Transcript
Hi Mike Matchett with Small World Big Data. We are here talking about hybrid multi-cloud workload portability. What does that really mean? It means if you want to really be hybrid, or you want to repatriate workloads, or you want to take advantage of multiple clouds, or you want to have a good business continuity plan that really works, you need to be able to move your instances around with the data that goes with them and everything else around your hybrid infrastructure. It's complex at best. So hang with me and we're going to talk to Data Motive here in just a second. Hey. Welcome, Samir. Welcome to our show. Thank you. My pleasure to be here. All right. Data motive. You guys are tackling. I mean, in one word, we could call it replication. Workload replication in two words. But there's a lot more wrapped around that. But first, just tell me a little bit about how you got involved in this end of the IT market, moving workloads around and taking advantage of clouds. How did how did how did you get dropped into this? Yeah. So? So Mike have been in this world for like 25 odd years, right? And in the last ten, 15 years have spent a lot of time around data centers and helping large organizations doing transformations. Right. And every time when we talk about migration or D.R., the one question was, can I get my when I'm migrating, can I get 100% data consistency? Second is how do I reduce the impact of business services to my customers, right. These were always the two questions and I don't believe we had the right answers. Right. We always kind of said, yeah, we can do this, we can do that, but we can never guarantee a flat line as part of migration or D.R.. And that's where this whole data motive came into picture, where we said, let's try to solve this problem. And that's how it all started for us and for me, myself. All right. So you now can can with data motive say, and you've got it on your your background there ten minutes. We can move your workload from any place to any place if we're already involved there doing the replication in the background. Yes. So what data motive brings to the table, Mike, is we are a cross-cloud replication and recovery engine, right. Essentially, we can take your entire workloads, lift and shift them from any cloud to any cloud. And we bring a flatline recovery of ten minutes, right? Irrespective of which cloud you're coming from, what is the type of workload and what is the size of the workload? That is what data motive brings to the table. All right. We're going to get into we're getting into the clouds that you can move between here in a second. But first just define workload a little better for us. When you say workload what do you mean. Yeah. So workload for us what it means is the entire virtual instance that is above the hypervisor, which includes the guest OS. It includes the platform. It includes the applications, the configurations, any agents deployed on on that instance, plus any data that is attached to that instance. So anything, anything in storage, anything being written and read, that's all part of this as well. So that's where your consistency comes in. It's not. Yeah that's correct. All right. So this is a pretty big challenge because you know in the past you'd have to maybe take a VM, put it in a key asset, put it in a quiet mode, shut it down, make a copy. That's the copy that you can actually then move somewhere else and start up again. And you interrupt business service. And you basically take things out of the flow of what they're trying to do in production. That is so correct, Mike. What we at Data Motive is that we are an online incremental agentless replication engine. Right? So what it means is that you don't have to shut down your source. Workload will continue to replicate. We don't sit on the workload. So there is no performance impact. There is no security impact. Right. And the best of that is there is no no element of streaming of any kind when you are replicating the entire workload, right? Today, you have to even if you get your instance ready, your streaming data right where and and hence the consistency is becomes a question mark. Right. With data motive. No no no no. Streaming will copy your instance as it is right as you see on the source. And we will copy it on the target and replicate and recover on the target. All right. And now you're able to offer this ten minute, this ten minute recovery or SLA, regardless of which cloud I'm coming from is in target. Right. So you're supporting what are the different clouds you can move in and out of. So today we support VMware and Azure as the source cloud. So your workloads can come from any of these clouds. And we support VMware AWS Azure and Google Cloud as the target clouds. So it's a matrix. Any of the source cloud can go any of the target clouds. And each time you will see a recovery of ten minutes, ten minutes. I think that's pretty amazing, right? That you can that you can just make that statement and stick to it. Because most tools we've seen that do replication or backup and restore or something, you know, they would never say, you know, you only have this many minutes. They would say, you know, it's good for ransomware, cyber resiliency or something because you can recover in within days instead of weeks or months, right. They're not saying ten minutes. This is pretty fun. Yeah yeah yeah. So and then and then I understand also the way you're doing this, you're going to lower you're going to lower the, the basically the TCO of what someone might normally do. How do you sort of attack the cost side of this for someone. Yeah. So. So the way. There are two parts to it, right? So if you if you look at steady state replication, right. The two major cost elements are the egress