Latest / The MonkCast / VKS: One Platform for VMs and Containers with Audrey Bian
Transcript
- speaker-1: Hello everyone and welcome to the MonkCast. I'm Rachel Stevens and with me today I have Audrey Bianne. She's a Principal Product Marketing Manager with Broadcom. Audrey is the PMM for vSphere Kubernetes Service or VKS. And today we're going to be talking about how she helps teams simplify their Kubernetes journey. Audrey, could you please introduce us to who you are and share a little bit more about what you do with Broadcom? speaker-0: Sounds great. Thank you, Rachel. Hi, everyone. My name is Audrey Bien. I am a principal product and marketing manager at VMware, now part of Broadcom. So I've been with the company for seven years. Actually, this month is my seven-year anniversary month with the company. ⁓ I started with hyperscaler solutions, and now I focus on vSphere Kubernetes service. speaker-1: And I think one of the things that can be tricky for people who do not follow the VMware portfolio closely is that there have been evolutions in names and packaging, and especially in our industry of three-letter acronyms. Sometimes it's hard to keep up with all of the changes. So for people who are maybe not familiar with VKS, let's set a baseline. ⁓ I'd love for you to help us run through what exactly is VKS, where does it sit within the VMware stack? And what are the specific unsolved problems that you are trying to help solve in the enterprise? speaker-0: Yeah, that sounds great, Rachel. So at a high level of Vsphere Kubernetes service, we go with VKS. It's VMware's enterprise Kubernetes runtime that we built directly into VMware Cloud Foundation, which is BCF. So BCF is our unified private cloud platform that bridge library, computer storage, networking, and cloud services all together in a single platform. So what VKS does here is It brings the fully conformant Kubernetes so organizations, can run modern containerized applications and the traditional enterprise workloads on the same platform. the problem we are solving today is there are so many enterprises today, they are operating like two separate platforms, such as like one for traditional VMs and another one for Kubernetes platforms for modern apps, which means they need to get like a new infrastructure, new tours, new operational systems, that fragmentations create just operational complexity and this slows down the application delivery. So with VKs, as we can see from this slide here in the box, we provide open source services to our customers and also extendable for like third party. services as well. ⁓ basically bridges the gap that platform engineers, they are provided with self-service while also continuing to use the same infrastructure, security policies and operational tooling they already trust in VCF environment. Basically, our goal is very simple to help enterprises modernize their applications faster without forcing them into re-use their infrastructure or operational practices. from scratch. speaker-1: Wonderful. And you started to actually inch towards my next question. One of the things that we talk about a lot at RedMonk is that different personas within the enterprise experience technologies differently. So the experience of using Kubernetes is not monolithic across all the different users. And it's a different experience depending on where you sit in the org chart. It seems like you have different personas in mind based off of how you're starting to talk about this, but can you give us a breakdown of the personas you're building this product for? speaker-0: Yes, happy to let me go to the next slide. So there are three key personas we will talk about today. There are like first decision makers, like a CEO, CTO, and the platform engineers. The last is our cloud admin. speaker-1: Okay, let's pause and interject. What is a cloud admin versus a platform engineer? speaker-0: Yeah, that's a great question. Actually, our customer also asked a lot of these questions. Can help me to like differentiate these key personas? So in VCF model, Cloud Admins and Platform Engineers, basically they operate at different layers of the stack. So for Cloud Admins, they are more focusing on running the underlying infrastructure, such as compute storage networking and lifecycle management of the VCF platform. For platform engineers, then they build on top of that infrastructure to create a developer platform. So such as we're using services like VKS to provide like self-service infrastructure, modern Kubernetes, and a standardized environment and automation for application teams. So you can think of the cloud elements they operate in the cloud, and the platform engineers, they're enabling. developers to build and deploy their applications on it. speaker-1: Given that the people in all these roles will have different day concerns, especially if we're kind of differentiating between practitioners and the decision makers, let's maybe dive in a little bit more. So if we start with the cloud admins that we just went into, they've spent a decade or more kind of mastering the vSphere universe. What is the actual quality of life benefit that they have by having VKS baked into VCF rather than managing Kubernetes as like a separate entity? speaker-0: Yeah, that's a great question, Rachel. So ⁓ here I have a slide for you. In this slide, we're going to highlight what's in it for our Cloud admins. The first is unified platform, like I mentioned earlier, for VMs and containers. So customers can run traditional and modern workloads on a single VCF platform with consistent operations and fewer infrastructure silos. And also the second part is a simplified Kubernetes operations. As we all know, Cloud Admin State, a lot of them coming from the VR admins. So because of the unified set of the APIs we offer, so Cloud Admin State now can provide, can provision, upgrade, and scale Kubernetes using very similar VMware workflow and existing infrastructure skills. And also I want to mention, help them to change their career path. and without too much of the learning curve. And then the third part is the centralized governance and policy control. Basically, it enforces the security, compliance, and operational standards while also enable our team to move faster. We also support the multiple Cobra Danny versions in each release. So our teams, are not forced to upgrade ⁓ in a fixed time. They can stay in a version ⁓ as long as they they want and the works for