During this podcast, VMware partners heard from Mike Wilkerson, VMware Cloud Specialist, who provided an overview of VMware Cloud on AWS.
Speaker 1: On behalf of VMware and Carahsoft, we would like to welcome you to today's podcast, creating new digital possibilities for your customers with VMware multi cloud solutions. Today's speaker is Mike Wilkerson.
Mike Wilkerson: Thanks for joining us here and taking some time out of your busy day to let us share a little bit about what we're doing with cloud services here at VMware, kind of want to set the stage for, you know, kind of what's going on in terms of driving the market, or driving the trends in government it towards cloud computing, they really, at the crux of it is kind of the expectations for the for what the technology that government organizations rely on to get business done are changing, due in a large part, to the emergence and rapid adoption that we're seeing of cloud computing. You know, most government organizations are looking to cloud infrastructure and services, to help modernize and really revitalize, you know, sort of your mission services and your citizens’ services. In addition, customers and moving to the cloud to take advantage of things that will help improve productivity, improve efficiency, quality of service, help retire technical debt, hopefully, shrink it complexity, lower overall operational costs, and deliver new digital IT services, among other benefits. You know, at the end of the day, really the one thing that's sort of the holy grail, that sort of the magic to the kingdom, or the key to the kingdom has to be the applications that it is providing, both to internal users, both to, you know, mission users and to in a lot of organizations, constituent services. And so the real trend has been towards application modernization. And the reason why is when you start to look into modern applications, it's really a combination of your existing applications, and new applications that you're building and rolling out. When we talk to customers, many of them are going through some sort of app rationalization exercise and process and deciding how they're going to transform their application roll call it landscape, whether it's to replace, form an existing application, or to, you know, fully modernize an existing application and leverage a new infrastructure model, either, you know, looking at refactoring and the micro services or containerizing them, or, you know, deploying a multi-tier hybrid type of application, or in some cases, building a completely net new cloud native app, or just after just replacing it with some sort of SaaS based platform. All of these decision points are things that around transformation that require some sort of no conscientious decision, based on, you know, choices around infrastructure and management choices. We see a lot of organizations are choosing a hybrid infrastructure approach to support the needs of applications. In fact, I read a recent study that said 93% of the organizations are committed to or at least expressing an interest in a long term hybrid cloud strategy as a way forward. And sort of like, you know, what does that mean? You know, early on hybrid, we really were limited to two choices, either on premise private cloud, or a public cloud. And a hybrid were sort of the commingling of the two and some sort of architecture that that worked and made sense. Not necessarily simple. But, you know, that's what we had. Well, now we've got a variety of choices. Now, with the advancement in cloud computing. The first being, and it's really, you know, hybrid is really about tapping into the right platform or combination of platforms, that makes the most sense to you, and your mission and your organization. You know, if we look at things from the left, like public cloud, public cloud gives you, you know, this unique, innovative, dense, differentiated set of services to build that new applications. Most all public clouds are available in some sort of on demand consumption model to allow for, you know, an easily an easy the flexible page ago, consumption based model. You also get the advantages of elastic capacity, rapid scalability. You know, in most cases, public clouds can be scaled, and deployed in a much faster way than, say, traditional local clouds. Then you have, you know, kind of, you know, you have a local cloud, which is essentially providing that same public cloud service, but you know, closer to an on prem type of capability. And really where you have that need is where you have applications that either due to some sort of geographical, like if you're GDPR, if you're in the UK, in Europe, or some sort of data proximity requirements, security requirements etc. The data can't be away from the computer. So you have to pull those resources locally. And now we've got this concept of an edge cloud. And really, that's another data proximity use case where, you know, users in a remote location, need to have the data close to the applications and close to the users, either due to latency concerns, in some cases, the ability or capability to work without a LAN connection, kind of remote office scenario. And then lastly, is this notion of the private cloud, which is pretty self-explanatory. Most, if not all, organizations have built a quite a substantial private cloud premise over the years. So existing hardware investments, they're really purpose built for a specific application. They're completely controlled, governed and managed, you know, by you as the IT organization. So, you