 From around the globe, it's theCUBE with coverage of KubeCon and CloudNativeCon North America 2020, virtual, brought to you by Red Hat, the CloudNative Computing Foundation, and Ecosystem Partners. Hi, I'm Yuppie Scott, and welcome to theCUBE's coverage of KubeCon, CloudNativeCon 2020. So I'm joined today by Matt Kicksmiller, he's VP of strategy at Pure Storage, as well as Michael Forenzi, he's the senior director of product marketing at Portworx, now acquired by Pure Storage. Fellows, welcome to the show. Thank you. Thank you for having me. I want to start out with, you know, the kind of lay of the land of storage in the CloudNative space, in the Kubernetes space. What's, you know, what's out, what's happening, what are the trends that you see going on? Matt, if you could shed some light on that for me. Yeah, I think, you know, from a pure point of view, obviously we just thought of customers really maturing their Kubernetes deployments and particularly leaning towards persistent applications. And so we noticed within our customer base that there was quite a lot of deployments of Portworx on Pure Storage. And that inspired us to start talking to one another, you know, almost six plus months ago that eventually ended in us bringing the two companies together. So it's been a great journey from the pure point of view bringing Portworx into the Pure family. And, you know, we're working through now with our joint customers integration strategies and how to really broaden the use of the technology. So it's quite exciting times for us. And of course, it's good to hear that the match goes beyond just the marketing color, like the brand color. Absolutely. Yeah, we, you know, the fact that both companies were orange and, you know, their logo looked like kind of a folded up version of ours. Just started things off on the right foot. Imagine it in heaven, heaven, right? So I want to talk a little bit about the, you know, the acquisition, what's happened there. And especially, you know, looking at Portworx as a company and as a product set, it's fairly popular in the cloud community. A lot of traction with customers. So I want to zoom in on the acquisition itself and kind of the roadmap going forward, merging the two companies and adding Portworx to the Pure portfolio. Matt, if you could shed some light on that as well. Yeah, why don't I start and then Michael could jump in as well. So, you know, we at Pure had been really working for years now to outfit our all flash storage arrays for the container use case and shipped a piece of software that we called PSO that was really a super CSI driver that allowed us to do intelligent placement of persistent volumes on Pure arrays. But the more time we spent in the market, the more we just started to engage with customers and realized that there were a whole number of use cases that didn't really want a hardware based solution. You know, they either wanted to run completely in the cloud, hybrid between on-prem and cloud and leverage bare metal hardware. And so, you know, we came to the conclusion that, you know, first off, although positioning our arrays for the market was the right thing to do, we wouldn't really be able to serve the broader need for storage for containers if you did that. And then, you know, the second thing I think was that we heard from customers that they wanted a much richer data management stack. You know, it's not just about providing that this is the volume for the container, but, you know, all the capabilities around snapshotting and replication and mobilization and mobility between on-prem and cloud were necessary. And so, you know, Portworx really brought to bear not only a software based solution into our portfolio, but really that full data management stack that platform, in addition to just storage. And so as we look to integrate our product lines, you know, we're looking to deliver a consistent experience for data management for proven entities. Whatever infrastructure customers might choose, whether they want to run on all flash arrays, white box servers, bare metal on VMs or on cloud storage as well. You know, all of that can have a consistent experience for the Portworx platform. Yeah, because, you know, data management, especially in this world of containers is, you know, it's a little more difficult. It's definitely more fragmented across, you know, multiple clouds, multiple cloud vendors, multiple cloud services, multiple instances of a service. So the fragmentation has, you know, given IT departments quite the headache in operationally managing all that. So Michael, what's, you know, what's kind of the use case for Portworx in this fragmented cloud storage space? Yeah, it's a great question. You know, the use cases are many and varied. You know, to put it in a little bit of a historical perspective, you know, I've been attending Kube-Cons either internal or logical for about five or six years now, kind of losing count. And we really started seeing Kubernetes as kind of an agile way to run CICD environments and other test dev environments. And there were just a handful of customers that were really running production workloads at the very, very beginning. If you fast forward to today, Kubernetes is being used to tackle some of the biggest central board level problems that enterprises face because they need that scale and they need that agility. So, you know, COVID's accelerated that. So we see customers say in the retail space who are having to cope with a massive increase in traffic on their website, people searching for kind of, you know, the products that they can't find anywhere else. Are they available? Can I buy them online? And so they're re-architecting those web services to use often open source databases, in this case, Elasticsearch, in order to create great user experiences. And they're managing that across clouds and across environments using Kubernetes. Another customer that I would say, kind of a very different use case, but also one that matches that scale would be Esri, which unfortunately the circumstances of becoming a household name are a lot of the COVID tracking ArcGIS system to keep track of tracing and outbreaks. They're running that service in the cloud using Port Works. And again, it's all about how do we reliably and agilely deploy applications that are always available and create that experience that our customers need. And so we see kind of, you know, financial services doing similar things, healthcare, pharmaceutical doing similar things. Again, the theme is it's the biggest business problems that we're using now, not just the kind of the low-hang fruit as we used to talk about. Yeah, exactly because, you know, storage is a lot of the times it's kind of a boilerplate functionality. It's, you know, it's there, it works. And if it doesn't, you know, the problem with storage in the cloud data space is that fragmentation, right? Is that enormous, you know, on the one hand that enormous scale on the other hand, the tons of different services that can hold data that need protecting as well as data management. So I want to zoom in on a recent development in the Port Works portfolio where the PX backup product has spun out into its own little product. What's the strategy there, Michael? Yeah, so I think, you know, fundamentally data protection needs to change in a Kubernetes context. The way in which we protected