 Live from San Francisco. Extracting the signal from the noise. It's theCUBE covering Nimble Storage, the power of predictive analytics. Now your host, Jeff Frick and Stu Miniman. Hey, welcome back everybody. Jeff Frick here with theCUBE. We are live in downtown San Francisco at one charity at the Nimble Storage Predictive Flash launch. It's an exciting event. We just had a few key presentations with analysts, customers, some executives, and we're excited for our next guest who's out in the field, you know, getting it done, getting his hands dirty. Justin Giardina, the CTO of Island Internet Solutions. Welcome Justin. Thanks for having me. Absolutely, so welcome to theCUBE. So you're here, give everybody a little information on Island, who aren't familiar with the company. Okay, so Island was actually, we've been around for about 26 years now. Company was started in 1995. We have our roots in web hosting, co-location, managed data center services. And really for the first 10 or 11 years of the company, that was the main focus, that got us into about 50 to 60 data centers globally. In 2007. How many data centers? 50 to 60. 50 to 60. And around 2007 is actually when I came on board and our initial offering was to, before actually the cloud was a buzzword, was to actually offer a managed and enterprise grade hosting solution based on virtualization technology. So back then we had things like vSphere 3.5 and we saw really the dawn of all this virtualization stuff. But again, the main focus was to start taking some of these footprints out of these larger data centers, virtualize them. But also giving an enterprise level experience to the customer. So what does that mean? Things like VPNs, firewalling, multi-tenancy, things of that nature. So if you fast forward between 2007 and where we're at today, our main focus is infrastructure as a service still, disaster recovery as a service. We actually won some Gardner and Forster, things like the Magic Quadrant around DR as a service. And then last year we launched our infrastructure as a service with advanced security platform. So in this platform, you can do things like host VMs and whatnot, but we offer a whole suite of security services on top. So for instance, if you fail over to the Island Cloud or run a workload in the Island Cloud, you get things like intrusion detection, vulnerability scanning, anti-virus and so forth. And I promise this is the last thing I'll say, but on top of that, we have what we call our Cloud Console. So we have so many underlying pieces of infrastructure and APIs, things like vCenter, Ness's trend, Veeam, the whole gamut of the ecosystem. And what we've done is we created an overlying platform on top of all this using big data and predictive analytics, so that when you're a customer of Island, you basically have a one-stop shop and one pane of glass in front of all this infrastructure. Wow, so that was a lot of information, Justin. I'm wondering if you'll help unpack for us a little bit. Going from a managed hosting company to an infrastructure as a service provider, I mean, that's a very different skill set. And if you look at the value stack of what you're offering to your users, I mean, you're offering a lot more value and they're usually expecting pretty low prices. So can you talk through us, why you guys went through that transformation, what it took to go through the transformation, did you wholesale when you brought in bringing new people or how did that all come to be? So I think, I mean, if you rewind back to 2005, we started seeing things in the data center like power and space, price, things like that started diminishing. And then we really started seeing the uprising of virtual environment. So you started to see people in the enterprise say, oh, well, I have this web server, I have this low-key application that I'm a virtualized now. You know, I'm going to go ahead and try this, whether it was on the initial offering from BMWare, BMWare server, or actually ESX. And as far as the company goes, I mean, having roots in the hosting business and the data center business, it was actually not as hard as you would think because understanding that level of business, understanding things like SLAs, understanding customer support in remote data center environments really was a catalyst for us to get this. The trick was, how do we take and put out to the world the message of, okay, remember when you had a two-use server and you had 50 of those in Iraq? Well, hey, you don't need that anymore. We can give you this infrastructure at a reasonable cost as well as a high SLA. All right, so you've now got a global footprint with this. Can you talk a little bit about, what does your IT staff look like? What kind of skill sets do they have? And how many people do you have managing all these environments? Okay, so our company is rather small. We only have about 60 people. But I'll say that the success of our company goes back to our staff. We have a bunch of very, very smart people. And these people have allowed us to be able to automate things, to be able to really take advantage of all these APIs I talk about and really automate and streamline the IT process. So if we want to get into in the conversation, we can talk about how we term our pods and how