VMware Cloud Foundation’s Shift to Self-Service Private Cloud Consumption

Hello, everyone. My name is Vincent Ricchio and I'm a technical marketing, uh, manager here at, uh, Broadcom. And, uh, today what I'm gonna focus on is, uh, what are we doing with VCF nine and how to consume the cloud.

So basically what I'm gonna focus on really is the automation of that private cloud investment that you made. You know, how do you deploy your applications and services? How do you, um, you know, like set up tenancy or different organizations that have different needs, all that kind of stuff.

And then how do you consume that ias, right? That, or how do you use IAS on top of this cloud? Um, as well as maybe some other ways that we can consume, such as, you know, anything as a service and so forth.

So, really glad to be here with you today. I've always decided to do these, and, um, I'm really looking forward to showing you kind of where we're going, what we've done recently with the product and automation and, uh, and just kind of the overall strategy, uh, that we have to help you really deliver, uh, a private cloud to your organization. So, I do wanna go over a couple of things first before I get into the, the meat of the slides.

And I also want to, uh, I'm also gonna do a demo. So I'm gonna go through, uh, just a handful of slides here, and then I'll show you a little bit about, show you in the product and, and, and emphasize some of the things I'm gonna be talking about. Um, but, but this slide is really just kind of an intro slide that I wanted to bring up and a sense of what are some of the newer things that we're kind of doing, or what are some of the areas that we're really kind of focusing on when you, when you think about consuming the private cloud, self-service, um, and, and automation in general.

Um, and part of that is, you know, if we look at the bottom down here, we can see that there are services, right? Those SUDC services, things that we're familiar with, vSphere, uh, VNSX, right? Others, we we're gonna talk about some other services as we go along as well.

Um, but, but the, but the reality is that we're still continuing to do that, still continuing to consume that. But on top of that, what we're doing is we're introducing some new, uh, new experience in terms of automation, in terms of, uh, consumption. 2, and eight point 18, there are still a path for customers to go down where they can continue that experience.

And we call that organization type VM Apps organization. There's two types of organizations in the product that we're gonna talk about, basically, tenant types, if you will. One of them is the VM apps, where customers that are existing customers will continue to have that same experience that they have today when they upgrade.

There'll be some consolidation of UL and some minor differences, but functionality will all be there. But what I'm gonna talk about today really, is the new experience. Um, we have a new Oracle called an all apps org.

This is gonna be a new to v VCF nine. And, uh, essentially it's very focused on the supervisor and deploying Kubernetes custom Kubernetes objects, um, onto a supervisor cluster or, uh, a supervisor itself, not necessarily supervisor cluster, sorry. But, um, so, so the idea is you could have multiple clusters in a supervisor, and then you can, uh, start to deploy to that.

So we'll get into that. So when I talk about all these pieces, some of them will bleed over into the other types of org, but I'm gonna on the new experience, which is the Kubernetes, uh, uh, experience for us. So when we talk about that, one of the main things here is I wanna talk about is the services.

So you're gonna see this a lot coming up from our, from not only our marketing, but other things in the product is VCF services and partner services. So what we mean by that is, um, and you're gonna see it in the demo today, I'm gonna actually deploy using the VM service and vks service. So these services are bubbled up into automation from vSphere.

Um, and essentially what we can do is we can consume them and start to deploy our applications and services. And those services could be, like I mentioned, the VM service, vks service, be like a network service, a storage service. There's other extendable, extensive extended services like, uh, Argo, cd, um, uh, uh, Valero Harbor and so forth for like image registry, cd, continuous delivery, um, things for service meh, things for databases like DSM for databases of service.

So all these services bubble up and are easy to consume. And then partner services allow folks to build out some new stuff. We are actually doing some stuff with a couple of vendors now.

Um, so be on the lookout for that. I'll talk, uh, a little bit about those as we go along. Um, in fact, I have a slide on it and then blueprints.

Okay? So, um, we're continuing on with, with our blueprints. The, the main difference here with the blueprints now is that, uh, in this new experience, it'll be focused on custom resource definition objects, uh, that are based on Kubernetes.

