How can you avoid vendor lock-in in your cloud strategy
Video Summary: CIO's are actively pursuing strategies to leverage public cloud so they achieve maximum flexibility. The flexibility to choose which provider and to migrate between providers is paramount. In this short video podcast, Zerto explains how their technology helps enable the commoditization of public dloud providers and what's new in their latest 6.0 release. Transcript follows:
Mike Matchett: Hi, I'm Mike Matchett with Small World Big Data, and I'm here today with Rob Strechay from Zerto. Rob's gonna tell us a little bit about their 6-0 launch that they just put out, and how it's helping Zerto go from a hypervisor based company to one that's more involved with actually helping IT succeed across clouds and continuous availability and do bigger and greater things. Welcome to the show, Rob.
Rob Strechay: Thanks Mike, glad to be here. It's always fun to talk with you. Yeah, I think what Zerto's doing and what we're bringing to the table is really helping CIOs execute on their plan to move to the cloud. I think a lot of people have been talking about cloud and how they're going to do it. I think a lot of people have dipped their toe in, maybe, to some extent and they're trying to figure out what is the best way to put a plan in place so they don't have that vendor lock in. And what Zerto is doing is really bringing a platform out that helps those customers really be able to execute a multi cloud, hybrid cloud plan going forward.
Mike Matchett: All right, so when we were talking, over time you guys have been enhancing these capabilities. But a couple of years ago, the idea was really just to get from hypervisor to hypervisor so I could recover things from one place to another. And now we're seeing that people over time have gone from I want to get from my on premise environment into a cloud, and you helped them with that in a couple of ways. And now ... What have you added in six? Right? The ability to come back.
Rob Strechay: Yeah, so I think what we've been doing over the last couple of years, in 2014 we actually took customers to AWS and allowed them to use that as DR as a service in the target. In the past year, we actually did to and from Azure. Now we brought out the, get out of AWS and go home again, kind of a part of it. But we didn't stop there, we allowed our customers to actually go if they need to between Azure and Amazon, Amazon and Azure, back on prem to Hyper-V or VMware, kind of that any to any story.
Having that platform and being able to deploy wherever you want to go and not be locked into one particular cloud or hypervisor really changes the game. It's that any to any capability so that we've been hearing a lot about people looking to move from one hypervisor to another, but how do you do that with reliability and the capabilities. How do you test it out, how do you utilize that? A lot of people are using DR and data protection as a way to start down that path. So it was pretty natural for us to move into that and we had seen not a lot of people want to move between Amazon and Azure because of some costs that are associated with it, but they also don't want to be locked in. So maybe down the road, the ROI or TCO makes a lot of sense for me to move cloud to cloud and we are that platform, so they have that not locked in at that point in time.
Mike Matchett: So I had asked you about ... You know how people were really looking for this for repatriation and you said only a little bit, most of the CIOs thought it was really interesting are saying, no, I want that flexibility, I want to invest in tooling and I want to invest in solutions that allow me to get to a multi cloud, hybrid cloud agility, whatever comes down the pike, right? Whatever comes down. And I thought it was really interesting to emphasize that you're not simply wrapping something around these workloads, you're actually orchestrating and converting them as you go from platform to platform.
Rob Strechay: Yeah, correct. I think that's when people are looking to run in Azure, they're not trying to run as a VMware to VM, they want the tools to be right within the instance. Same with AWS, so it's not just producing a software defined target up there and keeping it in the same format and then you have to bring it back to wherever you started. We're actually allowing customers to go there and stay there if they're going to. So that if they want to use Azure as their secondary data center, they can actually fail up to Azure, and once say they have a data center in Boston and three feet of wet snow falls in through the ceiling, you know, what are you going to do? Well, you go to Azure, probably on the West Coast and you're there. Well, while you're running there, now I want to go Azure to Azure because I don't have another site, and we help them do that. All on the same platform, all on the same supported software.
Mike Matchett: So this is really saying, look, the world's gone, not just hyper cloud, but multi cloud whether people want to recognize it or not. In some ways, data protection is actually leading that multi cloud charge because you might want to recover in a safe place and you don't know where safe is until the disaster actually happens, right. So what's really some of the key capability for doing multi cloud other than, of course, being able to convert? What do I really need to be able to keep in mind if I want to do this effectively as a CIO?