costs. That is the cost for coming transferring data out. And the second is the infrastructure cost on the side or the alternate side. The way data replication engine works is that in steady state replication we will use only the native storage. We don't need new storage arrays or any of that kind. Whatever is available, we will use that right. And storage with with. When we do that, there is no compute. There is no network assets that are running. So essentially what is happening is that you are only copying data and using the storage services of that target cloud, right? In a sense, we are also a sustainable product because we reduce the consumption of compute and network on the target cloud, right? That that that brings a huge cost savings. Right. The second element is we have the engine allows you to compress data when you are transferring. Right. And if based on the type of data, we have seen anywhere between 30 to 70% compression. So essentially you are paying less for your data to be copied over. And that's how we, we, we we have seen up to 60, 65% reduction in costs without the compromise on the recovery of Selene. Yeah, that's pretty amazing. And I think there's some even bigger benefits here. When we talk about the fact that you do this kind of reverse, you call it incremental differential reverse differential replication. So if someone's failing over and then failing back, they're also saving on that because you're just even reversing the incrementals you're not copying back huge chunks either. Right? Correct. No data motive. The way we have built on the principle is that in an event, you always want to come home, right? Once the event is resolved, right? Today coming back home is the big challenge, right? Either you are recreating the entire infrastructure or you're like you said, you're copying the entire instance back home, right? We solve both the problems because we, like you said, we only do the differential reverse so your egress is less and your recovery time in the reverse is also ten minutes. So. So you can come back home securely, safely and quickly. Yeah. And I do think you just you mentioned somewhere as we were talking agentless. Right. So you're not you're not having to get installed on each instance and get involved in the security layers and stuff within each instance you're running on the side which which is great. Your appliances, I understand, are also pretty secure. Yes. So all our appliances are CIF compliant images. So they are hardened images in that sense. Also like this. Like they are inside the clients environment, right? We are not a SaaS product. We are hosted product, right? And our appliances do not need public IP to function. So essentially nobody even will know that data motive exists. We will work on private IPS and there is an SSL, TLS communication through SSL that happens between the two nodes. All right. And then you do have a play in cyber resiliency as well. And backup and recovery. Because once you've made these replicates for business continuity or or migration you can I don't know what do you call it. Manufacture the backups that people really need without having to have them also run separate backups and move those those data sets. That is so correct, Mike. Or maybe not the roadmap that we have. Probably next quarter we will bring features where we will do backup, snapshot management and ransomware. And like you said, based on the policies, we will stitch the data that we have in our journals to create those backup copies and ransomware copies and immutable copies. Right. The key part here is that with the automotive, organizations will have to move their data only once, and out of that they will get a D.R.. They will get backup, and also they will get ransomware recovery. Yeah. I mean, if you're doing all those things, I think your TCO is going to go up even higher, right? So correct. Yes. Yeah. Yeah yeah. Well we don't we don't have a ton of a ton of time here because it'd be really great to maybe get a demo or, or look at the UI and the rest of it. But you've got Clis, you've got the UIs, you've got a lots of ways to do policies for different things, I understand, and some analytics it would be interesting to look at. But just just first, how does someone license data motive that's unique I think. And and if they want more information about it where would you send them? So first. From a licensing perspective, we have a very simple licensing process which license based on per instance. Right. And that's it. There is no cost for data change or data transfer or anything like that. So it's kind of leading to a flat opex, right? For more information they can go to our website w ww.automotive.io. Or they can even just drop us an email at info@automotive.io. All right. This has been really kind of an interesting look under the cover here, Samir. You know, and with VMware getting acquired by Broadcom, people are looking for plan B to VMware as a whole. They might want to be able to use tools like this to migrate. But even if they're sticking with VMware, they might not want to stick with the VMware to VMware replication that they have. They might be re-architecting into a better, larger hybrid cloud. They might want to take advantage of some service provider hosting as well, or maybe even repatriation. We don't know. And this is that kind of neutral, cost effective migration utility that they could they could really probably use. So hey you guys you guys heard it here. If you if you have VMware and you're looking for some alternatives check out data motive. Any last words Samir. No. I want to thank you, Mike, for for taking the time and giving us the opportunity and pleasure to be here. All right. Well, thank you very much. Check it out. Take care guys.
