>>Hey, welcome to this cube conversation with Dell technologies. I'm Lisa Martin. I've got kosha ghost here with me. He's back on the cube director of product management for unified NAS solutions at Dell technologies. CATIA. Great to see you again. >>Yes. I raped a great to be here again. >>We're going to be talking about the major announcement that Dell technologies just made with their scale-out file storage system that has Dell EMC power scale. We're going to unpack the recent announcement, new features, capabilities, all that good stuff. Kaushik let's go ahead and start. Just give us that high level view of Dell EMC power scale. >>Yes, absolutely. Itself power scale is a high-performance scale-out file storage solution. Um, it's the successor to the Isilon family record, which as you guys know, I mean, there's one of the leading file solutions in the market today. Um, power scale one best, which is the file system that runs on power scale and also the Isilon family, um, is offers an exceptional simplicity, flexibility, and performance, um, which is what Isilon and Parscale is known for. I mean, um, if you look at Gardner's magic ordered one, Fs has been listed as the leader in that, uh, in the distributed and object file system. So, uh, so that basically is now our scaled. We launched our first Parscale all flash products last year. And then this year with this launch, we are sort of completing that portfolio, um, with, uh, with new hybrid and archive, uh, platforms. >>Excellent. And we're going to get into that as well. Let's go ahead and start unpacking this announcement. Walk me through some of the key things that are new and announced in this recent announcement. >>Yeah, except we just launched the hybrid archive platforms, um, on, as part of the Parscale family, are there two archive platforms and two hybrid platforms that we launched and, uh, they offer better CPU, performance, cash, and all that stuff, but, but we don't want to go into the speeds and feeds what I really want to hide breast is the, is the software capabilities that is far skull rings for starters. Um, it is, uh, it now includes inline data compression in 99 reduction. It's all built into it. Um, we support now new ransomware protection capabilities with, uh, with this product. Um, there's a new data protection capability that we now support with our, um, with our, our protected data data manager. Um, and, um, and then the, all the goodness of, uh, Iceland, one Fs and Parscale one Fs that sort of continues. >>I imagine, since the launch last year, cash took a lot of customer conversations that helped to drive this launch and the complete transition on the innovation of what we now see as power scale. >>Yeah. Yeah. I mean, uh, there, there have been some great conversations. People have been, um, people have been really waiting for this product offering because now, uh, they can basically combine those flash platforms that we launched last year with these hybrid platforms and can offer a really a solution that only gives you that performance, but also the, the cost and, uh, savings and the value that, um, that, uh, only our powers skill in Iceland can give you, >>Give me a good overview of some of those key capabilities that the existing Isilon customers and the prospective new customers of power scale are going to be able to take advantage of. >>Yes. So the new, some of the new capabilities as in line efficiency, as I mentioned earlier, that's now built into the product. Um, we have a line efficiency today on our all flash platforms. Uh, so now introducing it with these hybrid and archived nodes, what that means is that when you set up a mixed cluster with all flash and hybrid, when you gear the data down from the national hybrid, the data does not have to be rehydrated. They stay compressed, they stay in protected and so on and so forth. So that's one big advantage that you get. Second, um, these power skill hybrid type platforms were built ground up, uh, with our own custom hardware, unlike the flashback phones with be leverage powered servers for these ones, we use our custom hardware. And the reason for that is because what those archive and storage, the whole story we want that density, we can store up to 500 terabytes of usable capacity, effective usable capacity in, in these archive nodes in a single, uh, one U rack unit. And then, uh, of course, uh, from a software perspective, uh, we talked about ransomware protection. So, so we have a new capability with trends of bear protection. And then there's this new capability that we just launched in regard to backups, more efficient, more faster backups with, uh, with our, our protector will be bought predict family of products. >>Excellent. I want to dig into the ransomware and data protection in a minute, but I want to get a sense of the overall theme of the launch. You talked about this being the completion of that tech refresh some of the new capabilities and enhancements that customers are going to be able to take advantage of it. Give me that higher level kind of thematic look at this