So essentially, we're gonna take something like, let's say a VM operator, uh, which is a custom resource definition of Kubernetes, and we'll deploy a VM using that. In fact, we'll show you that in the demo so you can deploy your VMs right alongside, uh, your, your Kubernetes clusters as well as your, your containers. Uh, the other thing is workload operations, visibility.

Um, you're gonna notice a lot more of that in the product. Now, you know, as we've gone to BCF nine, we've, we've really spent a lot of time and effort in merging the products together and, and causing everything to act, you know, act like a single product as much as we can. And part of that is sort of that workload operations visibility coming from the operations metric metrics and so forth.

So, um, you'll see that a little bit, hopefully I have time to demo that at the end as well. And then one of the bigger new things is the tenancy and project management. Uh, right.

So with the tenancy and project management, what we're doing essentially is introducing a new kind of construct that lets us sort of have this tenant methodology, uh, to, to create something that we call organization. So in the product, they're actually called organizations, but you can treat them like tenants because we can do network isolation, we can do, um, various types of, uh, uh, you know, user isol, user isolation stores, like all kinds of stuff in there. And, and we'll talk a little about that.

And then project management as well to further, uh, isolate, you know, lines of businesses and, uh, and continue on that path of, of, of sort of that tenant methodology, governance and policy still in there. Approvals, uh, lease times, um, day two action policies. One thing we did introduce new, and I don't have a slide on it here, but one thing we did introduce new, uh, in bcf nine automation is, uh, I as policy engine, and that's actually a, a policy as code engine where you can actually use, uh, uh, validation emission control type stuff.

And Kubernetes, like that form, I think they use a CEL language, common expression language to actually build policies. So you could say, you know, VMs that get deployed have to have a label associated, or, you know, a development, uh, project, can't actually deploy or namespace or project can't deploy, uh, you know, development clusters with more than one node and stuff like that. So there's a number of kind of policies, and they can modify those in the code, uh, right there on the product.

So, really interesting. The other big area that we're focused on is content management. Uh, so content management, you're gonna see this in the demo.

And essentially what this is a, this is a really nice way to, uh, distribute content to these different organizations. And there's two content libraries. There's one on what we call provider portal, which manages all your organizations.

So kinda like your service provider type of person or enterprise IT admin that might be going in and managing all these organizations from the top level. Uh, they, they can actually distribute content, their content library, and then each organization can have their own content libraries where they have their own specific content. And we'll take a look at that a bit as well.

And then continuing down the, the event orchestration and extensibility capabilities with Orchestrator, um, and some other capabilities, uh, there as well. And then notice on the left hand side, there are personas, uh, and, and one of my slides is focus on that. Um, and also when we get into the product, there's different kind of, uh, you know, experiences that different personas may have.

So for instance, a user will have a different experience than like an advanced user, which will have a different experience than like, let's say the Orga or the enterprise ITM and, and certain in terms of things that they can do. So one of them is the end user, right? So I'm gonna, I just wanna preface this a bit because as we get into this new experience, uh, different users are gonna have sort of different use cases, right?

And if we start from the left, it's, that's really where we're trying to get to. It's getting these end users and developers a way to self-service the cloud through a catalog or UI and CLI, uh, all available, uh, get those day two operations in their own deployments, et cetera. Now, something that's new here is the organization admin, and then the provider or enterprise IT admin, the organization admin is responsible for the tenant organization once it gets created.

So they'll, they will go in and create the policies, they'll go in and set up the projects, and then they'll add the users to the organization. That'll ultimately be those end users on the left. Now, the provider or the enterprise IT in all the way the right is gonna be mainly responsible for, for, uh, accruing the resources out of vSphere that are needed for these organizations.

So they will provide quotas and resources through the supervisor zones and stuff inside of vSphere to actually say, okay, here's how much CPU, memory and storage you can consume for each organization and so and so forth. Okay, now I wanna just bring up this, I know it's a little bit of an eye chart, but it's a little important to kind of understand how we're really architecting this new solution, uh, in terms of especially consumption and ultimately getting to where folks can start to deploy BMS and V case clusters and then their applications. Now, I'm, I'm bringing up the supervisor, uh, pieces at the bottom here for a reason, reason, because without that, you don't have an all apps experience.

