How & why to move workloads from bare metal


Summary: In this short video, learn how & why having the ability to move workloads from different operating environemts, whether server, hypervisor or cloud provides advantages to your datacenter. With this type of flexilibity a variety of advantages impact resource utilization, licensing, DR and other aspects. 



Mike Matchett:                  Hi, I'm Mike Matchett, with Small World, Big Data. And I'm here today with Don Boxley, the CEO and co-founder of DH2I. We're gonna talk a little bit with Don about what they're up to, in terms of dynamicness, as he put, in the data center and moving databases, in particular around, and doing it in a smart way.

                                                      Welcome. Welcome, Don.

Don Boxley:                           Hey, thanks Mike. Thanks for having me on.

Mike Matchett:                  So, Don, I know that we've talked many times in the past, but you might be a pretty new company to a lot of people. Could you just explain to us what DH2I's all about first?

Don Boxley:                           Yeah, you bet. So we started our company back in 2010. And we started with the objective of helping customers to reduce the number of operating systems that they wanted to deploy, needed to deploy. And we quickly found that the best way to do that was to help our customers put together a more dynamic environment that allowed them to de-couple workloads and move them from one host to another host, anywhere, anytime, within their environment.

Mike Matchett:                  And we talked about like- the analogy that I can come with and most people are aware of is VMWare's DRS that would move workloads around within a hypervisor  environment. But as you just said, you guys took a much broader perspective of that and said, "You can't just look at one virtual hypervisor cluster, you have to look, today, at a much broader span of architecture," right?

Don Boxley:                           Yeah, absolutely. Certainly, we've kind of taken a more heterogeneous approach to it. So within our stack, a customer can move workloads from bare metal to hypervisors to the cloud in any fashion they want to. They only common thing that they all have in place is that they're running a supported version of the operating system that we support. And for us, that's basically and Linux or Windows server operating system, so it covers a broad spectrum of potential infrastructure that the customer might have.

Mike Matchett:                  And we're not talking about picking up and lifting a machine, a virtual machine, we're talking about the actual workload itself. You're gonna move the workload within a machine, which can have some really big, interesting- I remember, some really big, interesting benefits in terms of licensing and stuff. You're not propagating machines across the world and accruing license fees, you're moving the workloads within those operating systems sometimes.

Don Boxley:                           That's exactly it. We think that's the only way you really can do it, is by creating a distributed dynamic management framework around your infrastructure. So the goal would be, you've got certain machines that are optimized for certain workloads, and you just wanna make sure you don't have an overprovision. And to get to your point, when you can move a workload very quickly, within 30 seconds, from one machine to another machine, it fundamentally changes the way you think about how you deploy applications.

                                                      So instead of deploying one significant workload per machine, you can start stacking things to whatever density it makes sense. And when you are able to effectively stack because you can move things around really fast, the savings from machines, licensing, operating systems, it just naturally falls out. So what we try to do is, basically, this idea that we're coming to our customers. They generally have way more infrastructure than they really need. It's just built up over time, for a variety of reasons.

                                                      But now it's time to like, "Okay, we need to transform ourselves." And that's where we come in and help that customer to say, "Okay, what are the key workloads? Let's get 'em into this fabric." And then we'll start taking stuff away until we get to the optimal place for them, with respect to performance and availability costs that they're trying to achieve.

Mike Matchett:                  It's all really interesting. And you guys were kind of pioneers in this idea of multi-cloud and looking for the optimal deployment patterns and making best use of infrastructure, even beyond a lot of other companies we looked at that were doing- ones I worked at were doing predictive modeling. It was pretty cool.

                                                      Tell me a little bit about this latest concept of smart availability and a different way to think about ... Most people think about high availability as kind of a fragile tower kind of concept. How do you guys now think about smart availability, in your words?

Don Boxley:                           Yeah, yeah, yeah. So smart availability, really. Again, it's- for us it's creating this distributed dynamic management framework for your workloads. And when we say dynamic and distributed, the concept is that each one of your workloads can be encapsulated. And then once it's encapsulated, you can move it very quickly from Point A to Point B. So, when you can move a workload- imagine, you've got a very significant workload that you're running on this machine, and now you wanna move it to another machine because of performance, availability. Whatever the goal is, if you can do that in less than 30 seconds without compromising the outside-world connection stream to it.

                                                      So you've got people talking to that process-

Mike Matchett:                  So you're not taking it up or down or anything. It's staying-

Don Boxley:                           That's right. So people are still doing what they're doing, it's just moved from here to there, and you haven't compromised the data that it's responsible for. Now you've got great power, in terms of the ability to transform your environment. A classic that we see for our customers once they've got this framework in place, just bringing in new kit. So the decision's made, "We're gonna move from Vendor A to Vendor B." Not a problem. You bring in new Vendor B, you integrate that stuff into our fabric, and now you just drag and drop workloads off of A into B. So customers can make that migration in seconds, versus what before would take them maybe days and weeks to have to plan and think through exactly how we're gonna do this without disrupting the business that we have to take on to run our business.

Mike Matchett:                  You know, Don, I just wanna ask you one last thing. If someone- I mean, this is kind of a new concept for these people- someone wants to try it or learn more about it, do they ... go to your website and look around? What should they do?

Don Boxley:                           Yeah, it's simple. They just go to our website. and we have- there's a button there for a free trial. Just click on it and someone will get in touch with 'em, and we'll get 'em software, get 'em all set up. Usually we do a 30-day trial. Customers, no cost to them. They can bring it in. Test it out. See if it works.

Mike Matchett:                  And I understand that once somebody's tried it out, they almost don't wanna ever give it back. They're like, "No, no. This is really-"

Don Boxley:                           Yeah. It's pretty sticky. Truthfully, Mike, anybody who's trying to lose weight, which we all are, right? But if you've lost 20 pounds, you kind of wanna keep the 20 pounds off.

Mike Matchett:                  Keep the 20 pounds off.

Don Boxley:                           So, when they bring us in, we reduce customers' infrastructure significantly. They don't wanna lose it. They don't wanna go back to the way they were.

Mike Matchett:                  All right. Well, thanks, Don, for giving us that update. I appreciate you being on the show. Hope to talk to you again soon. Thank you for being here.

Don Boxley:                           Okay, Mike. All right. See you, bye.

Mike Matchett:                  All right. And thank you for watching. And we'll certainly have some more coverage of DH2I in the future. Thanks.