news. >>The big team of this is basically finishing that Parscale family that we started last year, right? So we started with launching the whole flash. Now with this hybrid and archive. Now we have the flu family done, um, all products, not support in line efficiency, so we can move the data around, you get the same, uh, data doesn't get high rehydrated. Um, you are, you can make it part of a single cluster. Um, and you get all the performance benefits, um, the scalability benefits of one Fs, um, and new data management capabilities, the, um, so all of that, that we started all of that goodness that we started with our skill, all flash. Um, we soft, continued now with this, uh, with this platform. >>Got it. And I know you guys did your own internal study and I'd like you to share some of the results with the audience, you guys compared power scale to competitors in traditional NAS in flash only NAS in mixed NAS, San and software only NAS give us a snapshot into what some of those results were for power scale. >>Yeah. I mean, uh, the big take away out here is that, um, that when it comes to power scale, um, they, we don't have a competitor when it comes to scalability, right? Uh, the fact that you can now work, uh, on petabytes of capacity under a single namespace, a single file system, and also give you that performance. Um, we, there is none to today, right? And, um, and then there may be some which can do those also, but then they don't have the enterprise capabilities like replication, um, and, uh, the rich enterprise capabilities that one, if fit, sets so off performance, scale capabilities and all the, uh, the simplicity of one Fs. And that's basically what the unique thing about our skaters >>W performance scale and simplicity, three things that I'm sure enterprises, small, medium businesses in any industry appreciate you. You talked about the, um, what's new in terms of the hybrid notes and the archive nodes. Can you help us understand what workloads does nos are best targeted for? >>Absolutely so hybrid and archive. What we have realized is that not every data can be a compressed RDU, right? So, so it's not, we would love customers to use our all flash products. They get the deduplication, they get the compression, then it lowers the cost. And clearly then you get the performance and the cost, but there are workloads like media and entertainment, video surveillance, where you will not be able to compress or that guest, rather than it being for a very expensive flash. You could put those data sets in our lower cost archive platforms as an example. And if you have situations where look, I need some performance, but there is a lot of old data and you can actually mix and match it also. So you're going to have those flash platform is giving that performance. And then you have our archive platforms, which is basically giving you the lowest cost storage for that data. And it is not so frequent giving access. >>And there's the flexibility there. So how can, this is the tech refresh? He said, this has been completed now a power scale from Isilon. How can existing Isilon customers take advantage? What are their next steps to be able to take advantage of the newer capabilities and technologies? >>Yeah, absolutely. I mean, one thing we, we, our scale has it, that's very different from others is that Parscale has this mantra called the no, no left behind. So if you are an existing Isilon customer, you can basically add these Parscale nodes to your existing Iceland cluster without breaking any donut. Then we put our scale, we lock them, magically redistribute, rebalance your workloads across these new nodes. And you sort of keep on expanding our cluster. And when you, when you feel like that, you can, uh, take out the older nodes, uh, at the time of your choosing, right? So that stuff, um, that's a huge benefit that we get. So in fact, in some customer environments, their data has been there for almost 10 to 12 years now, uh, uh, because they've never had to do a forklift upgrade. So that sort of continues with this family. Um, if you learn to learn more about it, I would encourage, uh, going to Dell technologies slash power scale, uh, or contact your Dell technologies, uh, rep >>Let's kind of wrap up things here with talking about, dig into ransomware. We've seen ransomware become a household word, the colonial pipeline, the meat packing organization that was attacked earlier this summer. We know that that a lot of data show that there's a one ransomware attack happens every 11 seconds. And of course we only hear about the really big attacks. Um, I've had the opportunity to talk to a lot of cybersecurity leaders lately, and they're showing that ransomware is up, you know, at least 10 X in the last year with this massive pivot to work from home now, work from anywhere. Talk to me about some of the focus that Dell has put in power scale now with perspective of ransomware protection and recovery. >>Yeah. So for ransomware product, we have to do things that we are doing. So one is this concept of a detection. So when an attack is