result of this? Obviously, having Amazon as a partner is a big deal. Yeah, yeah, definitely. I mean, utility model is obviously a big part of that. You can't just roll out onto a cloud platform and tell them that they have to buy a license that they're stuck with for life. I mean, it just doesn't work that way. Business agility, you know, and people are rolling apps out that only exist in the cloud for six months. It might be in a development platform and then they pull it back in-house afterwards. So that was a big change, obviously, bringing in these utility models, these cloud license programs. So that's had a huge impact on the way we engage with customers. But also being available in the Amazon marketplaces just opened another route to customers as well. I mean, they can self serve, they can come to Amazon and actually use their five services to be as secure and available as the traditional customers. Yeah, it's a whole other topic. I don't know that we'll go down there with the idea of the Amazon store being a channel. Yeah, yeah. It was a really interesting one. It's a real change, yeah. Yeah, yeah. So what's going on with you guys as a company? Give us the update on your engine software company. What's the big, big thing here at Amazon? What's going on around F5 for the folks that aren't up to speed on what's happening with F5? So customers are really happy that we're on Amazon's platform because they've been using F5 for a long time. I mean, our traditional customers been the global 2000 sort of organizations. They've been asking when are we going to be there on Amazon so we can take our tier one applications out to that platform. So for us, it's been excellent. I mean, our customers are really happy about this, that they can take this tested proven policy that they've been ratifying and working with us to develop and they can just move that policy with their app onto Amazon. So it's common, it's easier to support, it's not a new tool set, it's just giving them an extension of capability in a utility model. So yeah, it's exciting times. I mean, it has changed the way we look at the market, but we're keeping up with the customer demands and expectations, though. What's the customer's perception of the cloud? I have to get there someday. Oh no, drag my feet. Yeah. Seriously, what's your vibe? You do a lot of traveling, what's the sentiment? Yeah, there's definitely, each theater's got a different approach. I mean, you take Australia, where I'm from, it's been a while since I was out there, but it's huge uptake out there for AWS. And they've gone an extra step where they've got financial organizations that have actually ratified cloud where they can put financial data. And yet some of the other regions haven't done that. So it's not a global kind of blanket. Everyone's ticked it off and it's fine. We see little pockets of, this little Nordic country has gone right. All in, government's right to fight it go. And then another region's kind of, I don't share my storage with anyone. Yeah, yeah, yeah. We're still, it's a cultural thing, right? I mean, depends on where you are in the world. Exactly, last time I was up in Scotland, I met a provider that hosted totally isolated environments. So that's back to just managed services models, which was happening years ago because of organizations that won't share. But I think over time, we're going to see that model change. I think there's just some people that have taken a little while. They want to see if there's any epic failure in the media before they go, okay, it's good for me. There's always going to be a few people like that. But I see it as hugely positive. And I don't think it's a great solution. And what do you think that the old, incumbent winners from the past decades need to do to be successful in this new era? So anyone who doesn't match the kind of utility models in the way that they're building and addressing their customer base, I think it's really going to struggle. Anyone who's used to the idea of this tangible object, like a cup of water, I can sell that to you and it's still just a cup of water. That doesn't happen anymore. And there's your cup of water. Now we've got applications. This term I keep hearing around from development people about this phenomenon of spin and kill services. So an entire app or service that have a lifetime that could be an hour or two, just to do a demand campaign, spin and kill, spin it up and then literally hours later, kill it, gone. If you don't have a licensing model or a route to market that matches that. And people who have the mindset of killing and spinning. And the DevOps type people that can handle that mindset, then you're not going to make it in today's market. Business agility is really pushing all of these things. And I think it's really exciting. We've got something fun to talk about again. So talk about DevOps chart, okay. No, no, let's talk about that. The move to containers, Docker, all of this stuff. What are you seeing? When you talk to customers, are they talking to you about containers, what they want to do with them and how does that impact how you'll provide services for application delivery? Sure. And containerization I think is what's really driven this spin and kill kind of phenomenon where the speed at which they can actually throw out a new platform and a stack on top of it is amazing. But for that to really work, you can't just have that in your server tier. The entire data center needs to be able to handle that spin and kill scenario. So if I suddenly spin up 30 containers to just address a short need, maybe it was a launch that we know is just going to get hit, a video clip streams out and it's just that short amount of time. The whole network in front, the security services, the identity services, they equally need to make that jump and be able to jump back up. How do you deal with that? APIs. You don't want people managing that change at keyboards. I mean, one, just the amount of people it takes to do that. But two, points of failure. Keyboard is where a lot of risk happens. We ask our customers all the time, outages, what are they typically? Power phases, whatever. They open the admit. Actually, it's often at the keyboard. The protein robot. Yeah. You don't maybe copy something to stand by data center or you miss a step like that. There's your outage. So if we can do APIs, automate and orchestrate the flow of how we execute these things, that's how we're going to get there and solve these problems. So what's coming up for RSA? Last question before we break. RSA is coming up. What's the big security hot button that you guys are focused on? Obviously, app delivery, security, networking. You guys are a big part of that. I'm sure software. What's the story for RSA? What's coming up? What do you see as the big top line theme for RSA in terms of security in this area? Sure. So, security is, again, it's a problem for security people having to evolve as quick as these rapid changing data centers. I mean, how do you keep your security posture as parity with these services that come and go, you know, in hours, you know, the days? So that's going to be a tough thing for them to address. And a lot of its skills and expertise within the organization. We've moved beyond just data center firewalls that say, you know, allow port 80 or not allow it. We've now got application security. And that's where data theft happens in the application. Cross-site request forgery, SQL injection. That'll happen straight through network firewalls. So, yeah, there's a lot of push now up the stack. You know, people have to be very busy within HTTP, HTML, these kind of protocols. That's where the data's getting stayed in from. So, I'm hoping we see a lot more following that trend. It's something we've been on for a long time. So, the perimeter, does it exist anymore? Or it's perimeterless security, right? I mean. Yeah, I mean, now the app, half the processing is happening out in the browser. So where's the edge of that application experience? It's insane, right? It's everywhere within that session and every device it traverses. So a whole new paradigm needs to be created for security. It is, it is, yeah. Currently is broken. Yes. That's what we want to hear. Welcome to the party, people. It's broken. Security is effed up. But Amazon's got some great stuff. Yeah. And our application security, same things, all available through Bams and Marketplace as well. So, yeah. I'm sure Jassy and team railroaded CloudTrail to the table really quickly. That stuff's really critical. You've got to have the compliance. Yeah. You've got to have the tracking and everything in there. Definitely. Visibility's really huge. Well, security's, you know, no one wants to say it, but it's pretty broken. So, I'm looking forward to hear what's going on. And you guys at RSA. This is theCUBE. We'll be right back after this short break. I'm John Furrier, Mark Farley here live in San Francisco for Amazon Web Services Summit. This is theCUBE. We'll be right back.