The new experience doesn't work. You have to have that supervisor already created inside of vCenter. And so once you have that, it's gonna show up in VCF automation.

It'll just show up there. And then what you can do is you can start to build out regions, and regions are one or more supervisors, and then you carve out the resources from those zones underneath it. So what you're gonna do is say, maybe region US West has access to X amount of compute.

US East has X amount of compute, and those regions span clusters, they span B centers. So it really gives you an abstraction across your VCF fleet to consume resources and deploy onto. Now, above that, you can have an organization consuming all of that, or just pieces of it.

You can have an organization with one or more regions, and essentially, uh, then the organization can, can consume, um, or deploy onto those, uh, using the, the custom deep resource definitions for Kubernetes. Now, one space above it, which we'll get to in the demo as well, is the concept of projects. So each project lab, like users and namespace associated with them, and then those namespace are really ultimately the endpoint or where the user is going to target when they deploy.

So if I'm gonna deploy a virtual machine, I'm gonna pick a namespace and I'm gonna deploy into that namespace. And then what I'm gonna do is I'm gonna be able to go to that namespace using the UI or CLI, and then see that VM and then edit it, make my iterative, you know, changes and stuff like that if I need to in the code. Okay?

And you'll notice a couple different options at the top. There's some where you see a VM next to a, uh, a, a cluster, some that have the dotted line, that basically what you can do is you can create blueprints or even, uh, various other ways, but you, you can't create a blueprint, let's say, that can tie your VM to the cluster and you have one single app, uh, and so forth. Uh, so really nice there.

So I wanna just kind of bring that up a bit. I know, uh, I don't wanna spend too much, too, too much time on it. Um, then, uh, real quick, just going back to the services, uh, the services are essentially, uh, something new, like I mentioned earlier, but I wanted to bring this up a little bit because one thing that we really wanna start, what we're thinking about in, in, in a broad direction is, hey, this is more than just, you know, let's say deploying a VM or deploying a cluster.

We really wanna make consumption easy for, for our customers. And the one way that we can do that is to bubble up services like database services, load balancer services, um, you know, uh, networking services and stuff like that that customers can consume. So on the left hand side and the upper left, you'll see just sort of, kind of the outta the box standard services.

And we're gonna look at two of those and maybe, yeah, just two of those today, which will be the VC and VM service. Now, when I say those services, what I'm meaning is, okay, if you take a traditional Kubernetes, API, we, we prayed a bit, like I mentioned earlier, custom resource definition around it. And then the demo, you'll see when I deploy a vm, it's actually the API version for that Kubernetes manifest is called the VM operator.

It's the API version. So it's custom VM custom, uh, Kubernetes clusters. Uh, and then there's some other services which will actually deploy in the demo too, like a load balancer service, network service.

So what's kind of cool about this is you can expose, you know, your application as a load BA using a load balancer. This can use NSX or abi, um, you know, depending upon how you created or set up your supervisor. Uh, there's also some additional services on the right, uh, that we see.

So we'll see, like Harbor for Image Registry, there's some things for, uh, service mesh, et cetera, Argo CD for continuous delivery, uh, and, uh, and so forth. And then of course, AI services. So we're continuing to, to support, um, our Nvidia ai, uh, GPUs, uh, services and stuff like that.

So a lot, lot of talk about Kubernetes, they're managing within, uh, VCF automation. I work with a number of customers who are at various levels of maturity when it comes to Kubernetes adoption. Some of them may already have a platform that they've, uh, selected.

And, uh, I'm wondering, does VCF automation have the ability to manage other, any other Kubernetes platforms that are, you know, upstream compliant via API? Uh, if, if a customer needs to kind of do a crawl, walk, run, in terms of adopting VCF features or, uh, is, uh, Kubernetes management only available for customers using VKS for their Kubernetes runtime? Yeah, yeah, great question.