know, of course, you know, the notion of hybrid and being able to select, what cloud model works for you. simple as it sounds, does bring its own level of complexity, the first being operationally consistencies, how things work, and how you manage and the skills and tools required to do things, you know, even your workflow engines and your workflow tools for things that are on premise and your private cloud. And one cloud may not necessarily work. You know, in a public cloud, or in an edge cloud. skill sets may be different, you know, I've seen, you know, an interesting report that said, you know, almost 90% of organizations are citing that the number one challenge with moving the cloud has to do with the skill set. Yep, in terms of what they're managing the data, what they have to manage, you know, on premise or in the cloud, a longer timeframe, to be able to move workloads from on prem to the cloud, different tools and management, how I managed and do things, what my workflow engine is, and what my governance is on premise, or in my private cloud. It's just historically varied substantially what I'm doing in my public cloud, I have to have multiple sets of tools, and I have different security models. And with that comes, you know, inconsistent SLA is in capabilities. As I said before, private cloud, it's much harder to scale up as rapidly, or to respond as rapidly without over provisioning and overspending. As it is, say, in the public cloud, when you can spin up, you know, a server instance in a matter of minutes to respond to, you know, outages or unexpected outages, or bursts in compute cycle requirements, you know, maybe a pandemic response. Now, quickly, can you do that, that seems to be you know, something that's near and dear to a lot of us here lately, over the course of the past year. And then lastly, the other challenge we have is, like I said, incompatible security and governance, you know, I've got, maybe I have, you know, more control over my on prem firewalls, and rules and things that I do, you know, from a micro segmentation perspective, in my private cloud, that's different, or a different set of tools that you use to, to do that in my public cloud, whether it be, you know, some sort of firewall services, whether it be edge protection, whether it be you know, penetration testing, antivirus, etc., I've got all these different tools that are doing the same thing, but I have to have them because of that, you know, operationally consistency that we talked about is the foundation of the, you know, a big gap between private cloud and public cloud, or different clouds. There's some good news, you know, it's not, it's not all bleak, and it's not all, you know, challenging, and, you know, things that are gonna keep us, you know, pulling our hair out. But, you know, from a, an IT perspective, and that is, you know, you look at something like VMware, we provide VMware cloud foundation. And what we've done is we've teamed with multiple different cloud vendors, to be able to deploy our V cloud foundation suite of tools, which is our vSphere, vCenter, NSX, and V San tools that run in the public cloud enclave. What that allows you to do is, and we'll go into this in a second. But the goal is to give you that same consistent look and feel, and tools and user experience that you have in your private cloud in a public cloud. And that could be, you know, again, whether today it's in a public cloud, where it's completely customer managed and owned customer does not you do all the break fix working with like someone like VMware, you do all the patching, the updating, etc. Then there's the VMware managed cloud, which is we're going to talk about this more in depth with our relationship with Amazon with VMware cloud on AWS. That's a fully managed service by VMware. And essentially, what you've done is you've taken that line of demarcation and moved it up the stack. And what I mean by that is everything that sort of from that hypervisor layer, and I've got a diagram to go over this and kind of explain it. But the concept is everything that's from the hypervisor layer, horizontally and vertically. Horizontally we take care of. So all the patching the upgrading the break fix of everything at the from the hypervisor layer down to the hardware, that's all handled by VMware, we don't know anything in the we do not touch and manage and have access to all the virtual machines themselves that are customer out. Then there's the partner managed VMware clouds, which is essentially, that same v cloud foundation suite of tools is just being provided by them apart from a different partner as part of a VMware cloud provider network. Think of someone like IBM SoftLayer, or, you know, Microsoft, Azure, with their ABS service, or Google, or Oracle. The neat thing about all this is, from a private to public cloud perspective, having that same common set of tools at the hypervisor layer, makes moving workloads around between clouds a lot easier than it's been in the past. And I'll go over that here in a second. So, you know, we can't talk about benefits without also talking about some unique challenges. And here's what we're seeing, as we talked with customers are some of the challenges that are within government it today. And it's not germane to government, this is true, whether you're in a lot of cases, whether your state, local, federal, even in the financial sector, there's just perception that the public sector space is in a technology lag, right, wrong or different behind the private sector, that there's too