 Live from Las Vegas, it's theCUBE. Covering Veritas Vision 2017, brought to you by Veritas. Welcome back to Veritas Vision 2017. This is theCUBE, the leader in live tech coverage. My name is Dave Vellante and I'm here with my co-host, Stu Miniman. Dave Nettleton is here. He's the Group Product Manager at Google, Dave. Thanks for coming to theCUBE. Ah, thank you, really excited to be here. All right, let's talk storage and cloud. So Google Cloud Platform, we were at your show in March. Kind of the second coming out party, you know, Diane Green at the helm. Obviously, you guys are making serious moves in the enterprise. Give us the update overall and then we'll get into the storage piece. Yeah, well as you say, over the last couple of years, a big focus for Google has actually been shifting and focusing on enterprise customers. I think Gardner reflects that about a trillion dollars of IT spend is going to be affected by the cloud over the next three to five years. And Google has some amazing assets that it's developed over the last 10 or 15 years that we can bring to bed to really help meet enterprise customers' needs, help them where they are and really help transform their businesses for the future. So we're excited about that. Well, so how's that going? I mean, one of the big thrusts that we heard in March was, and we saw it, you guys have made some moves, bringing in people from enterprise companies in particular. You came from Microsoft to see a lot of guys from Cisco. We saw a lot of guys run around from EMC. So Diane herself from VMware, bringing in a lot of that enterprise DNA. How is the patient assimilating with those organs? Yeah, actually, that's been one of the most exciting parts I think of the journey. It's been watching the team kind of come together over the last year or two. As you say, bringing together that pool of talent that has entered one and created even new businesses in the past, it's amazing to see that talent group come together, Diane's doing an amazing job bringing the team together and building out all of the sales functions and other parts of the business that we need for the enterprise. Building out the partner ecosystem as well as obviously super critical. And when you marry that together with the technology assets that Google has, it really is giving customers unprecedented levels of capabilities in the clouds to operate their business in new more efficient ways. So Google's really well known for kind of the analytics piece of the business. Look at all the pieces that have spun out of what Google's done. I'm a networking guy by background. I said when GCP was launched, I said Google's network is second to none. Best network, really understand when the whole wave of SDN came out. Storage on the other hand, one of those kind of foundational pieces, but it's not the first thing that comes to mind. So give us a little bit of a pedigree of the group, what you're building, what differentiates Google from kind of the other infrastructures of service and cloud players. Yeah, and actually you teed it up beautifully because one of our in storage big differentiators is actually our ability to leverage the network. So let me talk you through that a little bit. So Google internally has been building out massive scalable storage systems for years to power the rest of Google. And as we take those to our enterprise customers, we'll find that we're able to leverage that core infrastructure together with global assets like our network. And two parts of the network actually I talked about. One is our wide area network. That allows us to actually, not only store data in regions around the world, but distribute that content through hundreds of points of presence direct to customers very, very quickly. Inside of our data centers we have software defined networks that allow us to separate our compute and storage. That really help us then scale these independently so that we can give massive flexibility and cost savings and pass that through to our customers. And how this shows up in our products, perhaps the best example is if you take something like Google cloud storage, which is our object storage product, that product is very differentiated in the industry in that it provides a single API that will meet use cases from global content serving for customers like Spotify and Vineo who want to stream media content around the world, streaming news, web, media, videos, all the way through to archival storage. Last year we launched our coldline storage class. And this is unique in the industry because it is archival storage that's online. And it has the same API and access as all of the rest of Google cloud storage. So I can take a single piece of data, a video for example, I could be streaming it out to customers around the world globally. And then after a month or two I might decide that I want to archive it. I can archive that down to our colder storage class and if a customer wants to serve it up again they have instant access to it. Yeah, what we're hearing from customers and something we heard in the keynotes here at the Veritas show is customer's cloud strategy is rather fragmented. By that I mean they're not all in on one place to spot certain companies say that. How does that impact your relationship with customers on storage? How do you interact with their SaaS environment, their on-premises solutions as well as what you have inside Google? Yeah, I mean I think fundamentally we believe the world is going to evolve to a sort of a multi-cloud world and that includes both on-premises and public clouds. And as part of that our strategy is to be the most open. And by being the most open that means we need to help customers be portable with their workloads. We need to help them bring their workloads to the cloud for when that's appropriate but also if it's appropriate to take it back to say on-premises to enable them to do that in a very first class way as well. And we think what will happen is some customers will go all in on a particular crowd will be particularly use cases and platform capabilities that will be very differentiated that they want to go all in on. And others will take a more portfolio approach and then partners such as Veritas and others are great for helping customers through information map help them manage that