And so as far as VCF automation, it's only VKS. Uh, if they have just a standard, you know, just download Kubernetes, uh, version out of the box, they won't really be able to manage it, uh, with the solution. Uh, now they could use some of the built-in scripting tools to do some stuff if they need to integrate with it.

Uh, but it would be custom, custom stuff, um, at that point, uh, out of the box. And what we, what we, what we integrated would directly would be the VKS, uh, uh, cluster or the supervisor on top of each here. So much like the, uh, RA automation of old external integrations are very much possible.

It's very extensible, but it's very much a DIY situation when you talk, start talking outside of the, Yeah, and, and this would probably be even a little more DIY, um, just, just because we are, we are a little more focused now on the, on the supervisor itself. Yeah, a great question though. But yeah, that would be, that would be a, a custom, uh, kind of thing, uh, if they were to go down that route.

But, um, you know, right outta the box, even on the existing, uh, uh, organization, they won't have that ability just yet. Or at this time With looking at those additional services as that essentially is a bunch of helm charts that I'm, I can then deploy into my environment. Is that a good way of thinking about it?

Or Is that even know? Yeah, that's a good way to kind of think about it, but that's not gonna be the process here. Um, there, there is a way to upload some files from the vendors, uh, if you, if you wanted to do that, if we don't have something outta the box.

Um, but generally most of them are going to be appliances, um, or services that will just get installed onto the supervisor. Uh, so the, the supervisor will either come with it, like Valero, there's a couple other ones that just kind, kind of get installed. Some of them will be additional appliances that get installed.

And then some of them will just be services that you enable on the supervisor itself using a YAML and, and, uh, uh, what they call kind of like a response file or something like that. So it's, it's not specifically like a helm chart specifically, but the, the idea is, is somewhat similar, I guess. Good question.

Um, I see that, um, you introduced the concept of a, of a project. So, um, can you specify better if the project is, uh, a simple, um, resources deployment or environment plus resources deployment, or application plus, uh, environments lifecycle and assist deployments? So, Yeah, good question.

So it's really kind of all, it's a ladder really. Um, so your project is going to be like, let's say the users, uh, the namespace they're gonna deploy to. And then once you create, like, let's say a blueprint, you will tie that to a project or one or more projects that you can share them as well.

Uh, but you could just say, Hey, this is tied to a project. And so everything that kind of then, uh, gets deployed and all of that will be revolving around the project. So once they deploy it, that deployment will be associated with the project.

Um, and then when they take their actions, everything will be kind of in that, in that project from, from just a sort of a, a, like a hierarchical view. Yeah. If that, that helps.

That makes sense. Yeah. Yeah.

When you create your project and then ultimately create your namespace, you're gonna, you're gonna choose, uh, some stuff about, you know, how large the VMs will be and, and you'll be able to choose like, some constraints around sizing and VM plus stuff and all that. And you'll see that in the demo too. So hopefully that'll, will that be bit, uh, but did that answer your question?

That, that kind of helps a little bit. Alright, well, let's go ahead and get started. So I'm gonna jump into a, an, an organization called ACME Sales, and I'm gonna, I'm gonna log in here as the organization admin.

Now for the sake of time, I'm gonna skip the provider admin experience. Um, but just know that the provider admin, what they did was they set up the networking, uh, and the region quota and the regions to consume stuff outta the supervisor for this particular organization. So I'm gonna log in as the organization administrator when I log in.

Um, I'm gonna basically just see like, uh, how many projects do you have with, have none right now, right? And just get a little bit of a overview. But what I'm gonna do is let's go ahead and just start building this out.

Uh, so a couple of things that we can do here when we first log into the organization, admin, I'm probably wind up gonna, gonna wind up realistically going to administer and doing my LDAP and bringing in my users and stuff. But what I can start doing is creating things like content. Um, so remember earlier before I mentioned that we had these content libraries.

One is in the provider, one is in the organization, one in the organization. I can create a content library, and this content library can be used, um, you know, for, for folks that wanna deploy VMs and stuff, this is where the, the VM images will come from. We can subscribe to an external content library if you want to, and that is actually a setting that the provider can do.

