 Live from New York, it's The Cube, covering Riverbed Disrupt, brought to you by Riverbed. Now, here are your hosts, Dave Vellante and Stu Miniman. Welcome back to New York City, everybody. This is The Cube, the worldwide leader in live tech coverage. Greg White is here, he's the senior director of cloud and infrastructure at Hertz. It's great to see you, Greg, thanks for coming on. Thanks very much, Dave. So we don't have to ask you what Hertz does, but maybe you could share with us your role there. Sure, basically my role is to help Hertz do a rather significant transformation. Our company has decided to go from a traditional model of IT managed computers applications and kind of one fell swoop to transform the entire thing into a cloud based, very agile environment in a three to five year timeframe. So what's the driver that's leading the business to push IT in that direction? Well, Dave, the driver is agility. As I've heard some of your people talk about in some sessions this morning, one of the things that causes most companies that are customers oriented is to be able to actually come up with new ideas and do methods for the end user experience to be enhanced. So from our company's perspective, that's being driven by our marketing groups for taking a look at how do we ensure our customers have the best experience that we can possibly give them. And looking at an older architectural style, that's not possible or it's very difficult or expensive. So our big challenge is to provide the necessary flexibility in cloud to allow us to implement some of the more innovative social and technological end user experiences. So if you like most organizations, you've got apps that are sort of built with infrastructure that's purpose built to support that app and you've got all the, a bunch of silos, you know, historically, this is kind of what the industry looks like. Is that accurately described sort of your environment and where are you headed? So paint a picture for us of sort of before and after. Sure. Well, most companies have evolved over a period of time and before I came to Hertz, I had to deal with a lot of different kinds of Fortune 500 companies and found a common thread in terms of their application architecture was whatever the soup of the day was kind of, you know, for a while it was dot net and it was Java, it was dot net and it's Ruby and several other things. So we have a whole different set of applications that are ranging from, you know, older applications to relatively new but most of them are not using cloud because we hadn't chosen to go down that path. So we have data centers around the globe and those data centers are used in a typical hub and spoke fashion to service our remote facilities and also to deliver services out to our customers. So what we're now doing is taking a big strong look at that entire chain and updating and modernizing that into the latest and greatest in the cloud space. So when you talk about transforming to the cloud, you're talking about your entire stack, I mean, the infrastructure, the platform and database services and the apps as well. I mean, is that what we're talking about here? Yeah, well it's actually a bit more than that. It's also the workflows that we have internally, approval processes, staffing is a retrain or replace type of a model. Basically taking the company and saying from A to Z what's it going to take to not only to design a new cloud infrastructure but also to operate efficiently and to make it as agile as possible in dealing even with new vendors. So when a new vendor comes with an innovative product we want to be able to seamlessly integrate that in with our long-term architecture for cloud. So Greg, maybe explain to us how important networking is to what you're doing. Obviously you've got locations around the globe, talk about the data centers and the cloud. How does the network fit into it and into your transformation? Sure, the network, we have a traditional hub and spoke network, it operates service well, we have a stable environment. That environment is about to be changed into or transformed into a cloud-based environment. The way we're approaching it is by creating a brand new environment, cloud and network and then looking and taking best practices and then look at how we are going to integrate the older technology network into the new. One of the reasons I'm interested in evaluating riverbed products is because of the innovative technologies that they have to support just that exact model where you can layer a new set of services on top of your existing network. You can also create what I call a proto-cloud which allows you to pre-define what your cloud's going to look like, how your applications are going to operate, how you're going to architect, especially mobile applications to leverage a global distribution cloud versus a very hub and spoke localized based cloud. Okay, great, great. Can you share what cloud services do you use? Are you going to Amazon, Azure, globally? Who do you? Our primary cloud is currently IBM Bluemix, IBM cloud. We also are integrating Azure for some of the things like our security services and some of our management components. We're also looking at some business intelligence and data lake services out of Azure. We also use other vendors like Amazon for kind of bought applications that are already hosted in that platform. We're using Salesforce for some of our CRM based services and we're using Oracle Cloud or some of our financials. So we basically are moving to what people call a multi-cloud strategy pretty much right away. Okay, so I guess one of the questions is, can't those cloud services just provide the networking pieces you need? You know, how does Riverbed help you solve kind of the management of that multi-cloud environment? Well, what I'm evaluating with Riverbed along with some other solutions, but specifically