overall portfolio. Could you explain that portability is Kubernetes a piece of it? Is that the primary piece of it? And maybe explain a little bit more how Veritas fits in too. Yeah, so the overall ecosystem is evolving. Kubernetes is obviously a huge part of that environment being able to potably move your compute around. In terms of relationship with Veritas, you know, for me it's all about helping customers solve the problems that they have and meet customers where they are. And if customers are leveraging multiple clouds either because they're using best of breed solutions through acquisitions, et cetera, they need the ability to be able to manage their data across all of those environments. And someone like Veritas with information map is a key partner for us in helping customers meet and manage their needs. So what does that mean for storage? So containers, obviously, for the application portability mobility, Kubernetes is sort of Google's little lever. You know, everybody wants to do Kubernetes and you guys are front and center there. So that gives you credibility in the cloud world. Not that you didn't have it before, but everybody now wants to sort of belly up to you on that. What does that mean for storage? Is that just sort of like an icebreaker for you guys? Are there other things that you're doing specific to storage to take advantage of your expertise there? Yeah, I mean, we want to make sure that customers have a really great integrated experience as they build out their application platforms. So, you know, we're always working with them to better define and understand their needs and build that out. It is a fast emerging, fast evolving space. APIs are still evolving fast. Different layers of the stack are evolving fast. So, you know, we continue to work with customers and just meet their needs through partnerships and also first part. And as you move up the stack sort of beyond sort of networking storage and compute into even database, you know, Google's got some amazing database technologies. Are you doing specific things in storage to take advantage of that? You know, making things run faster or more available or recover faster? Can you talk about that a little bit? Yeah, I mean, the underlying infrastructure at Google powers a lot of our external facing services. So we actually are able to reap very interesting benefits by managing on a single shed TI, technical infrastructure that we have at Google. But as that surface is up to customers, we have to make sure obviously that they can use it in the ways that best meet their needs. But we want to make sure that we integrate their solutions as easy as possible. So for example, Google Cloud Storage that I was just talking about is really well integrated with Dataproc, which is our managed to do product for running big data workloads. And also with something like BigQuery, which is our massively scalable data warehousing solution. So I can store a lot of my unstructured data in Google Cloud Storage and then leverage my entire analytics portfolio to operate over that. And again, a key part of that is the separation of computer networking that we were talking about. When storage is separate from compute and we've used a very powerful software defined network, then that lets us spin up thousands of nodes in something like BigQuery to operate over data and make a very seamless experience for customers. So Stu kind of touched on it before. When people talk about Google and Google Cloud, they point to two things. Obviously the Google Apps suite, okay, boom, you know, we're a customer, love it. Everybody's familiar with it. And the other is data, you know, the data king. And they probably kind of put you in those two boxes. Are you comfortable with that? Is that fair? Is that really the brand that you want? Are you trying to extend that? I wonder if you could comment. Yeah, I mean, obviously our strengths have been in analytics and machine learning and we find that that's the thing that customers are really looking to find ways to add net new value to their business. But we also want to make sure that we're a very trusted provider offering the various high levels of services. And it's not just the capabilities, but overall TCO. We want to make it much easier for people to develop net new applications on the platform. We talked a little bit about some of our open capabilities, but just in general, we want to make it easy for customers to get the best value out of their cloud. So we'll be doing, you'll see us doing more and more there. You know, things we've done have been like being able to create a custom VM images. You can dial up your memory and size, give you a lot of flexibility to really just hone in and solve the problems that you have. So help us square a circle there. And you talk to the cloud, I'll call it pure cloud, folks. People that, you know, born in the cloud, they develop cloud from day one, no legacy infrastructure. We talked to those guys who were like, wow, TCO advantages from a developer advantage, you know, the speed, et cetera. When you talk to the legacy enterprise guys, they'll tell you, oh, it's expensive in that cloud. A lot of people moving back from that cloud. No, of course we know the cloud growth is astronomical. The enterprise growth is flat at best. But there's two different, you know, exact polar opposites. What's just the truth? I mean, the truth is it depends on what you need, right? You know, we think cloud will be a huge disruptor to IT spend over the next several years. It already is. You know, Wineback fiber 10 years ago, I don't think people would even be thinking we'd be having the conversations that we have today. People were like, security, right? You know, I'm not even sure about this cloud thing. Seems like a shared COLO facility to me. I don't think I want to go near that. And it's taken us a while, you know, collectively as an industry to educate really what the cloud is, that it's actually a much more integrated set of services that helps people up-level what it is that they can do. But, you know, one of the biggest challenges we still face in the industry is just education, skills. You know, it takes time to learn new skills. It's encouraging developers, building, working with partners, providing solutions to IT that make it much more turnkey