GKE 1.37 adds native scale-to-zero, ending reliance on KEDA setups
rseroter · x · 2026-09-24
Google Cloud announced native scale-to-zero in GKE 1.37: sporadic workloads (batch processors, event-driven workers, dev environments) can scale fully to zero replicas, then restart quickly on GKE capacity buffers when demand returns. Compared to KEDA, GKE's managed approach removes extra operators and CRDs (KEDA fleets can exceed 10,000 lines of YAML) and internalizes the scaling signal path to cut cold-start latency. Framed as decoupling always-on infrastructure cost from workload readiness.
More from Infra
- French netizen proposes datacenters pay local residents compensation — clemnt · 2026-11-11
- GPU Rental Prices Hit Record Highs: B200 at $6.97/hr, H100 at $3.06 — inductionheads · 2026-11-11
- A Datacenter in Every Office: Building Work-Safe Local AI Racks — dee_hw · 2026-11-11
- "Compute is about power": the AI race's real bottleneck is energy, not chips — _sholtodouglas · 2026-11-11
- CoreWeave CEO says he canceled his 10b5-1 plan, halting insider selling — firstadopter · 2026-11-11
- UK chip start-up Fractile hits $7bn valuation, per FT — evisoft · 2026-11-11