Rob Strechay: Yeah, I think it's understanding who do you have a good relationship for or with. What's your strategy? Geo diversity is a big piece of what we're seeing, because we know a lot of customers, they have still regional data centers. One in Manhattan, one up in Connecticut, maybe out in Pennsylvania or Chicago, but they're looking, how do I get even more geo diversity in there. So I think that having a platform that can take you to wherever you to want to go, also knowing that not every cloud runs the VM the same way, so not every application inside that VM is going to run as well on Azure, as it does on AWS and vice versa.
We actually have a customer that uses both Azure and AWS. Certain workloads are going to go to Azure in an emergency, some workloads are going to go to AWS, because they've done the testing. They've used us and we create a failover test bubble there to do the conversions, bring those applications on line and they do their user acceptance testing. And what they found was, funny enough, some work better in one cloud than the other and they had better response times for their customers. So I think a big piece of what CIOs want to do is, don't just go there and then well, hey am I going to be okay or not and expect that everything is going to work day one. You need to actually put some effort in, put a plan together, and test it as well.
Mike Matchett: It's fascinating. I think we could talk for a couple of hours about how data protection is sort of flipping the whole IT idea on its head that that comes last, and really in a lot of ways if they did this first, they'd have a lot of flexibility in where workloads can be hosted and run and optimized, everything from cost to security and everything out on the line. Just one last thing, tell me a little bit about how you guys are looking at SaaS and thinking about delivering some of your services in a managed way.
Rob Strechay: Yeah, I think that's great. Great question, because I think a lot of people are looking for operational efficiency. And a lot of that is, well, great Zerto has been primarily an on prem software company since it started back in 2011 and what we've done is kind of in the last couple of years started to flip that on its head and deliver some of the applications as SaaS delivered. So we brought out our analytics about a year ago. It made a lot of sense. You have a data repository, it has a lot of data in it, we can put some analytics and some algorithms behind it and help people understand, for instance, what is the recovery point objective and how well have they been trending on their utilization of storage.
Also, it helps profile those applications that protected by looking at how many IOs they're doing, but we can do that in a way that we can deliver new features and functionality on a monthly, quarterly basis versus having to wait a full year, or half a year, as we do two releases a year in our February and August releases, well now I can get one in May and I can get one in June and I don't have to wait around.
I think that where people are going, they've gotten more used to using things like Salesforce and Office 365 and that's driven that comfort level that CIOs have with, hey, I want to leverage these. Many of our customers don't want to manage some huge analytical database on prem and for the management, they don't want to manage the management tools that help them do their data protection. So providing data protection and insight and multi cloud and hybrid cloud visibility, via a SaaS delivered application.
In this last one for instance, we added this past quarter here, we added network analytics because if you're going from place to place with a big workload, a big database, maybe I'm doing ETL every Friday night and repricing annuities in a database. I want to see what the fingerprint of that looks like. We're able to do that and provide that to our customers. Maybe they're doing dragging stuff back from a South American data center in Brazil up to Miami and using Miami as their target for disasters. Well what does that pipe look like? How full is it? Very long, expensive lines, you want to maximize those. So we helped to give them that insight and I think the trend of using SaaS to deliver that, we're going to continue to push the envelope on that and see where customers get the most value, because it ultimately is can they use it, can they get the value out of it?
Mike Matchett: I mean, it used to be that data protection was something you didn't want to change very often, because you got it in place and you tested it and you were just like [inaudible 00:09:59] has got to stay here for 10 years, because there's risk in changing it. And then what you guys are saying is no, look, we're going to change this every two weeks, but we're going to improve it and be agile and help you adapt and in fact lead you to optimize your workloads and your performance on the front line of how your workloads are delivering and it's not simply something you do on the backend once every ten years.
Rob Strechay: Absolutely.
Mike Matchett: Yeah, so well thanks. I know we could talk for a long time, Rob. Thanks for coming and talking today. Glad to have you on the show.
Rob Strechay: Glad to be here, Mike, thanks for having me.
Mike Matchett: And thank you for watching, this is Mike Matchett with Small World Big Data and we'll talk to you again soon. Thanks.