the team and upgrade as they are ready. The last but not least is we have very strong workload isolations because it built on the multi-layer isolation which improves the fall containment and also protect against cyber threats. speaker-1: that because there's then benefits for them career wise, there's benefits for them adopting new technology and there's benefits for the org overall. All right, so let's move on to platform engineers. These generally are folks who are trying to build like an internal development platform for their developers so that the developers don't have to think so much about the tooling and can deploy with ease. If I'm a platform engineer, why would I be looking at VKS as a foundation and how does that help me versus speaker-0: Exactly. speaker-1: other Kubernetes deployment options. speaker-0: Yeah, so basically, because we provided the enterprise Kubernetes platform, like I mentioned earlier, Rachel, which we have this fully CNCF certified and integrated directly into the VCF. So it allows our team to quickly spin up Kubernetes clusters with flexible version support and extended 24-month lifecycle support, which reduces upgrade pressure and customer state. get the same level of support for 24 months, which is the longest in the industry. It also integrates with common platform engineer tools like ArgoCD, Helm for GitHub workloads. So on top of that, also we provide the multi cluster management, like policy and a backup at a scale, along with the consistent API across the VMs, Kubernetes and services like ⁓ So the team can operate everything from a single platform. speaker-1: The open source project is releasing new things every six months. So for organizations that can be really hard to keep up with. speaker-0: Yes, and also I want to mention we also go, every four months we have a new VKS release. We want to make sure our platform teams, they are staying on top of the innovation. Yeah. speaker-1: And last but not least, let's talk about these kind of IT and executive decision makers, what appeals to them about the VK platform. speaker-0: So I think for a lot of decision makers, the top priority is the TCO. So we help organizations to modernize the applications. Also at the same time, we try to lower the TCO as soon as possible. Because the reason we mentioned, they don't need two separate platforms anymore. They can run VMs in containers on the same infrastructure. We're the silos and also reducing the license cost, operational overhead, all of these lead to a lower TCO for our decision makers. And the second one is fast time to value. So basically, NVCF enables a faster application delivery, so organizations, they can move from idea to projections more quickly, which means like projects, they deliver value sooner and start producing our ROIs earlier. The third part, I think all the decision makers, they all care about is the security and compliance. It means like better protection and in ECF we have also we emphasize on these six layers of isolation for VMs and containers, which minimize the noisy neighbor issues, which happens a lot on bare metal. And also data in ECF is encrypted, end-to-end, both at risk and intrinsic. At the same time, it's also reduced the Kubernetes skill gap. Like I mentioned earlier, Rachel, like the same API, we allow teams to use familiar VMware tools, and it's backed by our strong open source leadership with Broadcom as a top contributor in the CNCF ecosystem. speaker-1: Interesting. And I think that maybe that's not fully appreciated how much VMware is contributing back to the upstream. speaker-0: Yeah, as a VMware has always been the top five contributors to the open source. And I think we need to emphasize that a lot more. speaker-1: Let's close, Audrey, by talking about some of the features that you've shipped in your latest release. think it is VKS 3.6. So let's hear more about that. speaker-0: Yeah, the release just came out last month. So we are very excited about this release. There are three key pillars here. We talk about networking flexibility, optimized performance, and also we offer enterprise OS choices for our customers. So let's maybe take a quick look on each pillar. In the first pillar, we are giving platform engineers support for partner-provided CNIs. such as Isovilent and Caelium and Calico Enterprise from Taikera. So why this is important is because in our earlier releases, we can ask the standard package include only by two built-in CNI options, Entrez and Calico. So customers, they could choose between them, but the networking models was a little bit like fixed. So with VK3.6, we made a foundational changes. Our built-in CNIs are now delivered as add-on, and we introduced new variables in this built-in cluster class that allows our customer to select which CNIs they want to use at their customer creation time. And we will keep adding more options for our customers. And in the second pillar, we want to communicate to our customers is that Advanced kernel level tuning use 2D profiles, which allows our platform engineers to optimize Kubernetes nodes for performance critical workloads without they don't need to manually change it from OS level. So the key idea here is that the 2D profiles are managed declaratively by our VKS control plan. Instead of like manually tuning individual nodes or relying on Kubernetes defaults, customers now they can define the policy-driven kernel configurations and apply them consistently across the clusters. So the third pillar is our customer now they can bring their Red Hat Enterprise Linux Rail cluster nodes in EKS. So they can leverage existing their real investment like skills. and security toolings. So this is very important because it allows our customers to reuse their OS investment and operational expertise instead of forcing into a migration to a more like fixed OS choices for our customers. speaker-1: Well, it sounds like you all have been very busy. ⁓ So it sounds like if you are a cloud admin, a platform engineer, or a technology decision maker, the VKS team has a story that they would love to tell you. So if people are interested in hearing more, where should they go? speaker-0: ⁓ think the best way is to go to our webpage. Actually, we have a very easy to remember a link as a vmware.com slash vks. speaker-1: Okay, wonderful. Audrey, thank you so much for your time today. This has been a wonderful conversation. speaker-0: Same here, it's a pleasure. Thank you, Rachel.