OSS = Open Source ... Strategy!? Google Is Doubling Down on K8s in the AI Era, and You Should Too!
Talkat KubeCon + CloudNativeCon Europe 2025 · on Wednesday, 2025-04-02· at 16:15· Level 1 | Hall Entrance N10 | Room G·
summary
Google commits to doubling down on Kubernetes for AI workloads, claiming it must scale to the next trillion core hours and already supports large fleets (GKE uses Spanner to run 65,000 nodes). Concrete moves: CRDs GA in 2019, Spark-on-Kubernetes work in 2018, investments in DRA/accelerator integrations for GPUs/TPUs, and changes to upgrade and multicluster reliability. Strategy pillars are declarative APIs, extensible framework orchestration, and differentiation on price‑performance via deep cloud integration.
AI-written summary · about
chapters (19)
- 0:00Open source strategy at Google
- 1:04Google’s open source history
- 2:29Kubernetes disruption phase
- 4:14Why Kubernetes succeeded
- 5:59Extensibility in Kubernetes
- 7:44Ecosystem expansion
- 10:33Stability in Kubernetes
- 11:57AI era disruption
- 14:07Mission-critical apps
- 15:51Path dependence in adoption
- 17:17Reliability in Kubernetes
- 18:42AI/ML frameworks
- 20:26Kubernetes infrastructure model
- 22:55Google’s K8s strategy
- 24:45Stickiness in open source
- 25:47Kubernetes evolution in GKE
- 27:14K8s for AI/ML layers
- 28:40Future investment in K8s
- 29:00Q&A
related talks
20mArchitecting for Innovation: Building AI Powered Apps and Platforms
09mConversation with Sarah Novotny, Head of Open Source Strategy for GCP, Google
35mThe power of Open.
14mKeynote: Building the Infrastructure that Powers the Future of AI
42mThe State of Open Source Cloud
27mOpen Source in an Age of Transformation for the Energy Industry
23mThe State of Cloud Native: The Shift Towards AI
20mOperating Frontier Intelligence at Scale