So if you don't want your organization to be able to do that, you can restrict that. Uh, but essentially we're just gonna connect to this content library. We're gonna pick a region.

So I could pick one or more regions where there's storage classes here for this content. So essentially the storage class is gonna say, how much con consumption, uh, can I, or how much storage can I consume? I can add multiple regions here if I want to.

Um, and then I'll hit confirm. Now, what this is gonna do is this is gonna gimme the ability to publish content to these organizations. Again, I could do this from the provider as well, um, but I could go ahead and do it, uh, from here.

Uh, now the other thing I could do then is go to my images and I can see all my images are here. Uh, so those images could be used, uh, when I go ahead and do all my deployments and stuff like that. Now, before I create my project, okay?

I want to define a couple things because I've not done anything in this environment yet, really, except probably my LD a stuff, uh, in my content library. Now I need to go ahead and start building some things out, making sure that's really customized the way I want it. One thing that we can do here is we can create namespace classes.

So as we get into this, you're still to see more Kubernetes type type methodology, uh, and, and, and terminology as well. So a namespace class are essentially templates that will be used inside of the namespace. But these are kind of unique for us because not only can I go in and just say, Hey, we're gonna create a namespace class with certain limit, memory limit, say, you know, you can only consume so many resources out that region.

But we can also add VM classes, which can have different capabilities. So for instance, I may just have one VM class inside of vSphere that is GPU enabled or has other capabilities that you can kind of check box when you create these VM classes inside of the system. And there may have capabilities that I want for that namespace.

Um, and usually like, kind of the example we normally give is, you know, GPU enable, like if they're gonna do something with Nvidia, right? Or some kind of GPU thing, you picker storage class, you can have more than one. This will just be generally what, how much con some, how much space am I gonna give this namespace out of the storage, uh, what are they gonna actually be able to do?

So that'll be my default namespace. There's a few out the box, but I'm gonna just do default here. Um, and then what we can do is go up to projects.

So you just asked about projects. So here's where we're gonna create a project. So before we start adding users into the system, I need to go ahead and create a project.

So I'm gonna go ahead and do that called customer portal. I'm gonna add my users. Now I notice there's different types of, uh, user types here.

I'll just add Connie. And, and Connie will be a project administrator, meaning Connie can add users to this project. And then we can also add other types of users, like project user, advanced user, advanced user just sees more things to deploy from, like what we call services ui, which I'll show you here in a second versus just the catalog, okay?

Because there's also a catalog, okay? Now, once I've got the project created and I've got my users in there, I can start creating namespaces. So lemme kind of do this, I'm gonna do one real quick, and then I've got another one that gets created.

So I'm gonna create this development namespace, and then essentially I'm going to, uh, uh, go in here and just create, pick that default namespace class that I just created. So what that means is that any of the VM classes that I chose when I, when I picked that namespace class just now, that will show up when folks go to deploy to this namespace. So it, like it said, it could be a custom VM class, it could be, you know, something with some properties in it.

And so that way I can really, you know, get very granular on what they can do here. Then that region, like I mentioned that abstraction across my VCF fleet. So they're gonna be able to, to deploy to all kinds of clusters here.

Uh, and then, uh, and then the VPC. Now, I really didn't talk very much about the VPC. We've had some integrations with the VPC historically, but this is a, a really new in the product in terms of how much we're, uh, using it now.

So this is gonna create a lot of isolation from the networking standpoint because this VPC, uh, we can actually connect this, uh, this namespace, this, this particular organization or project to, uh, transit gateways that will go up to maybe like a dedicated provider gateway, which is de, you know, can be connected to a T zero and nine x. So we can really get real network isolation between these organizations. Um, and then we just pick a zone, um, that we wanna consume from and so forth.

All right, lemme just do that again. 'cause this one just does two of 'em. Uh, we'll do a production one as well, because you can have multiple name spaces and each one could go to a different D VPC if we have multiple VPCs.