much of a reliance on traditional infrastructure, meaning there's been a big investment made into, you know, on-premise hardware and software that's being leveraged today, that there are, you know, one of the things that we're seeing as you definitely have is, there's definitely, you know, siloed and incompatible IT systems, meaning, you know, a lot of times a lot of the hardware and the software were sort of purpose built purpose bought, specific for, you know, a specific workload or workload type. So you don't have a lot of there's some incompatibilities as far as being able to, you know, repurpose things, from one platform to another. Another thing that we're seeing a lot more than we've seen in the past, has been some real scrutiny on any it spending, particularly in cloud, where customers and oversight agencies want to see what your real, what sort of ROI Are you getting on every dollar that you spend seeing a lot of that lately. And then the ever present challenge of security, compliance and resiliency, you know, we've got our recent node, it was a private sector problem, but it was a problem on the last, you know, when data is when systems get hacked, you know, how much havoc that can wreak, on a company on a country on an economy, if security is lagging over if there's a lapse and compliance or security. So, again, all things that are not necessarily unique to government, but our challenges within the government it space that we're seeing today. So with kind of all this as a backdrop, VMware, and Amazon got together a couple of years ago, probably back in 2017. And said, Look, VMware, you know, as to market leaders, you know, VMware, sort of being the de facto standard, or the market leader in private cloud enablement, through the vSphere, and V cloud foundation suite of tools, you know, you could argue that we have, you know, VMware had the lion's share of sort of the private cloud enablement, but we lacked that region to the public cloud space. Conversely, Amazon and AWS, the fact the definite world leader in the public cloud space, you know, around the globe, broad, rich set of tools, red set of applications, super flexible, and easy to consume, you know, cloud services, extremely scalable, but had challenges in reaching back into the private cloud space. So the two companies got together and said, What if we could do this? What if we could take our V cloud foundation suite, and allow customers to run that as a managed service offering, inside of, you know, an Amazon data center? And that's where VMware cloud and AWS got its start. So what are some of the advantages? You know, going back to some of the challenges. The first is this notion of seamless migration? Really, what that means is to move a workload in simplest terms from a VMware powered private cloud, or infrastructure on premise infrastructure, into VMC. on AWS, you're using the same exact tools that most VMware customers use today to move workloads around whether it's v motion, whether it's Site Recovery Manager, and that's to not only move workloads up to the cloud, but to bring them back. So you could have an application running in the cloud that's running in, you know, vSphere on-premise. Today running in the cloud and the amount of time it takes to do we do a V motion next, exercise across your connection? At the same time, bring it back. There's no need to have to rebuild, re architect retool, refactor anything in that workload. In some cases, customers don't have to you don't even have to read IP based on their networking schemes and architectures is still there. As a service, so as I said before, it's that same, you know, it's that entire v cloud foundation suite, hardware software, everything consumed as a single entity that's fully managed by VMware, no more hat, no more VMware administrators having to patch vSphere having to patch NSX having to do brake fix on a host. That's all taken care of as part of the service by VMware consistency, how you manage your V cloud, your instantiation of vSphere. And tools in the cloud, is exactly how you manage it on premise. It's v center, okay, there's nothing magic to it. It's simply vCenter. So it's the same tools, it's the same capabilities, it's the same, if you're leveraging some sort of workflow engine to do orchestration, like a ServiceNow, or like a V realize, the VMC on AWS instance disappears as other end point in that overall topology. From a security perspective. The VMware cloud on AWS and Gov cloud is in its final, we're in the final stages, literally in the final stages of achieving our FedRAMP high CTO, we have our comments back from the PMO. We have our kickoff call scheduled for next week. And the comments backwards, extremely low for a high offering. So our expectation is to be published in the marketplace by the end of this month, that's June. And then lastly, modern application support the platform itself was what I like to call sort of future proof. What I mean by that is, along with running traditional legacy virtual machine workloads, you can also run, take advantage of Kubernetes or run some sort of micro services type of workloads on the same platform and coexist on the same platform as traditional virtual machines. So it supports both. Here's really more of an overview and topology of what the service looks like VMC on AWS is consumed and purchased by the host. And what that mean, and what that includes is, it's the vCenter, vSphere, vCenter, NSX, and all the underlying hardware and networking plumbing that comes along with it. Again, it's