happening, we want it to be able to detect with date at an attack is happening and take some corrective measures, right? And so we have this product called Sabrina eyeglass, which is exclusively built for, uh, uh, built for, uh, uh, our scale and using this product they use, we use AIS, uh, to basically to figure out that if an attack is happening, we detect it. And based on that based on policies, we can then either, if it's happening with only one user, we can start off, um, uh, start off, uh, prevent, uh, sort of lock it down that particular user profile or, or take other corrective actions taking meaning set up. So that's one aspect of it, which is about the detection of it and taking some quick steps. >>Then there's a second aspect of it, which is all about recovery, right? So, so we do have a replications event. If the customer chooses, we can have reputations set up from your Parscale, uh, production cluster to another cluster. And, um, and in that replication, uh, we can introduce an air gap so that, uh, any, anything bad thing is happening here does not get, uh, uh, does not get replicated to that, uh, remote in mine. So, um, so, so those are the two ways, one detecting, and second basically protecting it. Um, and not only just protecting it, but ensuring that air gap, um, capabilities, data as well, so that, uh, the ransom reason not replicated there as well. >>Absolutely critical. Given some of the things that you and I mentioned a few minutes ago in terms of the explosion of ransomware, which hopefully in our remote, remote work hybrid environment, as more technologies like this come out from Dell technologies and its partners, we'll start to see those ransomware numbers go down. Lastly, I want you to just restate, you mentioned a URL where folks can go to learn more information. Now you've got several different links to point folks to, can you go ahead and remind us what those are again? >>Yes, absolutely. I mean, uh, the easiest you are to go to is Dell technologies slash flower scale. I mean, if you're, that's a one-to-one URL and I'd like you to remember, once you go there, there'll be videos, articles, blogs, and you can, uh, look through a much going and then whatever you want from them. >>Excellent contract. Thank you for joining me today. Talking to me about what's new with power scale, congratulations on the completion of the refresh, a lot of new capabilities and, and, uh, technologies that your customers, existing Isilon and feature perspective, power scale customers are going to be able to take advantage of, look forward to hearing in the next few months in customer success stories. Thanks for your time. Thank >>You >>For Casha gauche. I'm Lisa Martin. You're watching a cube conversation.

>> We just started reselling Veeam We now have a combination of a very strong technology portfolio, deep integration, and a commitment to good market partnership. The combination, we think, will be very exciting for HP, Nimble, and Veeam customers in the years to come. (relaxed electronic music) >> Announcer: Live from New Orleans it's theCUBE covering Veeam On 2017. Brought to you by Veeam. >> Welcome back to New Orleans, everybody. This is theCUBE, the leader in live tech coverage. We go out to the events and we extract the signal from the noise. I'm Dave Vellante with Stu Miniman. Dave Russell is here. He's a vice-president and distinguished analyst at Gartner. David, good to see you. Thanks for coming on. >> Hey, good to see you guys. Nice to see you again. >> So, we were talking off camera. I mean, you are probably the number one known backup, data protection analyst in the business and have been for quite some time. You've seen it all. Give us the state of backup, recovery, data protection, availability, whatever you want to call it. But where are we today? >> You know, in some regards, I don't know if we're any different than we were 28 years ago when I got into the business. The interesting thing is, my wife actually got into this before I did. We were both mainframe developers of backup at IBM and I didn't really want to get a real job. Maybe you could argue I still don't have a real job but what I wanted to do is to stay in grad school forever and I started doing backup there in grad school for undergraduate computing lab. And about six years ago, I showed my wife some of the polls that we do at Gartner events. We can do realtime feedback, what's your greatest challenge, what are your issues with backup? And then she said that was kind of interesting. Two years ago, she came to an event we did in Las Vegas and afterwards she came up and I was hoping she was going to say, "Hey, you did a good job." She said, "What in the heck have you been doing? "These are the same problems when I left the industry 20 years ago to be a mom." Everybody still has too much data, too little backup window, the cost is too high, the complexity is too great. So a lot of infrastructure changes but not a lot of the same pain points have shifted dramatically. What has shifted, though, is cost is even more important than it ever was. Obviously, we could talk about volume of data but now we maybe want to have multiple copies of even our backup data. We want faster access to that backup data. 