Bart Farrell: Who are you? What's your role and where do you work?
Yasmin Rajabi: Hi, my name is Yasmin. I'm the COO at CloudBolt. And I'm here with StormForge by CloudBolt to talk about Kubernetes right-sizing.
Bart Farrell: So what news are you bringing to our audience today?
Yasmin Rajabi: I'm excited. We have a bunch of new things that we've announced this week. The first one is we made our Kubernetes cost allocation GA. We've added new reports in the UI and added support for network costs so that you can see all of your costs down to the container level You get that visibility and also the recommendations to go out and automatically right-size them. We also added MCP support to the product so you know you can use tools like Claude and wherever you're asking all your questions doing your AI-ing these days and also ask questions with StormForge, change configurations, do kind of your day-to-day operations. And then we've made a few enhancements to our HPA algorithm. We also four years later got officially we're issued the patent for our HPA algorithm that allows us to adjust the requests of CPU and memory while preserving your scaling behavior by patching the HPA target utilization. So we're very excited about that patent. And we've also added the ability to detect group workloads. So think like a blue-green deployment, a canary deployment, or just AI workloads that are short-lived but are all related. The product can detect the usage patterns and then group them together and providing the recommendations. And something we technically released about a month ago, but we're very excited about is in-place pod resizing. So with the latest version of Kubernetes, I now support in-place pod resizing as GA. And the way that we've adjusted our product is to use that as a default. So the default experience is to resize in place. And then the product will make decisions based on should you, if we can't. right size in place, then defer it to the next restart or restart immediately. You have a few options as well. So we're very excited about that. It really helps people roll out the software and be confident in how it will get deployed, especially for workloads that are less tolerant to restarts.
Bart Farrell: What specific challenges do these announcements address?
Yasmin Rajabi: so honestly, there's a lot of tools that will give you recommendations. There's not a ton of tools that will help you automate that. reliably, but there's quite a few solutions out there that will give you the recommendations. What we found is that the difficulty to go from recommendation to actually applying it is not really a tooling challenge. It's more of a trust thing. And you can only move as fast as the speed of trust. So all of these features kind of go into helping build that trust for either a platform engineer rolling out the software or a developer that's receiving the automated right sizing so that Any organization, no matter what the scale they have, knows that the right-sizing solution will improve their reliability while also making them more efficient, therefore saving them some dollars in exchange.
Bart Farrell: How does this announcement change the landscape compared to what existed before?
Yasmin Rajabi: Really what this changes is that traditionally where a lot of kind of hurdles people would have to jump before they could. automatically right size and with in place pod resizing a lot of those go away the fear of okay what's going to happen when my pod restarts when you patch the workload to update the requests and limits for CPU memory was always something that people stressed about so we're very excited that now the default behavior is to resize in place and help kind of allay some of those fears
Bart Farrell: so for our open source community are these announcements are the things that you've announced are they open source And if so, where do they fit in the CNCF landscape?
Yasmin Rajabi: So our product is proprietary, but we work alongside all the open source tools in the environment. So we work very well with the HPA or with KEDA or Karpenter, which we love as well. And then with the GA release of In-Place Pod Resizing, as Kubernetes as a project matures, we like to make sure we're staying on top of the new capabilities and supporting them within our product as well.
Bart Farrell: Can you break down StormForge's business model?
Yasmin Rajabi: Sure. So StormForge's pricing model is based on vCPUs, so it's a total amount of requested vCPUs. We aggregate that into the product so you can see exactly what you'd be paying for. And the pricing scales as your vCPUs go up and down so that you know clearly what you're getting out of it. And typically, for every dollar a customer spends on our license, they save between $4 to $8. in returns.
Bart Farrell: What key advantages set StormForge apart from similar solutions in the market?
Yasmin Rajabi: So last week we were issued our patent. Four years later, we've been doing Kubernetes automated rightsizing since 2019. But we first applied for the patent for our HPA algorithm in 2022 and we're finally issued it. So we're very excited about that. That's really what differentiates ourselves between other competitors. It's how we preserve your HPA scaling behavior when we vertically right-size the pods. And then another thing is that we are fully SaaS-based and try and keep our agent very lightweight. We run just a couple of pods in the cluster and we automatically right-size those as well so that you don't have to pay extra for an optimization solution that runs in your cluster.
Bart Farrell: Looking ahead, what developments can our audience anticipate from StormForge and the announcements that you mentioned?
Yasmin Rajabi: We're really excited about some of the future roadmap items that we're working on, especially around improving reliability, taking a look at your HPA replicas and making suggestions, taking a look at the different workload types and how we can make more dynamic recommendations. So we're excited about things like CPU throttling response and being proactive on the different maybe reliability needs you might have in addition to kind of the cost savings.
Bart Farrell: What's next for you, Yasmin?
Yasmin Rajabi: What's next for me? I'm hoping lunch because I'm a little bit hungry, but I'm excited about the progress the team's made. I'm excited about kind of some of the things that we're hearing from our customers and what they've been able to do this year. So we have some exciting case studies coming up. pretty soon.
Bart Farrell: And if people want to get in touch with you, what's the best way to do that?
Yasmin Rajabi: Yasmin at stormforge.io or Yasmin at cloudbolt.io. Just an email away and would love to chat.