writing on top of AWS bare metal, we're running we're running on AI three and AI three in AI three een AWS hardware. It's not nested on top of EC two, it's straight on the bare metal. The only thing we don't have access to by design or the virtual machines themselves that are customer managed and maintained. Everything from the above the hypervisor is customer owned. We provide a capability that we call hybrid link mode, which allows you to link your on-premise v center into your cloud instantiated version of V center to give you that single pane of glass management that allows you to do your emotions that allows you to do everything that you do on premise and vCenter. You'd also do it in the cloud that appears as one pane of glass. Like I said, if you're leveraging some sort of third party tool like ServiceNow, or like a V realize type of tool for workflow orchestration. Essentially, VMware cloud on AWS, you're familiar with Amazon's terminology, we're just a VPC inside of AWS data center. So BMC would appear as another endpoint in your workflow engines. Another thing that's unique about our relationship with AWS is if you have existing AWS services that you're using today, like EC to like lambda like RDS, etc., any of the services that they have in their catalog, via NSX, we provide a 25 gig high speed low latency link that's depicted by this arrow here. What that means is that gives you access to all your native AWS services that you're running. So what we see a lot of customers do is they've got some things running in Amazon back here. Let's say they're running databases on RDS, they'll move their on prem, tier one or layer one application servers up into VMC and link it across this high speed link. One big advantage here across this interface is as long as you're in the same availability zone in the same region, any data that translates transmitted across this high speed link, there's no e gress cost for that. The only time you would incur egress would be anything that comes from the VMC on AWS VPC, out to on premise or after the internet. So you can imagine if you have a two tier model today, where you're doing a lot of transactions between, you know, on premise and into, like an Amazon data center, by moving those workloads into VMC. You cut out that egress cost. We have a we have it's fully supported and fully staffed by VMware, we have a support center out in Colorado, and a team here in Virginia, that's us citizen US soil that's, that provides a break fix support. It's available in multiple flavors, one being on demand, so it's essentially a pay by the hour. The other is you can reserve hosts for either a one year or three year term meaning you know, the As hosts, and that's all the hardware and the software included, is licensed to you for a period of time for either a year or three year period. As I mentioned, we're FedRAMP in process, we should have that by the end of this June. And we also touched on, you know, having direct access into those back in native AWS services. This truly is what differentiates differentiates the VMC offering from other VMware powered clouds is the fact that it's the only one that's fully supported and managed by VMware. And having this direct access to native AWS native services is a tremendous asset. So let's quickly look at some use cases that we're seeing in the government space. And these are not in any kind of numbered order is going from left to right. The first is app modernization, we do have a lot of customers that are looking to you know, instead of just moving to the cloud, they want to fully rewrite applications, and containerize them or, you know, leverage microservices or leverage Kubernetes. Again, they can live coexist and run on the platform, cloud migration. What we're seeing here is we've had, we've had several customers that have shut down data centers, not all of them, but they've shut down like one or two remote data centers. And they're moving all those workloads into VMC on AWS, as their second site, some call it data center evacuation, we have a tool we call hybrid cloud extension, hc x, which allows us to do essentially think of multi-threaded parallel v motions to do to do data center moves. In fact, we have one SLED customer who evacuated an entire data center over a weekend. That was I think it was over 900 to 1000, virtual machines that were all moved out and evacuated into BMC, over a weekend or two day period. The next use case is data center extension. And this is sort of that bursting capability, where you have an on your on premise data center, and you want to say you need some extra compute networking and storage cycles, maybe for just a short period of time, maybe, maybe you work for an organization and Treasury who gets really busy in April around the 15th. And there's a lot of process that has to go on. Once instead of you know, designing and building and overprovisioning on-premise assets or capital assets before you know worst case scenario, you know, you can scale down a little bit, you know, build your build your own framework, build your on-premise infrastructure for your steady state and then have a plan and be able to leverage the cloud, this option, you have to burst out because you burst out use it for when you need it, but only pay for what you use. You're not over provisioning assets, they're just kind of sitting there underutilized. We also have customers that are using it as an extension of their on prem data center. Just to test Dev, we have one customer who's using the cloud, to do all their