Um, there's also a little bit more integration with the NSX with automation. Now, in fact, when you create one of these, uh, when you create an organization inside of automation, it creates an NSX project. So you can go to that project and see all the VPCs that the organization is consuming, uh, from an NSX standpoint.

Like you can go and look at the topology, you know, change your connectivity profiles, and there's some things you can do in here, um, as well around that. Okay? Then we can get some information here.

So let me just kinda show you here, we can go into, uh, uh, the services, and then once we have the services here, we can see that, you know, uh, this namespace is available. So we just created that namespace. And so now what we can do is we can go in, um, as, uh, the user and then start to consume some of this stuff.

So let me just kind of get over there real quick. Okay. Now what I'm gonna do is I'm gonna log, in fact, me sales as a user.

So I'm gonna log in, is Mary, who is going to be an advanced user. And so the advanced user, remember I, me, I mentioned users, advanced users a little bit earlier, the advanced user will see the services and the catalog, uh, capabilities, uh, in here. So when we go to the services, we can see that development, uh, namespace that we created.

I can see all the services available to me, virtual machine, Kubernetes. So let's go ahead and deploy a virtual machine. So when I deploy a virtual machine, again, I'm using that VM operator to deploy this.

I can do this through the CLI as well, using it a Kubernetes manifest and the cube CTL command. Um, or I can also do this through a, through a blueprint. But if I want to go in and say, Hey, look, I wanna really get very granular in how I do stuff and I want to download the emal, then here's how you can do it.

And what's nice about this interface is that as I go through here and start picking things and changing things, um, I can, uh, we'll, we'll see the Kubernetes YAMBLE manifest change on the right hand side so they can download this, reuse it, um, and, uh, and also, you know, use it as a reference if they're building their own blueprints and so forth. Um, I'll pick that VM class. Remember earlier we, we chose that, uh, VM class.

So these are the, the, uh, uh, NAYSAY class, I'm sorry, these are VM classes that are part of that for that namespace. Um, and we'll go ahead and hit next here. And then what's nice about this too is when I deploy this vm, this is actually a virtual machine.

If you notice the name of it's my sql, um, we're gonna actually install my SQL on this vm. And, but the other thing that I could do with this VM is I can add a persistent volume, just like I, like I would in Kubernetes, right? So I could come in here and add a five gigabyte persistent volume.

Um, I can expose SVM as a service or an application using a load balancer. So I could come in here and say, okay, I'm gonna, I'm gonna, um, um, you know, add SSH in my SQL ports. So now I've got the MySQL 33 0 6 port opened up.

So now I can access this from the outside world as long as I'm on a public subnet, which I'll change here in a second. And then, uh, we can start to, you know, say, okay, now this, this load analysis server should be out there for us, uh, to consume. Now, the other thing we've done as well is because we're deploying a virtual machine here, using this custom code, like this VM operator code right here that you see, we also have the ability to inject cloud and net scripts.

Now, we could do this before, but we, we do have some nicer ways to do it now. One is, we have some guided inputs, but you can also just like hit raw configuration and then copy and paste your cloud in its script here, and then it gets injected into the Kubernetes YAML script. So here I, I'm able to, you know, add users, install my SQL on here, and that's kind of what I'm doing.

And I'm also gonna add a network interface because I want to publicly, uh, uh, expose that load balancer service, right? Using a public subnet. So I'm gonna do that here real quick and just hit save, and then go ahead and hit next.

And, uh, we'll, we can deploy in this VM once we're ready. Now, the thing I can do before that is I could download this, I could go ahead and download vml, open it up in a, in an editor, um, and so forth. Um, I can also copy and paste this into a blueprint.

The other thing that's really nice about this interface too, from, for me, for someone that uses it, is that when I deploy something after the fact, I can click on that VM and then see the yaml, the deployment yaml, and then edit it. So if I wanna edit like a label or make some change to the Kubernetes Gamble, I can do that, um, later on. Okay, so let's go ahead and deploy the vm.

Uh, once we deploy the vm, we can see that's powered on and we can get some information about it, uh, and so forth. And then if we go over here to the overview, then we can do the same thing with the Kubernetes service. So I can deploy a, uh, a VKS cluster, uh, right from here as well using this ui.