for Riverbed is the WAN optimization component and some of the tools like Net Profiler that allow us to get a better handle on some of our applications and what their dependencies are, how chatty they are back and forth because all the applications are running out of a single data center. As we move to cloud, those applications are going to increase in their latency as they begin talking to one another across the network. So we're looking at using some of the Riverbed products to not only track that, but also to optimize that and make that a little bit more efficient. So you've got, like a lot of, we see this too as sort of laughing when you were describing your HR cloud, your Salesforce cloud and et cetera, et cetera and some infrastructure clouds as well. And do you try to minimize, I mean most customers do, try to minimize data movement, right? But you've got to, like you said, talk to the clouds, have to talk to each other. So how do you manage that? That's a new challenge that you guys are now facing, right? It used to all be inside your data center under your control. You got different security protocols and procedures and how do you guys manage all that stuff? Well, in my opinion, we've hired some of the best people in those industries. I mean, our security team is probably one of the best ones I've ever worked with in my life. And so in terms of taking a look at security on both core legacy and how we're going to go to cloud, the same thing with our data. As we move into normalizing all of our data, which as any big company has data that's scattered all over the place for different applications, we're looking at ways of normalizing all the data and making sure that it's accessible from cloud to cloud. And we're using, I would call best of breed applications along with some of the best talent that I get I've worked with. And you use Riverbed, as you said before, to minimize the latency between clouds. Is that, did I get that right or? Yeah, well, we've used Riverbed Steelhead for WAN optimization in the past. The reason that I'm currently evaluating the Riverbed technology is because of things like the SD WAN that's now available and all three of the clouds that I'm using currently along with Steel Central, which allows us to have a high level of visibility at every layer and every time the packet moves, we can get in and we can look at everything from A to Z. The ability to auto deploy and to make sure that we leverage our high cost in networking, for example, is something that I want to be able to make recommendations to our team, our board, that we don't necessarily have to have a big expensive MTLS-only network. We can have an internet-based network and have a lot of flexibility, save some money, and provide even a better service to our customers. So it's kind of a win, win, win. If you can pull all these pieces together into, again, this proto-cloud and prove it out prior to deploying major applications into it and making a production cloud. And the goal is to move everything to the cloud or when we talk about cloud, are we talking about a sort of a hybrid on-prem? No, our goal is to shut down all of our data centers. Okay. So when you say, well, what are we doing? Well, we're shutting down all the data centers because they have a cost associated with them. And so we want to make sure that we have an ability to take all of that application, set all the hardware, and where it's appropriate, move it to cloud. If it doesn't happen to be appropriate, we'll still have probably some Colo equipment in various places. But the overall goal is to transform us into a state-of-the-art cloud-based company. But so Hertz has decided that running data centers is not its business. Correct. That's exactly right. And that came from the top, presumably. Yeah, and you can go look that up. Our CEO put out a statement about that before they hired me. And, you know, we got board of director approval, and so our marching orders are to create a cloud that is state-of-the-art, that provides maximum agility for the company to service our customers. And those mission-critical applications that we're talking about here in your portfolio, I mean, obviously, you got your reservation system, I suppose you got a fleet management system, and what are the, paint us a picture for the application portfolio, the mission-critical stuff. Sure. The mission-critical stuff is obviously the highest priority because that's where any company gets their bread and butter from. So we typically are, we're looking at best-of-breed vendors as well. So we have a fairly aggressive outsourcing program to obtain and apply the skills that these other companies have learned in deploying this kind of technology. So things like the fleet system would be something that would be outsourced. The res-rent system would be outsourced. When we say we outsource it, so we're providing an architectural overview to those companies so that they will be in kind of compliance with our long-term cloud and architectural strategy. So that way we're not just saying each one can go off and do kind of whatever they want in general. We're trying to make sure that we have five, 10 years, 15 years from now, we still have a clean environment that hasn't been scattered and still is agile. That's one of my architectural goals is to provide that kind of design. And are you having to make your significant custom modifications to those apps, or is that something that you try to avoid doing? Some apps are going to be lift and shift. Some of the apps are greenfield. In other words, we're going to just take a look at the business requirements, reevaluate the business requirements, and then redeploy software using things like microservices, Docker containers, and things like that to provide maximum flexibility and reusability. We're also looking at making sure our development stack is set up for