we could take advantage of what technology is out there today. But we're very lean and mean. Yeah, so I guess how much of that has been in-house built and how much you leveraging open source tools and various projects that are out there? That's great. So as far as the underlying infrastructure, we're a Cisco UCS shop, as you know, we're a Nimble shop. We deal with the major vendors out there, Veeam, Zerto, VMware. And we do leverage their technology. The value we bring to the table is that we have developed a whole middleware platform on top of all these infrastructures. And we've written our own code using 100% open source software that we developed in-house. It's not something we outsourced or anything that actually talks to all these APIs and endpoints. And then we abstract all that data in our data warehouse. And we've actually written an API in front of our data warehouse that you can either program if you're a programmer or our portal can leverage this data to provide the analytics and data that a customer would need. Okay, and so how much do you leverage the predictive analytics that Nimble has? So for Nimble, I would say the InfoSight really plays true to the operation side of our business, meaning the guys who are running the underlying infrastructure. What I will say, though, is that when we leverage our own code and we can leverage Nimble's API, vCenter API, we're actually looking at all this data and presenting it to the customer so that they can make an intelligent decision. And that may turn into things like latency or IOPS from a storage level, or try to really compare that a VM is having trouble. Where is that trouble, right? All right, Ken, you talked to us a little bit about your users and how they're thinking about the services you offer. Are many of them ones that you had with hosting? What do they think of the likes of Amazon and Azure? How do those play into what you're doing? Sure, so I mean, you'll hear me say this many times. I think of Azure, Amazon, Rackspace, I think those are all great services. We're not out to pick fights with competing vendors, but we bring a different level of service to the table as a cloud provider than some of these other providers can provide. And what I mean by that is our full suite of products. Whereas a traditional user may go to Amazon, it's been up an instance, and it's got a public IP. We really don't see that as a need for enterprises. They may have a use case for that, but what an enterprise comes to us for is they have mission critical applications. I need things like an SLA. I need to be guaranteed uptime. I need to be guaranteed performance. And oh, by the way, I have this multi-tenant application, it's on totally different networks. I have a load balancer, I have VPNs. These are very complex architectures. So what we bring to the table is we can entertain those types of deals using software-based stacks, software-based networking, and the whole gamut of things we can get into. But that's where we differentiate ourselves in the cloud. And so are you helping them migrate some of that legacy stuff into your cloud to operate it as a cloud? Definitely. So we partner with Veeam and Zerdo, so sometimes customers will leverage those technologies to migrate into the cloud. But we've also written internally a way for a customer to actually maybe create an OVA from vCenter or Hyper-V or export a VM from Hyper-V and actually import that into our cloud. But we also, what's a little bit different about us is we also allow the reverse to be true. So whereas you hear about the Hotel California thing with Amazon, you can go in, but you can never leave. With us, we actually allow the customer to import and export from the cloud. Well, the other conversation we have a lot is, obviously Greenfield, new applications, new opportunities, it's easy to spin it up, make it kind of cloud native. But there's all this old stuff that's running businesses that may or may not make sense to migrate to a cloud, may or may not make sense to get that cloud economics. But in theory, if you could easily, cloud economics offers a lot of benefits, but it's this migration issue that is not easy. That's what we hear a lot about when customers come to us. There's cloud native applications. So we run an open source product called Cassandra. And this is the backend and powerhouse of our data warehouse, right? That application is made to be in something like Amazon. So for instance, there's multiple nodes, nodes can fail. You don't get that kind of stuff with exchange, right? You may be able to set up an HA cluster or a SQL, you may be able to do HA, but it's not cloud native. So what we're starting to see is, A, the larger vendors are coming up with scalable applications to do that. But for the time being, guys that are in their enterprise on legacy applications, we're making it easy for them to get it into the cloud, also extend their data center into us. They might be a VPN or maybe a private line and really helping them accommodate that move. So they feel less like, hey, I'm putting this somewhere and I have no control over it. Yeah, so Justin, I'm wondering if you can speak a little bit about the partnerships of some of your infrastructure providers. So, obviously Nimble, but you're also heavily