Transcript
Hi Mike Matchett with Small World Big Data and we are here today talking with Storpool. We're going to talk a little bit about how you can address this challenge of of VMs and Broadcom and stuff when you have to provide rock solid 100% availability, disaster recovery kinds of situations, whether you're an enterprise with lots of robos or different different stacks of infrastructure or an MSP with lots of lots of clients scattered everywhere doing everything. How do you provide them D.R. natively and easily and efficiently. So we've got an answer for you coming up in a second. So hang on. Oh hey Alex, welcome to our show. How are you doing? Hi, Mike. Thank you so much. Doing great. How are you? Good, good. So it's been it's been a it's been a little bit since we've talked to Storpool. And you've got some great things that you want to show us today about what you've been up to. But first, let's just step back a little bit and tell folks what is store pool? Where do you fit in this grand scheme of storage? Things of course. So StorPool is a company that's been on the market for 14 years. At this point, we have over a million users storing data in our storage solution around the world. Uh, we are self-funded, uh, profitable and growing company. And, uh, where we stand is that we provide a primary storage solution for the benefit of, uh, service providers, SaaS, SaaS companies, enterprises, uh, and, uh, all kinds of, uh, other companies around the world. Yeah. So we're gonna we're gonna look at that and the architectures that you guys provide and encourage, starting with that sort of multi-tenant cloud perspective, really, I think, as we'll find out, helps even enterprises today and even medium enterprises, even smaller customers with their storage, because you start from that idea of multiple stacks, multiple locations of operations, multiple things, and then make it easy across that. So we're going to dive into that a little bit. Um, so this is uh, this is block storage and it's basically software defined, right? This is what the position is exactly. It's a software defined storage product. Uh, it can be used with KVM hypervisors, with VMware, with Hyper-V, uh, with, uh, storage protocols that are suitable for each of those, uh, in the KVM stack, we typically deploy with the block protocol, which involves, uh, our own client driver, uh, with VMware Hyper-V, we work with, uh, iSCSI. And more recently we have, uh, we introduced an NVMe implementation, uh, about two years back, uh, that that is an option as well. All right. So you just you mentioned the fact that you've got this Multi-protocol platform you can support multiple stacks, multiple different kinds of hypervisors in one solution, which is a benefit, to be sure, for a lot of people who are otherwise challenged to provide storage uniquely for each thing. Uh, what are what are some of the other problems? I mean, even specifically with VMware and Broadcom that people are looking at today, what are they running into on the on the storage side? Uh, one of the core problems that we've seen more recently, uh, not just on the storage side, uh, but uh, in the ecosystem as a whole is that you have out there in the world, uh, two main alternatives, uh, to the VMware Broadcom ecosystem. The first is, uh, the KVM based cloud alternative. And the second is the Kubevirt based cloud alternative. Uh, in our in our view, uh, KVM based clouds are the way to go. Uh, so we really focus on, uh, the problem that that customers have going to, uh, that companies have going to this, uh, KVM based cloud, uh, approach. And One of the key things that is missing there is the ability to do disaster recovery. Uh, so. So how do you, uh, protect, uh, large scale cloud infrastructure environment, uh, deployed on premises so that, uh, wherever, whatever you're doing in a primary site, uh, you can somehow recover into it in a recovery site or remote recovery site, uh, and continue working. Uh, if something goes wrong with your, uh, main site, whether it's a flood, a fire, uh, a mass power outage, any kind of disaster that might strike it. All right. So people are people are saying, hey, you know, we're no longer we're no longer, uh, able to afford those VMware licenses and whatever they're doing. Or maybe they don't even want us as a customer anymore. Uh, we're going to go do something else, like OpenStack or Portworx or, you know, or whatever else to set up our own KVM, you know, clusters in all these places. But D.R. becomes a problem for them and challenge. Right? Because you're otherwise now probably faced with writing a lot of scripts or doing a lot of things manually, you got these run books. Now for D.R.. People are going to have to do things under pressure and push a lot of buttons, and things can go wrong if you have people in that loop doing that. Uh, so, uh, so, so yeah, I believe that, I believe that, and I think also you mentioned MSPs who have multiple tenants and stuff. So this is not something that you want to be. If you've got a problem, you don't want to be doing this manually. You want this to be fairly automated. So how how how does the storage system how does Storpool help with D.R.. Uh, the the thing that we're doing is storage accelerated, uh, disaster recovery, uh, built in to the primary storage platform. Um, it helps you do efficiently things like creating and managing virtual machine recovery points with metadata included, uh, doing things like, uh, protecting those virtual machines in a geo redundant way. Uh, so sending, uh, virtual machine data to a recovery site, uh, remotely. Um, it would help with, uh, now we're able to help with, um, things like test, failover, uh, doing, uh, disaster recovery planning, uh, actually failing over in case it becomes necessary during a failback, uh, all these processes that are, uh, really difficult to automate, we've automated them, um, and we've made them simple. All right. So this is this is, I mean, reminiscent of, of some fairly advanced solutions where you are talking about this managing things at sort of the VM level a little bit. Right? I'm moving things not necessarily what we normally think of with block storage, which is, you know, take this petabyte of data and replicate it there, and that can bring up the whole environment or whatever's in that petabyte of data. Right. We're talking about managing managing it at a much more granular level that would people people want to do with D.R., which is basically down to the virtual machine. Right. Exactly. We're