Application vs Infrastructure-Level High Availability with vSAN

This episode of the Virtually Speaking Podcast features hosts Pete Flecha and John Nicholson joined by returning guest Pete Koehler for a deep dive into vSAN and the evolving conversation around application versus infrastructure high availability. Pete Koehler highlights recent updates to vSAN Ready Nodes that significantly reduce hardware requirements, including up to 67% less RAM and 50% fewer CPU cores for vSAN storage cluster hosts—bringing notable cost and efficiency benefits. From there, the discussion shifts to availability design strategies. The trio explores the differences between application-level availability (features built directly into the app, such as database clustering or microservices patterns) and infrastructure-level availability provided by the virtualization platform. They cover when each approach makes sense, how they complement each other, and why a balanced strategy delivers the best resilience. Whether you’re a VI admin, an architect, or an application owner, this episode breaks down the real-world tradeoffs and helps you understand where the platform should handle availability—and where the application should take the lead.