involved with VMware and Cisco. Does Cisco InterCloud, VMware vCloud Air Network, are those tie into what you're doing? And they're kind of early in their journey to cloud and I'm curious how the partnerships with those three companies specifically work. Definitely, so we're a Cisco CMSP, we're also a VMware vCloud Air Network provider. We've been working with VMware for a long time. We drank the Kool-Aid with a vCloud director 1.x and we've been developing against it for a number of years now. What I can say is that Cisco ICF and InterCloud footprint actually brings a different level of service for a customer. It's got specific needs. It's great for the user base. Also, vCloud director, there's a lot of native integration between vCenter and vCloud Air partners and things like that and we understand that. So we want to be open to the technologies that these vendors are providing to their customers. At the same time, the challenge for us is to provide that single overlay that we've been talking about. Meaning, for instance, if you're doing DR with vMerserto, we want to be able to show you all the metrics around your DR event. We want you to be able to press a button in our portal that says go ahead and failover and all those things. So we hook into all these vendors APIs and we extract the data and then have a common presentation. So it brings up, I guess, so many people think of a company like yours and they say, well, can't you just use white box switching, KVM, virtualization and some software-based cheaper storage? Why nimble Cisco in VMware? So that's exactly right. As I said before, we're a pretty mean and lean staff. There's no argument out there that the VMware technology is tried and true. I mean, we rely on this every day and if we can access it programmatically, it really, we take out that idea of hiring a staff just to run, say, a Ceph cluster or a Swift instance. I personally don't think there's anything wrong with those technologies. It's just for Island, we needed to stick with that. The second thing about that is when we need to provide a high-level SLA and when customers sign contracts with us that say, you need to perform this way and there's no deviation from that, you can understand why we have to go to some of the better players. All right, so you said the company's run on blood-sweetened here. Sounds like you guys have a lot on your plate. Yet, you still have time to, you know, volunteer as a systems admin for open-source community projects. Tell us a little bit about that. Yeah, so I'm from Houston. I'm originally from New Orleans but my wife and I moved to Houston after the hurricane and we started the Houston Ubuntu users group. So we keep a mailing list. The meetings are not as much as we would like them to be but the idea is to help people onboard into Ubuntu. We're a very heavy Linux shop over at Island. As I mentioned, our entire platform other than the vendor stuff like VMware is built on open-source technology and it's just a way for us to give back. We also, I don't have it on my LinkedIn, of course, but our developers and our developer community also submit, they actually take one day a month to totally devote themselves to the open-source projects that they work on. So we're very, very involved in giving back to what's freely given to us. That's excellent. Thank you for, you know, contributing to the community. So give us kind of your perspective on how things on the ground have changed with customers, you know. The cloud thing's been going on for a while. The data and analytics and mobile's long on its path. What are some of the conversations you're having with customers? Where are they on their journey in this whole process and what have you shared some of the insight that you see? So I can say that when we started in 2007 it was very dev and test and putting your toe in the water style things. Nowadays that conversation's really not even on the table. The conversation is, you know, hey, how can I bring this complex situation to the cloud? Not can you run SAP or can you run Oracle? The customer or the user base that we're seeing has actually bought off on it. When they find a cloud provider like us they're actually able to get what they want from a contract perspective, can see the performance of things like that. What also is great too is that the industry and the community in general is building the tools to actually make hybrid cloud be an actual thing, right? So we're starting to see things like you mentioned Cisco ICF, whether or not you're Cisco shopping, whether you're not believing their vision, they give you a very nice package thing to say, okay, I'm going to put this secure switch in the cloud, I know how to connect to it, I can extend my prem to the cloud. And it's just one example of actual vendors being able to make that easier. So I think the discussion about should not, should not go is now when. Right, right. Well, Justin, thanks for stopping by, spending a few minutes with us. Thank you very much. Absolutely. I appreciate it. Justin Jardina from Island Internet Solutions, I'm Jeff Frank, he's Stu Miniman. We are at the Nimble Storage Predictive Flash Launch in downtown San Francisco. We'll be back with our next guest after this short break. Thanks for watching.