applications in the past was very closely related to the way in which we protected servers because we would run one app per server. So if we protected the server, our application was protected. Kubernetes breaks that model. Now an individual application is made up of dozens or hundreds of components that are spread across multiple servers and you have container images, you have configuration, I mean, you have data. And it's very difficult for anyone person to understand where any of that is and cluster at any given moment. And so you need to leverage automation and the ability for Kubernetes to understand where a particular set of components is deployed and use that Kubernetes native functionality to take what we call application aware backups. So what PX backup provides is data protection engineered from the ground up for this new application delivery model that we see within Kubernetes. So unlike traditional backup and recovery solutions that were very machine focused, we can allow a team to back up a single application within their Kubernetes cluster, all of the applications in a namespace or the entire Kubernetes cluster all at once and do so in a self-service manner where integrated with your corporate identity systems, individuals can be responsible for protecting their own application. So we marry kind of a couple of really important concepts. One is kind of the application-specific nature of Kubernetes, the self-service desire of DevOps teams, as well as with a pay-as-you-go model where you can have this flexible consumption model where as you grow, you can pay more. You don't have to do an upfront payment in order to protect your Kubernetes applications. Yeah, I think one key thing that Michael hit on was just how this application is designed to fit like a glove with the Kubernetes admin. I see a lot of parallels to what happened over a decade ago in the VMware space when VMware came about and needed to be backed up differently and a little company called Veeam built a tool that was purpose-built for it. And it just had a really warm embrace by the VMware community because it really felt like it was built for them, not some legacy enterprise backup application that was force-fitted in this new use case. And we think that the opportunity is very similar on Kubernetes backup and perhaps the difference of the environment is even more profound than on the VMware side, where the Kubernetes admin really wants something that fits in their operational model, deploys within the cluster itself, backs up to object storage, is just purpose-built for this use case. And so we see a huge opportunity for that. And we believe that for a lot of customers, this might be the easiest place for them to start trying to portworks portfolio. You've got an existing Kubernetes cluster. Download this, give it a shot. It'll work on any infrastructure that you've got going with Kubernetes today. Especially because looking at the kind of breakdown of Kubernetes in a way, data is infrastructure provisioned, data is placed in cloud services. It's no longer the cluster admin necessarily that gets to decide where data goes, what application has access to it. That's in the hands of the developers. And that's a pretty big shift. It used to be the VI admin, the virtualization admin that did that, had control over where data was living, where data was accessed, how it was accessed. But now we see developers kind of taking control over their infrastructure resources. They get to decide where it runs, how it runs, what services to use, what applications to tie it into. So I'm curious how portworks and MPX backup kind of help the developer stay in control and still have that freedom of choice? Yeah, we think of it in terms of data services. So I have a database and I need it to be highly available. I need it to be encrypted, backed up. I might need a DR, an offsite DR schedule. And with portworks, you can think about adding these services, HA, security, backup, capacity management as really just, I wanna check a box and now I have this service available. My database is now highly available. It's backed up, it's encrypted. I can migrate it. I can attach a backup schedule to it. So within a Kubernetes cluster, some apps are gonna need that entire menu of services. And some apps might not need any of those services because we're only in test dev page. Everything is multiplexed into a single cluster. And so being able to turn off these various data services is how we empower a developer, a DevOps team to take an application all the way from test dev into production without having to really change anything about their Kubernetes deployments besides a flag within their YAML file. Makes it really, really easy to get the performance and the security and the availability that we were used to with VM-based applications via that admin now within Kubernetes. So Matt, I wanna spend the last couple of minutes talking about the bigger picture, right? We've talked about PortableHPX backup. I wanna take a look at the broader storage picture of Cloud Native and kind of look at the pure angle on the trends and what you see happening in this space. Yeah, absolutely. So like a couple of high-level things I would kind of talk about. First is that I think hybrid cloud deployments are the de facto now. And so when people are picking storage, whether they'd be storage for a traditional database application or a next-gen application, cloud native application, the thought from the beginning is, how do I architect for hybrid? And so within the pure portfolio, we've really thought about how we build solutions that work with cloud native apps like Portworks but also traditional applications. And our cloud block store allows those to be mobilized to the cloud with minimal re-architecture. Another big trend that we see is the growth of logic storage. And if you look at the first generation of logic storage, object storage is what, 15 plus years old and many of the first deployments were characterized by really low cost, low performance, kind of the last retention layer, if you will, for unimportant content. But then this web application thing happens and people started to build web apps that used logic storage as their primary storage. And so now as people try to bring those cloud native applications on-prem and build them in a multi-cloud way, there's a real growth and a need for high performance kind of application-centered object storage. And so we see this real change to the needs and requirements on the object storage landscape. And it's one that in particular, we're trying to serve with our FlashBlade product that provides a unified file and object access because many of those applications are kind of graduating from file and moving towards object, but they can't do that overnight. And so being able to provide a high performance way to deliver unstructured data, whether the app wants object file or both is very strategic right now. Well, that's insightful. Thanks. So I want to thank you both for being here. And I look forward to hearing about Portword Compure in the future as is acquisition integrates and new products and new developments come out from the pure side. So thanks both for being here and thank you at home for watching. I'm Yuppie Skach. Thanks for watching the cubes coverage of KubeCon Cloud Native Con 2020. Thanks. Thanks, Yuppie. Yeah, thank you.