talking about granular control down to the VM level. Uh, you can have, uh, let's say that someone has, uh, we have several protection models. Uh, one is active passive, another is bidirectional. Uh, a third one is uh, many to one, uh, model. And the final one is multi-site mesh. Um, in the case of, uh, for example, the multi-site mesh, uh, you could have one user with ten virtual machines in a given location and, uh, they have, um, for example, uh, two recovery sites as options. Uh, they could send five virtual machines to one site, five virtual machines to another site. And, uh, in this way, their virtual machines are geo redundant. So they have control. The user has control over, uh, which sites, uh, their VMs are going to be protected into. Um, with this said, uh, we partner with, uh, service providers, uh, typically, uh, to make this possible. So the service provider is the one that provides the, uh, things like multi-tenancy. Uh, the, uh, DRaaS is built into their cloud portal. Uh, so we are not looking to, uh, we are looking to partner with with the customers we already have, uh, and we're not building out something, um, that is competing with them, right? Yeah. Yeah, yeah, it really it really augments what they're doing. Uh, and I guess, in a way allows them to leverage their staff far more efficiently, uh, by doing this, sort of in this. So, so is this is this a big implementation then to roll out this D.R., these D.R. engines and the D.R. layers that you guys are doing? Um, and does this require, you know, a whole nother, uh, implementation project for people? Uh, it definitely requires some work. Uh, because, uh, we are talking about something that brings a that US intelligence into the service provider's cloud. Uh, what we've done is automate all the grunt work. Uh, so the more most difficult parts of how do you make, uh, how do you get the VM metadata? How do you move it? Uh, create the VM recovery points, uh, and move them to a remote site. How do you, uh, make sure that the metadata is consistent in the remote recovery site? All these difficult questions have been solved. What's left is, uh, to integrate with the higher level API. Uh, because, uh, the disaster recovery engine provides a simple, central, persistent, uh, control mechanism, uh, with which one can interact to develop, uh, to and provide, uh, and manage, uh, the whole life cycle of a comprehensive disaster recovery solution. Yeah, I like that, I like that. So. So even an MSP can in an enterprise wants to be a service provider to their divisions and departments and users, too, right? There's a kind of whole point of. Yes, I always say service provider, but for me, service provider is both an external service provider and an internal. Internal. Server. Anyone who is, uh, handling the IT needs of a whole set of customers, whether they are internal to the organization or external. Yeah. So now you're so. So in order to augment that, you're actually building in this sort of core functionality that people use to sort of look to their VMware expensive VMware licensing and extensive, you know, the VMware D.R. solutions and things as they get off this and start to do more KVM. I mean, I think you even do some Kubernetes stuff too. But as you do this, this, uh, this KVM support that's native to different KVM stacks out there, uh, people can now, now get this enterprise DRaaS functionality or this, this, this cloud provider level functionality. Yeah, the KVM stacks, whether you're talking about, uh, whether you're talking about all the open source ones, uh, like OpenStack cloud stack Proxmox. Uh opennebula. Uh, or some kind of proprietary system that's built to manage KVM based clouds. Um, the pain, the pain there is really, um, noticeable. And, uh, that we want to, uh, we are seeking to solve it in the most efficient way possible. All right, all right. Uh, so just going to kind of wrap up here a little bit, and I just want to talk about, um, uh, you know, we talked a little about the efficiency. We talked about this. You know, I think the interesting thing here is as sort of an analyst looking at this is if someone has deployed a lot of KVM stacks for better, I don't want to say silos necessarily, but they've got a bunch of KVM deployments, different places. They're managing those, uh, maybe as independent entities for some sense, but they can now leverage a common storage layer and I'd say storage pool, right. A storage pool, right. Uh, kind of from a management perspective, uh, to get their D.R. services working to go from place to place to place so they can build that mesh kind of idea you had, which for an MSP might look a lot like availability, zoning and things like that, where they start to build out different regions or if they've got a lot of, uh, remote, uh, robo kinds of edge kinds of locations, they can create a disaster recovery site in the middle somewhere, maybe in the cloud or maybe in a hybrid location where all their different edges can recover into, depending on which one's having problems. Right? So it's like you don't have to have a recovery site specific for every, I would say edge, but every smaller data center, you can have one recovery site. So it looks a little bit more like n plus one in terms of what your investment is, rather than two n, which I think, you know, for those of you following that, that's a lot of money difference between n plus one and two n. So I like that the storage system is actually supporting that a lot. So um, anyway, if someone wants to look into more of this, uh, and find out what you guys really recommend and how you do this, what would you have them do? Uh, Alex. Um, I would, uh, say that they should check out, uh, our website. Uh starpulse.com. Uh, they can always get in touch on, uh, any of the social media sites if they want, with me. Uh, just look. Look for my profile and reach out. All right, so there you go. If you are in that pool of candidates out there who have been struggling with KVM replacement for your prior really expensive VMware licenses, or even just kicking the tires on it and say, like, how do we handle this? How do we manage this? Here's an enabler, right? You can get storpool, you can get the disaster recovery that's built in with it, uh, to help you deliver those like cloud service provider level of services. And if you are an MSP or cloud provider, you probably already know about this. So, uh, check out storpool. Uh, thank you so much for being here, Alex. Thank you. Mike. All right. Take care folks.