development and QA work. They're doing all the dev work in the cloud. And they're promoting some of the back on prem for production. Some they're leaving it out there for production, but they're using it for test Dev, we have some customers using it for labs, we have some customers using it for like training environments, they have to quickly spin them up for like 30 days and then take them down. And they're able to do that. The fourth use case is disaster recovery. Whether that is providing Dr. for net new applications, whether that's some customers like looking at the cloud, looking at VMC to augment existing disaster, best disaster recovery. Some of them are using it to completely retire their existing Dr. and using the cloud as a DR site. It's really up to kind of, you know what the applications the mission and kind of what your overall strategy is, but you can you know, we have users using it for multiple different things. We do have a managed Dr. As a service component in our commercial region where it's hold the component itself is fully managed, that will be making its way in the Gov cloud probably sometime in 2023. Because it is a shared service, and it's Amanda, it's a SAS offerings is going to have to go through the FedRAMP certification. And then lastly, something that's probably been the most popular use case, for the past 12 months has been running virtual desktops in the cloud. And that's customers that either want to fully retire their own Prem VDI farms and move them to the cloud. They want to build a certain number in the cloud to be able to burst out to when their on prem allocation when they're on prem resources start to get taxed. Or they want to have a balance of both to be able to load balance across both an on prem VDI form and a farm that's in the cloud. And I want to take that lead into an actual real customer that we have. I won't name the agency. But this organization in reached out to us in March of 2020. Right when the pandemic started, things hadn't been closed yet. But you know, they were rapidly heading that way. They had a large number of users that were spread out all over the country that we're going to be you know, that we're going to be got to come into an office and had to get worked on. And we're looking for a way to they couldn't they already had VMware VDI platform in their data center. But given the circumstances, they weren't going to be able to procure hardware fast enough, let alone get people in the building to be able to provision it, the rack and stack it and get it turned up and get it configured fast enough to be able to support the remote users that were most likely going to be remote within two weeks. So they came to us with an kind of with this question of, can we run virtual desktops in the cloud. So we did a quick proof of concept with them. Everything worked fine. And then this customer went from zero presence in the cloud, to almost 2000 desktops within a matter of three weeks. And during the entire time, they were out and weren't able to come into the office, they were able to spread their spread their user workload and ship the workload across, if users started to log in, they would, they would first hit their on prem data center first. And as that started to scale up and reach a certain threshold, their load balancers will start to then provision users in the cloud. And it's funny, it got to the point where some of the users got savvy enough to know that they were being run out of the cloud. In their performance, it's actually better. So they were asking for that on a full time basis versus, you know, kind of luck of the draw when they logged in. But this was, you know, this was this was a really good, really good success story. And it was really good to be able to support a customer who had a real mission need and a real crisis, to be able to help them solve a problem be up and running, you know. And I got to thank, you know, in the years, I'm dealing with government, going from zero to 2000 users in about two and a half weeks. That's got to be some sort of record that we sat there. But it's a great story. From a partner perspective. And from a procurement perspective, we work primarily with our friends at Carahsoft. We also work with a lot of the same reseller partners that you all work with on a daily basis, we're on set most grown a lot of contracts, you know, GSA on schedule, 70, sube, other contracts that will, and that's, that's growing, you know, by the week by the month, especially as we now get close to our FedRAMP CTO. In addition, this is another list of keeps growing exponentially on a monthly basis, number of third party vendors, third party technology solutions that are integrating directly with BMC, whether it's writing directly to it or integrating it through the API set, you know, around all different segments of technology. One more use case from one customer was talking about. This is another customer in the defense space, they were doing some security work. And we're looking at looking for ways to and they couldn't obviously get equipment brought into the data center and stood up last year. So they looked at the cloud as a place to be able to provision a couple of hosts. You know, once they procure, they were provisioning time to build a nest to build a software defined data center. You know, in the cloud. A typical configuration is the base 303 hosts that requires a minimum takes about two hours from start to finish, we're able to do that to them. And they started using it for using that lab for remote users