very strict DevOps security and also with continuous integration capabilities so that when one of our business units decides that they want to have a change, we can reduce dramatically the time to market for those changes. How has mobile Greg changed or affected your infrastructure strategy? Mobile is getting a lot of attention because whenever we take a look at the customers who are using the services, most of them are mobile. We already have a lot of mobile capability today. When you get a car, it'll let you, oftentimes they'll send you a text message and see if you want to change to a different car or whatever, so we've already got a fair amount. We're really looking to leapfrog that up into the next generation where most people are very linked to the kind of the Hertz experience in terms of the technology, so that from the moment they get off of a plane or the moment they make a reservation, they have a relationship with us. So it's kind of leveraging and taking the idea of social networking and integrating that in with mobile application delivery. All of those things have to be architected into our platform for those new ideas to come along. So like Facebook was here and then there's LinkedIn, do we want to use that? There's a lot of other social media that at some point may well need to be linked into the experience of ours and any other company. So essentially we always talk about flattening the network. I mean essentially you're flattening your network and you're connecting clouds, right? We always talk about North-South versus East-West traffic. Is that what's happening inside of Hertz? Yeah, the architectural goals are to make sure that we have an ability to be agile and the way to do that is to flatten things out. So we have the general cloud and then I have what we call kind of a hyper-converged or a fog layer that I'm trying to finish designing, which then that pushes the edge of the cloud away from the cloud, which might be considered floating above the business and pushes it down where it's right at the level where the transactions are occurring. For example, at the local rental facilities or even on a mobile device. So it's taking that cloud and pushing it down. So Greg, I wonder if you can tell us a little bit about operationally inside Hertz. If you know, getting rid of the data centers sounds like you're changing skill sets, you know. What happened to kind of the, what happens to the personnel there? How much of it is retraining and repushing around? Is there a change in kind of the headcount going forward? And I guess- Well yeah, that's really probably more of the CEO thing than my thing. I can just tell you that my observation is that Hertz made the decision that they needed to revamp their processes and so they have revamped staffing as well. So we don't have a whole crew of programmers like a lot of companies might have had in the old days because we're not a programming company. So we don't do software development. So what we do do is architecture for the business. So I call the corporate architecture mirrored in with the cloud architecture and blending those and then getting a team of people usually external to create it and then operationally the people internally will run that part. So the way the business operates has changed dramatically in the last year. Yeah, I mean something we've been talking about for years is ITs traditionally spent, you know, the vast majority of their resources on running the business and you need to focus more on growing the business and transforming the business. So it sounds like you have a clear top-down mandate to where your focus is. From the day I started, that was crystal clear that we are about the end customer experience. That is who we're serving. And I think the team that I'm currently working with is doing a really good job of embracing that. So I feel very fortunate to be able to work with a company that's doing this level of transformation. Greg, final thoughts on disrupt, things you're learning here, things you've heard that are exciting, things you can take back to your organization. Well, for one thing, what's exciting is the whole, my thought of Protocloud, I've always liked the model things ahead of time. So there's a lot of tools here, like for example, SteelCentral, is a tool that allows us to monitor a lot of what's going on in terms of the cloud. SteelConnect is something else that's quite exciting because it allows us to prototype at a complete global network without any hardware whatsoever. And we can basically go in, instead of using some architectural tool and drawing boxes and drawing lines, you can literally architect a global solution with tools like that. And then Fusion. Fusion is, Fusion, for any business that designs their applications to leverage that fog layer as that's how I refer to it, that's a powerful tool because it moves all the data mastery out to the very edge of the cloud, provides a lot of HA, a lot of BR capability and a lot of sustainability where your downtime is minimized. And that's really important when you have a business that relies on all these remote offices to generate your business. So overall, I think that in my opinion, Riverbed is dead on. And the other thing that's really good to find out is that their strong partnerships with IBM and Microsoft are very, very helpful to me as I go forward to evaluate all these technologies. Yeah, you mentioned BlueMix, the software integration had to be good news. Oh it was, I saw it up there ago. Was that aimed at me? Because that's what I'm doing that tomorrow. So that was really good news. All right Greg, always good to have the practitioner on theCUBE share your insights, really appreciate it. Thanks Dave. You're welcome. All right, keep it right there everybody. Stu and I will be back with our next guest. We're live from Riverbed Disrupt in New York City. Right back.