'Cause we want, now, backup to be a high-availability replication solution not just the tape in the vault somewhere. So there's now speed requirements on our backup. So, I could keep going forever but I'll just let it out to say that as an industry, we still have many of the same challenges that we've always had, arguably for decades and decades. Now, the challenge is the cat's out of the bag, meaning the rest of the business sometimes is aware of just how costly this is. Just how difficult this is from an op-ecs perspective. We can't go hire five, 10 smart people to do this. >> And the backup window, is it correct to say it's essentially disappeared? >> Yeah, there's some organizations that really feel like we don't have a backup window 'cause if we just take a step back, what is really backup, nevermind how you could use it for other use cases like DevOps. Backup, if you state it in the most unappealing terms, it's how much data are you willing to lose, how much time are you willing to take to go get that aged copy of data. And, of course, the rhetorical answer would be well, I don't want any of those bad things to happen, right. But at the end of the day, that's really our frustration. >> David: And I want it back instantly. >> Yeah. >> Okay, so that's obviously putting great pressure on the businesses. So when you look at Veeam's ascendancy, I've been saying it all day and I'd like to test this with you, it sort of coincided, obviously, with VMware and when people had to sort of rethink their VMware backups. You just did a webinar entitled "Backup: Fix it or Ditch it." I feel like a lot of people went through that, answering that question in early VMware days. So, give us, what was the conclusion of that webinar? >> Yeah, well, the number one thing is frustration. And we've done a lot of drill down on what are you frustrated on. Number one is cost, number two is complexity and we could even break this up by large enterprise, mid-size, and smaller enterprise but there's a lot of similarities. So now, where do you come out on fix it or ditch it? The answer for many organizations, is a little bit of both. And what I mean by that, this is kind of mind-boggling, I think, is that backup space used to be sweep the floor. If you were in an incumbent vendor, you wanted to kick out any other solution, if you were an organization, you wanted to collapse from three, five backup products to one backup product, and if you were an emerging vendor, what do you want to do? Go kick out the incumbent vendor. But now, an organization says, "You know, maybe we'd like "to completely change, but we can't. "So we're going to try and fix what we've got." And that's usually what I recommend, at least try and get the value out of what you've already bought and deployed But we're going to implement something else, too. So, there's probably 15 years or more of trying to collapse the number of solutions. Now an organization says, not 'cause I want five solutions but because through pain, basically, not getting my needs met, I'm going to continue running two solutions or expand to two solutions. And you could argue Veeam invented that. They came in on the virtual end, exactly to your point, and then it was a land and expand. We see this happening, though, in the industry overall. >> Dave, I have to think that just the current state of cloud is compounding what you're talking about. Customers have their own data centers, they have virtualized environments. I think Veeam said this morning the average customer they have is only 75% virtualized so they've got 25 physical. Everybody's got SASS, everybody's using some public cloud, at least for some test data. Veeam says that they can now go everywhere but most customers are probably doing piecemeal deployments. Everything in IT is additive. What do you see, how does cloud impact that space in general? >> Well, my biggest fear on the cloud aspect, whether it's software as a service or public cloud, someone's going to rent you infrastructure, is that we're going to learn some lessons the hard way. Again, meaning that most organizations typically think well, if we went to software as a service, they'll take care of it. We have no responsibility anymore or didn't we "get rid of that problem" meaning backup or DR. And the answer is no. You're still the owner of the data. And where it gets shades of gray is that SASS provider's going to give you some level of protection, some level of backup. Chances are they're not going to give you everything you had when you had that email system on premise. So my fear is that organizations are going to suffer an outage and realize there is still a need for additional protection. Right now, many organizations, they're running a bit exposed or don't even realize that they're running a bit exposed. >> Yeah, what is the state of those SASS providers and public cloud providers? Is Veeam still best of breed to go in those