>> From Boston, Massachusetts, it's theCUBE! Covering Actifio 2019 Data Driven. Brought to you by Actifio. >> Here we are in Boston, Massachusetts. I'm Stu Miniman, this is theCUBE at the special, at Data Driven '19, Actifio's user event. Happy to bring on a CUBE alum who's a partner of Actifio, Phil Buckellew, who's General Manager of IBM Cloud Object Storage. Phil, thanks for coming back. >> Great, great to be here Stu. >> All right, so object storage. Why don't you give us first just kind of an encapsulation of kind of the state of your business today. >> Sure, object storage is really an extremely important business for the industry today because really it's a new way accessing data, it's been around obviously for a decade or so but really, it's increasingly important because it's a way to cost-effectively store a lot of data, to really to be able to get access to that data in new and exciting ways, and with the growth in the volume of data, of particularly unstructured data, like 103 zettabytes by 2023 I think I heard from the IDC guys, that really kind of shows how important being able to handle that volume of data really is. >> So Phil, I go back, think about 12 years ago, all the technologists in this space were like, "The future of storage is object," and I was working at one of the big storage companies and I'm like, "Well we've been doing block and file," and there was this big gap out there, and kind of quietly object's taken over the world because underneath a lot of the cloud services there, object's there, so IBM made a big acquisition in this space. Talk about, you know, customers that I talk to it's not like they come out and say, "Oh jeez, I'm buying object storage, "I'm thinking about object storage." They've got use cases and services that they're using that happen to have object underneath. Is that what you hear from your users? >> Yeah, there's a couple of different buying groups that exist in the object storage market today. The historic market is really super large volumes. I mean, we're unique in that IBM acquired the Cleversafe company back in 2015 and that technology is technology we've expanded upon and it really, it's great because it can go to exabyte scale and beyond and that's really important for certain use cases. So some customers that have high volumes of videos and other unstructured data, that is really a super good fit for those clients. Additionally, clients that really have the need for highly resilient, because the other thing that's important the way that we built our object storage is to be able to have a lot of resiliency, to be able to run across multiple data centers, to be able to use erasure coding to ensure the data's protected, that's really a large part of the value, and because you can do that at scale without having downtime when you upgrade, those are really a lot of core benefits of object storage. >> Right, that resiliency is kind of built into the way we do it and that was something that was just kind of a mind shift as opposed to, okay I've got to have this enterprise mindset with an HA configuration and everything with N plus whatever version of it. Object's going to give you some of that built-in. The other thing I always found really interesting is storing data is okay, there's some value there, but how do I gain leverage out of the data? And there's the metadata underneath that helps. You talk about video, you talk about all these kinds there. If I don't understand what I've got and how I'd leverage it, it's not nearly as valuable for me, and that's something, you know really that one of the key topics of this show is, how do I become data driven, is the show, and that I have to believe is something critically important to your customers. >> Absolutely, and really object storage is the foundation for modern cloud-native data lakes, if you will, because it's cost-effective enough you can drop any kind of storage in there and then you can really get value from those assets wherever you are, and wherever you're accessing the data. We've taken the same technology that was the exabyte scale on-premise technology, and we've put it in the IBM public cloud, and so that really allows us to be able to deliver against all kinds of use cases with the data sets that clients want, and there's a lot of great innovation that's happening especially on the cloud side. We've got the ability to query that data, any kind of rectangular data with standard ANSI SQL statements, and that just really allows clients to unlock the potential of those data sets, so really good innovation going on in that space to unlock the value of the data that you put inside of object storage. >> All right, Phil let's make the connection. Actifio's here, IBM OEM's the solution. So, talk about the partnership and what customers are looking for when they're looking at their IPs. Sure, so, quite a ways prior to the partnership our object storage team partnered up with the Actifio team at a large financial services customer that recognized the growth in the volume of the data that they had, that had some unique use cases like cyber resiliency. They get attacked with ransomware attacks, they needed to have a standard way to have those data sets and those databases running in a resilient way against object storage that can still be mounted and used, effectively immediately, in case of ransomware attacks, and so that plus a lot of other traditional backup use cases is what drew the IBM Cloud Object Storage team and the Actifio team together. Successful deployments at large customers are really where we got our traction. And with that we also really began to notice the uptick in clients that wanted to use, they wanted to do test data management, they wanted, they needed to be able to have DevOps team that needed to spin up a replica of this database or that database very fast, and, you know, what we found was the combination of the Actifio product, which we've OEM'd as IBM Virtual Data Pipeline, allows us to run those virtual databases extremely cost-effectively backed by object storage, versus needing to make full replicas on really expensive block storage that takes a long time. >> Well yeah, we'd actually done research on this a number of years ago. Copies are great, but how do I leverage