Um, so if I go in here to create, you'll notice, um, I can do a custom configuration and the YAML also gets built out on the right hand side for me, um, as well as some changes, some things I can do here. So I can pick the cluster class, I can pick the Kubernetes release version that I want. Um, and, uh, so if I just pick like the latest here, uh, yeah, yeah.

And then we can add labels, stuff like that. We can also do like, like interesting things like, you know, sort figure rotation for them so that way they don't have to, you know, rotate their certs, uh, for, for the different services to talk to each other. Um, and then also, uh, just, you know, we'll add a storage class here as well.

Um, so that way when we, when we deploy our worker nodes and our, uh, uh, our, uh, control plane nodes and stuff, we have some, uh, options there. Now, the one thing we wanna do is we'll wanna deploy our control plane nodes. We can do one or three here, and then there's the VM class.

So this will be your, I'm sorry, not your worker nodes, but your control plane nodes. So that'll be like your CD servers, your cube servers, all that stuff on the back end. Um, and then, uh, we'll pick that, uh, yeah, we'll do that stuff for, okay.

And then our node pools, which is gonna be our worker nodes. So we can pick, you know, multiple worker nodes if we need that capacity. We can also pick what VM class, how big we want these worker nodes to be.

So if we know we're gonna, you know, we, we know, hey, we don't want to have to scale out too much and all that kinda stuff, which you can do after the factory. You can scale up these nodes if you want to. Um, but, uh, or scale out, right?

Um, you know, add more nodes and stuff like that. Um, we can do that, but this is essentially what's going to, uh, be where our pods will live, right? So we'll change that to two and then we'll finish that.

Um, so that's basically deploying the VM and deploying a VKS cluster, uh, using the, the services ui. And I guess I'll pause there for just a second. I, uh, and then go to the catalog VM plus I could add GPUs into that VM class and have a, an AI application running on this.

Um, well, well, not, not the supervisor cluster, obviously, but in the, in the workload notes, I could have the, uh, GPU added as well and run my ai AI application on this. Yeah. Yeah, right.

So the right, if we need like, like let's say GPU, uh, capabilities or something like that, um, then yeah, that would be an option, right? There's a number of check boxes on those. Again, there's like a bunch of different things you can do to integrate in with third party or like different stuff.

So you'll see all these options in there and they'll be like, okay, you know, connect to this and like some kind of GPU card or some graphics card or something, you know, so there's all kinds of things you can kind of do in there. I think it connects to different vendors, and then those capabilities could just be in that class. So you could call that class like, you know, my custom class and it could be inside your namespace class.

Um, and then you would just use that if you needed, right? Um, and, and that could be part of your, your, your application. So it's just a matter of bubbling everything up really, and, and making sure we can consume as much as possible using A DUI here.

Okay, so before I end it here, I wanted to, uh, go ahead and go through the catalog service, um, because this is the other way to consume. So there's really kind of three main ways to consume. One is the services UI just showed you.

Um, and now ultimately what I could do with that is I could go on the CLI, I could start doing stuff. I just don't have time to go through the whole process of deploying everything in the CLI right now. But there is an option to do that.

We have the V-C-F-C-L-I, which is very powerful plugin based system, uh, I think would be great to kind of do almost a, uh, uh, uh, love to do a deep dive with anyone on that eventually, because it's really some powerful stuff. Um, but the catalog is also still there. And we do have the ability to version control blueprints and push them to a catalog.

So if your users are not familiar, let's say enough with Kubernetes, or let's say they're not, um, uh, maybe maybe a little lower tech users, or you have use cases where you have more of an anything as a service type of thing where maybe these are folks that just need to fill out a form, uh, folks could still go in and, uh, and do that so they can go in and pick their dropdowns and stuff. And we still have things like, you know, conditional dropdowns in the form, custom forms, uh, you know, data grids and all that kind of fun stuff as well, um, that they can do. They have bullon values and stuff, so they can just go in and submit, um, as well.