environments or are we starting to see them all offer their own native pieces? >> Well, I think we're in a transition period because there's a number of third party solutions that can be good at handling this and you'd have to believe that ... So, take Microsoft for example. They're in the unique position of having had on premise applications and now having public cloud and so eventually, someone's going to say well, here's all the things we did for exchange on premise. Why can't we get all that availability beyond 60, 90 day retention if we go to SharePoint Online or exchange in Azure. There's a tension that's taking place right now. Right now, at this point in time, though, I think if an organization really wants to protect their data like they have and they're used to having been doing on prem, they're going to need a third party solution, whether it's Veeam or someone else. >> David, I want to ask you about your magic corner on data center backup and recovery software. It struck me that ... I don't want to overdo it. I know you guys are very sensitive about each quadrant and how customers should interpret that but we all do the same thing. We go right to the leader. People fight to be in the upper right. And it struck me that Veeam was the only relatively smaller company that sort of knows their way in there. And they're known for SMB but in the magic quadrant you were saying this is really the upper end of M and larger organizations. So what is it that sort of sets leadership apart and how is it that Veeam was able to get in there with those established, much larger players? >> Yeah, that's a great question because exactly what you said, the competitive response would have been isn't Veeam just deployed in small environments? And collectively, we take about two and a half thousand end user inquiry calls a year in backup. So we started seeing a number of trends a couple of years earlier that hey, Fortune 500 companies are deploying Veeam and it's not in the plant in Mexico City or in a small, little area. It's in the Detroit Motor City in the data center and we're seeing a bid for six figures or higher, in some cases. So that's when we started realizing, hey wait a minute. The point of being cast an enterprise supplier is to actually be in the enterprise. They're already in the enterprise. So that's what we started to notice and finally we said another issue we have with putting some of the leaders in quadrants, are they really leading the market or pushing the market? And we really felt that Veeam had kind of crossed over the point last year when we issued the quadrant in June that they were causing the market to shift, whether it was having better virtualization capability, changing to socket-level pricing, addressing ease of use. They were doing things and give sort of "extra credit" for a provider that can not only sense what the market is looking for but kind of push the market. >> Can you explain the socket-based pricing a little bit and how that affected the market? 'Cause I know a number of vendors have made some pricing changes. IBM in particular sort of said everybody can buy anything and use credits there and that was, I felt, a move to keep the install base where it is. Veeama interpreting was different with the socket-based pricing. What was that, did it have an effect on the market in any other way? >> Yeah, the short answer is it absolutely effected the market because you look at the number of heterogeneous backup vendors that have come out and now offer socket-based pricing. So they're doing this in response to Veeam. And what we see now is the organization, depending on who the buyer is, they have no idea what terabytes are. I know what server deployment we have, meaning how much socket we've got so it was just speaking to that constituency in a buying motion that they understood. >> Stu: Something they could quantify. >> Exactly. >> Veeam made a number of announcements this morning and some prior to the show. Anything jump out at you? CDP's one of the ones we've been talking the most. Maybe you could give us your quick competitive analysis of how that looks. >> Well, CDP was near and dear to my heart. In 2005, it was September 2005, almost the same day Microsoft came out with their data protection manager for CDP, Backup Exec came out with CDP. >> Stu: I was trying to remember when Kosha came out because I was at the company that acquired Kosha. >> Yeah, sure. So Kosha, Topio, you know, it can go on. And CDP, around 2005 and 6 was really a lot of buzz, going to change everything. The problem was it was difficult to do because thee infrastructure didn't facilitate it. So, back then you had to split the volume manager and have multiple rights. Now, today's announcement on CDP where you don't have to have a lot of extra infrastructure but it's the hypervisor that's splintering this off for you. IL filtering that's making this easier, making this actually achievable. I think that's going to be really compelling. Most people here I've been talking to say this is going to be great for critical applications. There