that right? From the developer team it's, I want to have something that mirrors what I have in production, not just some test data, so the more I can replicate that, the better. Phil, please, go ahead. >> There's some really important parts of that whole story, of being able to get that data flow right, to be able to go do point-in-time recoveries of those databases so that the data is accurate, but also being able to mask out that PII or sensitive information, credit card data or others that you really shouldn't be exposing to your testers and DevOps people. Being able to have the kind of-- (Phil laughs) >> Yeah, yeah, shouldn't because, you know, there's laws and lawsuits and security and all these things we have. >> Good, good, absolutely. >> So, Phil, we're talking a lot about data, you've actually got some new data to share with us, a recent survey that was done, should we share some of your data with us? >> Yeah, we did some, we did a, the ESG guys actually worked with us to build out a piece of research that looked at what would it cost to take a 50 terabyte Oracle 12c database and effectively spin up five copies the way you traditionally would so that different test teams can hammer away against that data set. And we compared that to running the VDP offering with our Cloud Object Storage solution. You know, distances apart, we had one where the source database is in Dallas and the destination database is in Washington, D.C. over a 10 gigabyte link, and we were able to show that you could set up five replicas of the database in like 90 minutes, compared with the two weeks that it would take to do full replication, because you were going against object storage, which runs about 2.3 cents per gigabyte per month, versus block storage fully loaded, which runs about 58 cents per gigabyte per month. The economics would blow away. And the fact that you could even do queries, because object storage is interesting. Yes, if you're using, if you have microsecond response times for small queries you got to run some of that content on block storage, but for traditional queries, we look at, like, really big queries that would run against 600 rows, and we were half the time that you would need on traditional block storage. So, for those DevOps use cases where you're doing that test in development you can have mass data, five different copies, and you can actually point back in time because really, the Actifio technology is really super in that it can go do point-in-time, it was able to store the right kind of data so the developers can get the most recent current copies of the data. All in, it was like 80% less than what you would have paid doing it the traditional way. >> Okay, so Phil, you started talking a little bit about some of the cloud pieces, you know, Actifio in the last year launched their first SaaS offering Actifio GO. How much of these solutions are for the cloud versus on-premises these days? >> Absolutely, so one of the benefits of using a virtual data approach is being able to leverage cloud economics 'cause a lot of clients they want to do, you know, they want to be able to do the test in dev which has ups and downs and peaks and valleys when you need to use those resources, the cloud is really an ideal way to do those types of workloads. And so, the integration work that we've done with the Actifio team around VDP allows you to replicate or have virtual copies of those databases in the cloud where you want to do your testing, or we can do it in traditional on-prem object storage environments. Really, whatever makes most sense for the client is where we can stand up those environments. >> The other thing I wonder if you could expand on a little bit more, you talked about, like, cloud-native deployment and what's happening there. How does that tie into this discussion? >> Well, obviously modern architectures and ways of Agile, ways of building things, cloud-native with microservices, those are all extremely important, but you've got to be able to access the data, and it's that core data that no matter how much you do with putting Kubernetes around all of your existing applications you've still got to be able to access that core data, often systems record data, which is sitting on these standard databases of record, and so being able to have the VDP technology, be able to replicate those, stand those up like in our public cloud right next to all of our Kubernetes service and all the other technologies, it gives you the kind of full stack that you need to go do that dev in test, or run production workloads if you prefer from a public cloud environment, without having all of the burdens of running the data centers and maintaining things on your own. >> Okay, so Phil, everybody here for this two day event are going to get a nice, you know, jolt of where Actifio fits. You know, lots of orange here at the show. Give us the final word of what does it mean with orange and blue coming together. >> Well absolutely, we think this is going to be great for our clients. We've got, you know, tons of interested clients in this space because they see the value of being able to take what Actifio's done, to be able to virtualize that data, combine it with some of the technologies we've got for object storage or even block storage, to be able to serve up those environments in a super cost-effective way, all underlined by one of our core values at IBM, which is really trust and being responsible. And so, we often say that there's no AI, which all of this data leads up to, without information architecture and that's really where we specialize, is providing that governance, all the masking, all of the things that you need to feel confident that the data you've got is in the right hands, being used the right way, to be able to give you maximum advantage for your business, so we're super excited about the partnership. >> Phil, definitely a theme we heard at IBM Think, there is no AI without the IA, so, Phil Buckellew, thanks so much for joining us, sharing all the updates on what IBM is doing here with Actifio. >> Great, great to be here. >> All right, and we'll be back with more coverage here in Boston, Massachusetts at Actifio Data Driven 2019. I'm Stu Miniman and thanks for watching theCUBE. (futuristic music)