And, uh, and then go ahead and deploy it. Now what we're kind of doing in this particular, uh, scenario is we're deploying into production from the catalog and, and using that services UI for dev. So it's kinda like, you know, you might have some folks that are in development testing, maybe they go and use Services UI and the V-C-F-C-L-I to de to develop and deploy things or develop and, and kind of test and, and then you can take your production name space and then, you know, curate all that, throw it into a catalog, and folks can just kind of come in and consume that, right?

Um, so that's kind of one of the ideas here. And then what they can do is they can, uh, you know, kind of go into their, um, uh, deployment that they did and see the topology of their deployment. So this is if you went into the catalog, uh, and deployed it from a curated blueprint, uh, and, uh, and, and for more of your just general users.

And what's nice about it is they can kind of see all the objects that get deployed, take any potential day two actions, and then see if there was any errors or anything that showed up when it got deployed. This could be A, an internal developer platform, uh, kind of u user interface. This, this strikes me as a, a way for a platform engineer to start delivering, um, that IDP environment for their developers.

Yeah, for sure. I, in fact, I see that as probably a, a great use case here as I use VS. Coli.

I use other things a lot, but I've really like, kind of like this interface. I've liked the workflows and it's really nice to be able to switch between namespace, uh, very easily and see all the objects in those namespace. Um, because I have a drop down there with all the namespace I'm working with versus having to go in and, and, and use commands all the time.

Uh, so, so that has been pretty nice. Um, and then also you like, because we have the different personas, you can curate that content, create that platform, and then users with different roles come in and just request it, right? So that, that is really, uh, in fact, I've, I've had, uh, some customers in the past call this a developer portal, um, because, uh, they're, they're essentially looking at that and, and saying, here's all those services.

Now as we introduce all these other things, like I mentioned, Argo, cd, uh, DSM, your database as a service, which you may learn about some of these other sessions potentially. Um, but uh, uh, you know, all those things will be bubbled up too and getting bubbled up in the product. Uh, so, so that'll be very interesting too.

'cause then you can really, uh, deal with everything from your infrastructure to your backup, to your security, to your image registry, uh, to your continuous delivery and so forth and so forth. Um, and so it really does become a, a great way to sort of do that path has and, and, and is type solutions. Uh, where is the source of through for all the, every action we are doing here in this, um, you know, this automation is inside the Platform is Double okay, or we are able to export or integrate with a counter version or stuff like that.

Yeah, no, great question. Um, so couple things on that. Um, I guess you're, you're thinking about audit trail or hey, what, what are people doing in the system?

How do you track that? How do you track things that happen in the system? Yeah, yeah.

Great, great, great. Uh, great question. And so there's a couple of points during the setup of all of this where it's gonna ask you where to send logs for like networks and other things.

However, there's really nice log insight integration into this. So when you have the whole VCF platform, one is each individual product will have their own events and tasks and audit section usually, right? Automation does, operations does inside of automation.

There's events and tasks, there's audit sections where you can get quite a bit of info in there. But we also send everything to Log Insight. And Log Insight is one of our log tools that we've had for a long time.

Um, but it's a really, really great powerful logging tool, uh, that, you know, can really help you track and trace a lot of things that happen. So for instance, like if you have a service account or anybody going into doing stuff inside of BCF automation, you can track that inside of the Log Insight logs and really do your analysis there. Um, otherwise, you know, we've been just, we've also, I've also just used the, the internal audit and events and tasks and stuff like that.

In fact, each portal type has it. So if you're in the provider section of the product where you're really managing like an enterprise IT admin, managing all the organizations, you have an audit trail events task there with logs going out, and each organization will as well. So there's quite a bit there from that.

And then there's also the integration with the operations tools, um, which also gives you capabilities of like, alerts and, and, uh, config drifts, um, as well as a lot of auditing there as well. Um, and, and especially around compliance and stuff. So, yeah, I, I, you know, hopefully that answers it, but I, but I, but there are ways to, to do that, um, in, in terms of inside the product as well.

Alright, well, if there's no more questions, I appreciate everyone's time. I really thank you for, for, uh, hanging out and hopefully this was helpful for you. Thanks.