Why Kubernetes Still Needs VM Infrastructure
The idea that Kubernetes would eventually replace virtual machines entirely has not played out the way many predicted. Managing bare-metal clusters at scale has proven to be a complex and costly undertaking, and the industry is increasingly recognizing that VMs still have a critical role to play in modern infrastructure. In this conversation from KubeCon Europe, Mike Vizard speaks with Weiguo He, senior director of product marketing at Broadcom, about how VMware Cloud Foundation is evolving to meet the demands of cloud-native and AI workloads.
He describes how Broadcom is leveraging decades of infrastructure experience to address one of the most pressing challenges facing organizations today: dynamic GPU resource allocation. As AI workloads grow in both scale and complexity, the ability to efficiently provision and manage GPU resources across a mixed environment of containers and virtual machines is becoming essential. VMware Cloud Foundation is designed to make that process more accessible without requiring organizations to abandon their existing infrastructure investments.
The conversation also explores the economics of running Kubernetes in production. Hardware costs continue to rise, and security requirements are becoming more demanding. Rather than treating containers and VMs as competing approaches, He argues that the most practical path forward is running modern containerized workloads on top of a proven, resilient VM foundation. This approach gives organizations the flexibility of Kubernetes while maintaining the operational stability and security controls that enterprise environments require.
For platform teams weighing the trade-offs between cloud-native agility and infrastructure reliability, this discussion offers a grounded perspective on where the industry is heading.