to be able to come in and their testers to be able to go in and test new code. A lot of it had to do with security, patching and cyber tools. But they were able to meet that meet that objective not have any downtime, and not have any losses and productivity. This by leveraging the cloud to be able to quickly spin up and deploy environments for them to do their testing and QA work on. Okay, I want to wrap up here real quick with just a couple of items around just to kind of summarize what we talked about and give some data points around it. Some of the key benefits of VMC on AWS. As I said before, it's really the best of both worlds, you get the rich, robust tools of VMware Software Defined data center with V cloud foundation paired with Amazon's back end tool set of all their applications and their global footprint reach within the AWS architecture. Zero refactoring, the time and cost from migrating from on prem to the cloud is greatly reduced, sometimes years, months to even days. Could you ever have to rebuild or redo anything, or leveraging existing skills? As I said before, no need to have to relearn anything, it's the same look and feel is what you're doing today, you're doing the same thing and BMC. As I said, the only difference is where the data lies. Lastly, is around cloud economics. This was an IDC study that they did. But they interviewed, you know, several customers that were BMC customers, and did some drill down into kind of what they were seeing from an ROI perspective. They were seeing about a 351% ROI over three years with a breakeven point in eight months. And what that means that eight months means is obviously 46% faster migrations and 57% lower overall cost of migration. What that means is you see this thing called on this migration bubble here, that's something that everybody who is moving to any cloud of any type, you have to account for this migration cost. But what that means is it's that period of time by which you're essentially paying in the both buckets, because you still got to maintain either support or SNS on your existing on-premise footprint. While you're migrating to the cloud, we've also got to pay for that cloud footprint, the further to the left, that you can pull this migration bubble in, the faster the benefit realization comes in. What that means is, we have some ETS estimates that it would take about seven and a half years to fully migrate, you know, 100 applications based on some industry averages. And that's because you've got, you know, a lot of the reengineering, it has to take place, you can cut that down, the further you pull that to the left, the lower that migration cost bubble becomes. The other thing that you take advantage that we're seeing customers have the benefit up to is what some of the hardware that we have, we don't limit you to how many virtual machines, how many VMs, you can run in BMC. It's completely governed by the nature of the app, the workloads themselves, what your SLA is, are etc. We don't limit that. And the more density you get, obviously, the better the investment becomes over time. So we've talked about harbor cloud, we've gone over some of the benefits, we talked about being without an AWS. The one question we get all the time from customers, if this was great, how do we get there? What's step one? Obviously, number one is understanding. App rationalization is probably the number one thing understand what applications you have, what can go where what can be privatized, what can go in the public cloud, etc. You know, what can what's better suited for edge cloud, what's better suited for private cloud? Secondly, is doing some sort of analysis. We have a cloud economics team that can put together a specific protocols to Boehner a three year ROI, and T CIO for you based on your actual parameters, and your environment. And then lastly, we can share some of these out is we have some hands on labs are available for their essentially the VMware labs that we that we offer, they're free to use for customers and partners to go in and start, you know, playing with the tools and getting their hands on. This is what it looks like. In additionally, we have a whole fleet of professional services people here at VMware, that can help with some of the work help with some of the assessments to kind of answer these four big questions. Am I ready to go to cloud? You know, them help me with the rationalization help me do a cloud readiness assessment? What does my networking need to look like? You know, what can I move what fits best? Where what should my overall picture look like? What should my roadmap be? So journey, it's not a rip the band aid off and just go? How do we get there over time? What am I move? First, let's look at all of my application dependencies. Let's look at you know, my portfolio assessment. Let's see what I have. Again, where do I go? What can I go for cloud native? What needs to be migrated in place? What can I do a lift and shift? Then lastly, you know, how do I execute any education services around you know, taking on something new, there's a skill set that's that you don't have. We've got education courses on that around specific cloud technologies like around hybrid cloud extension for an example, or maybe NSX in the cloud, those sorts of things. That's all I had for today.
Speaker 1: Thanks for listening. If you'd like more information on how Carahsoft or VMware can assist your organization, please visit www.carahsoft.com or email us at vmware@carahsoft.com. Thanks again for listening and have a great day.