1 week ago
11
Few tools divide opinion like Kubernetes—equally loved and loathed. There has been a recent surge in companies moving away from it to try their own monolithic architecture and building in-house solutions. Then, there are many moving back to Kubernetes, and Portworx by Pure Storage is helping companies do that.
Venkat Ramakrishnan, VP and GM at Portworx by Pure Storage, has had a front-row seat to what he calls a “Kubernetes dial-tone moment”—where Kubernetes is no longer just taking off, but becoming the default across enterprise infrastructure.
Fresh from a bustling KubeCon Europe event in London, Ramakrishnan reflected on how enterprises are rapidly moving away from legacy platforms like VMware and choosing Kubernetes as their unifying control plane—and how Portworx is helping them every step of the way.
“We had folks from HSBC, Weasley, Roche—real global 2000 customers—joining in, plus some large banks and telcos that are still prospects. It was a fantastic exchange of ideas and best practices,” Ramakrishnan shared with AIM.
“We’re seeing a huge movement from VMware to Kubernetes,” he said.
“And honestly, if you had asked me six to eight months ago, I would have been a skeptic. But after working with several large customers, I’ve become a believer—Kubernetes can be the unifying control plane for both containers and VMs.”
What’s Unique about Portworx?
Recently, Pure Storage announced Portworx Enterprise 3.3, the latest version of its Kubernetes-native data platform, now enhanced to support virtual machine (VM) workloads through KubeVirt.
This upgrade allows enterprises to run and manage both VMs and containers side-by-side on Kubernetes—a timely move, as 81% of companies plan to migrate VMs to Kubernetes, according to a 2024 survey by Pure Storage.
By consolidating VM and container workloads on a single platform, Portworx offers a path toward modern, cost-efficient infrastructure, with potential savings of up to 50%, according to Pure Storage.
The introduction of RWX Block for KubeVirt-based VMs running on FlashArray or third-party storage enables scalable read/write access and improved performance.
Moreover, the release also includes enhanced support for KubeVirt platforms from Red Hat, SUSE, Spectro Cloud, and Kubermatic, with validated reference architectures and joint engineering efforts.
“We’re partnering very closely with Red Hat in this journey,” said Ramakrishnan. “They’ve done a fantastic job of integrating VMs into the Kubernetes experience using KubeVirt.”
Now, for those who might not be familiar, KubeVirt is essentially Red Hat’s way of bringing VMs into the containerised world. So, instead of managing two separate infrastructures—one for your containers and another for your VMs—you get a unified control plane on Kubernetes.
“Red Hat’s OpenShift and KubeVirt are helping us take customers from VMware and land them directly into a modern platform, without needing a full rewrite of every workload,” Ramakrishnan explained. “We’re seeing success with this model across industries—whether in auto, finance, or even retail.”
Why the Shift?
A combination of factors contributes to this shift. Ramakrishnan pointed to Broadcom’s acquisition of VMware and subsequent pricing changes as major triggers. “Many customers are looking for alternatives. Kubernetes isn’t just for cloud-native apps anymore—it’s becoming the infrastructure standard.”
Ramakrishnan outlined several “secular trends” fueling this surge in Kubernetes adoption, including CFOs pushing the cost pressure. “CFOs are asking CIOs and CTOs to simplify the stack, to run more with less,” he said. That’s pushing containerisation and replatforming efforts.
Enterprises are reconsidering their VMware investments. “We jokingly refer to them as VMware refugees,” Ramakrishnan quipped. These customers are increasingly looking to Kubernetes as a more future-proof alternative.
Generative AI is further reinforcing Kubernetes adoption. From running LLMs to managing vector databases and enabling GPU sharing, generative AI infrastructure is built on containers and orchestrated by Kubernetes.
Furthermore, with the rise of low-code/no-code platforms and code generation tools, modern apps are being containerised by default. “They just land in Kubernetes,” he noted.
“In telcos and with 5G and 6G, Kubernetes is the standard,” he said. And it’s not just the cloud or the data centres—factories and retail locations are seeing full deployments too.
In one standout case, a major US grocery store chain is running three OpenShift nodes per store—across hundreds of stores—with Portworx managing VMs at the edge. “We also have a satellite radio company doing the same thing, and one of the world’s largest automakers is in the process of transforming all their factories with Kubernetes. That’s a massive multi-year rollout.”
Enterprises are increasingly turning to cloud-native applications, especially Kubernetes, to manage their AI workload. According to another recent Pure Storage survey of companies with 500 employees and more, 54% said they were already running AI/ML workloads on Kubernetes.
During the training of AI models, there is often a large amount of data, while during inferencing, the data volume can be much smaller.
Murli Thirumale, GM (cloud-native business unit), Portworx at Pure Storage, earlier told AIM that Kubernetes addresses these issues by offering elasticity, allowing it to dynamically adjust resources based on demand. “This flexibility is inherent to Kubernetes, which manages a scalable and self-service infrastructure, making it well-suited for the fluctuating needs of AI and ML applications,” Thirumale said.
Is Kubernetes Really for Everyone?
Of course, not every use case fits neatly into Kubernetes. Ramakrishnan acknowledged the recent move by Gitpod and a few other companies like Juspay to leave Kubernetes, but sees it as a healthy sign.
“No technology serves 100% of use cases. Some performance-sensitive applications are better off as monoliths pinned to VMs. As the market matures, Kubernetes will crystallise around the most lucrative and repeatable use cases, and the rest will branch out. That’s a good thing,” Ramakrishnan said.
As enterprises adopt Kubernetes again, the demand for engineers who excel in the technology is also going to increase, and this will be a big challenge for enterprises.