were some shops I spoke with in the mid-2000s, you know, five, six, seven years, that said we use CDP even on general file systems and why? It's because if I keep making a delete and I call up the help desk and it's like, oh, Dave hit confirm to delete again. He called up to say can you get me my file back and it's the fifth time I've called this week. Well, data protection would allow us to go let him self-service perhaps, but definitely use less data. >> So, for Veeam to get that CDP granularity, if I could talk about that for a second. It's got to obviously rely on VMware APIs. Are you, I'm sure you're tracking this, but are you concerned about Dell EMC gaming the system? Historically, what have you seen there? Difficulty getting hands on SDKs? Trying to put the incumbent in an advantage. What are your thoughts on that? >> Well, you're right. Historically, especially at the storage rate perspective, proprietary APIs or sort of supporting SMIS but having quote "extensions" which are basically proprietary off to the side, were an issue. Here is a case where I think it's in the hypervisor's best interest, and soon it'll be in Microsoft's best interest with Hyper-V and you could go on and on about the other platforms to offer the capability as well. So there is a danger but I don't see how the sort of storage oligarchs are going to be able to fence that off in this case. >> Yeah, I call them the cartel. Is Veeam now, because of its ascendancy, part of that oligarchy? >> Well, I think you have to say approaching half a billion dollars in revenue, it's sort of like the enterprise question. How many enterprises do you have to get in before you enterprise? Well, how many hundreds of millions of dollars do you have to make before you're one of the big ones? >> What do you make of this messaging of Veeam, companies like Veeam, don't want to talk about backup anymore. Backups kind of past ... You see some start-ups like Datos the other day said no, no, we're not a backup company. Okay, and then there's shifting to this notion of availability. Does that resonate with customers? Is that the way customers are thinking about this or is it just sort of good marketing? >> It resonates with some customers. Now, personally, I like it 'cause to me availability is an umbrella. We can put backup and we can put disaster recovery and high availability under there. And maybe you can sort of find a way that DevOps and copy data kind of plays under availability. It doesn't actually work in all geographies. So, I was in Tokyo at a Gartner data center conference three weeks ago, I guess, almost. And they don't really, availability doesn't sound good and disaster recovery sounds worse because that meant you had disaster. So how much disaster recovery do you want to buy? Well, none because I don't want any disasters. So availability is a little regionalized. There are definitely some shops that just say look, I have a backup budget and that's what I need to go and do better. I have a backup pain point, etc. I think, though, whether it's replication and instant VM mounting and the notion of DevOps, we're seeing more and more organizations get their head around ... Whether they want to call it availability or something else but it's beyond backup. >> Well, what's come through loud and clear, however, is your point about cost. I mean, it seems like customers are still insanely focused on cost and that's because backup generally is insurance. So cost and complexity have to be minimized and a lot of the backup platforms that are out there are expensive and they're anything but simple. >> Yeah, and you look at the economics. We've seen negative pricing pressure on dollars per terabyte of backup software now for three years running. Now, list price and obviously, no one really pays list, but list price starting with just a small number of terabytes, some vendors were 10,000 dollars, some vendors were 14 and a half thousand dollars a terabyte and you and I go down to whatever shop and we go buy a terabyte drive, if you can find a one terabyte drive, for a couple hundred dollars. >> David: Four terabytes now. >> And obviously, the data written on it is where the real value is but you see the mismatch of I'm spending list price 14,000 dollars terabytes to protect 140 dollars worth of equipment. There's a problem here. So, whether you're the VP of infrastructure, the purchasing department, or just the backup admin that says I have a problem because I can't go buy now the agent for the database that I'm trying to buy 'cause we've already spent all this money on just the base backup platform. >> Yeah, there's really this 10 year pressure on all infrastructure pricing. Cloud, open source, is really putting pressure on that. So, David, thanks very much for coming on theCUBE. We really appreciate your insights and keep up the great work. >> It was great to see you guys. Thanks for having me. >> You're welcome. Alright, keep it right there everybody. We'll be back with our next guest. It's theCUBE, we're live from New Orleans, Veeam On 2017. (relaxed electronic music)