for them to use solutions so they don't have to learn, you know, deep developer skills or, you know, super high-end data science skills to get value out of their data. Yeah, one of the hot-button topics at this show has been GDPR. How does Google fit into the discussion? How are you helping customers get ready for that? Yeah, well, obviously we're very well aware of GDPR and are working really hard to make sure that we're going to be meeting the requirements for our customers as we move forward. We take security and compliance incredibly seriously. So, yes, expect us to see us having full GDPR compliance and then working with partners to make sure that customers can get the confidence that they need for their businesses. So, Dave, as a storage sort of technology guy, what are the big trends that you're tracking as it relates to storage that sort of are driving your Google's thinking? Yeah, great question. So, you know, more and more data more and more data is going to be coming out. Like data has traditionally been siloed. People haven't known where their data is. More and more of that data is now going to be shared within a single environment. And it's not just going to be in the cloud. That data is going to reach both on-premises and also all the way out to the edge. You know, IoT is going to be a huge generator of data. Being able to gather that data, manage that data, provide rich analytics over that data and machine learning and then push that intelligence back out to the edge. So that actually data that's produced can just be analyzed right there. It's going to be super important. I look to say that, you know, data is the fuel for analytics and ML. And that fuel is going to be not just in the cloud on-prem and all the way to the edge. And managing that is going to be super, super, super interesting. I think network, again, network once you start to bring low latency networks to your storage, you can actually start to do really new and interesting things with your data that you never thought of before. You know, if your data, if you can't access it quickly, your data's dark to you. It might as well not be that, right? It has to have things like, how have things like Flash affected sort of bottlenecks? I mean, you mentioned the network. People talk about the network is now the new bottleneck. How is that shaping your thinking? Yeah, I mean, so storage trends continue, densities get higher, speeds get faster. That's a trend that's been continuing. We've been tracking it, continuing to track it. For me, that just means then people will store more data and look to get more value out of that data. Sort of like the latent value of your data is often a function of how quickly you can run machine learning and analytics over that data and get value out of it. And we can do things now to analyze data faster than ever before. I was just thinking of an example the other day. I was running a query myself to look at storage usage. It's something I do regularly. And I ran the query to look at the results. Oh, that's cool. And then I was like, oh, how many rows of data am I querying here? I ran that query. Oh, that was like several billion rows of data that I've just analyzed in like four seconds. I have no idea how much compute power was ran up in the background to meet that query, right? But that's kind of the power that these new capabilities will enable over that data. Dave, how are customers doing with kind of the thing I want to poke at is in their own data centers, their utilization is usually abysmal. And the biggest problem we have is when you do a new technology you do it the old way. How are they doing at really taking advantage of cloud, getting utilization, utility? You know, I'm sure if they go all serverless and per microsecond it'd be much better, but you know, how are they doing? Well, so one of the beauties of the cloud is of course that it's a pay as you go model, right? And with storage and compute being disaggregated, we see customers can provision storage, pay per gig as they go. And then when they need to run compute, they just pay for the compute as they need it. They can shape custom compute instances in GCP, so they only pay for the compute that they need when they finish, they can shut them down. And if you're running something like, for example, a Hadoop workload where traditionally you are provisioning large amounts of compute and storage, sizing for maximum capacity, you no longer need to think about that anymore. You can just store data super cheaply when you want to run a large 100,000, 10,000 node, Hadoop cluster over that data, no problem, you spin it up, spins up in under a minute, run huge amounts of compute, shut it down, and you're done. And actually what we're finding is that this is leading, people are now having to ask new questions of how they manage cost and controls in their business, because this is an incredible power that you can give to businesses, but they also want the controls to say, hey, hey, hey, don't do that too often, or if you do, hey, I want to manage it and manage the cost and controls for departments inside of organizations, so we're building out the capabilities to help customers with that. The last question, Veritas we're here, what do you look for in a partner like Veritas? What do you want from Veritas partnership? So Veritas is a fantastic partner for us, they really help us do the two things that we strive for, which is meet customers where they are today and help them transform their business for the future. So for our integration with their backup, really helps customers in the enterprise just use existing products that they know and love and in a very turnkey way, use the cloud. That helps them manage the costs and meet a lot of demands that they have for their in their IT environments today super easily, so we love that. It also empowers them to do new things in the future, so the integration with information map, we love, helps customers identify new opportunities in their data and add new value to their business. All right, David Nettleton, Google, we'll leave it there. Thanks very much for coming. Thank you very much, it's been a pleasure. All right, we'll keep it right there, buddy. Stu and I will be back with our next guest. This is Veritas Vision 2017. You're watching theCUBE.