Post-Kubernetes / Decentralized Orchestration at Scale
The emerging pattern of abandoning Kubernetes' centralized, strongly-consistent coordination model (etcd, single scheduler) in favor of decentralized architectures — per-node sources of truth, parallel schedulers, optimistic accept/reject placement — to reach the millisecond-latency, million-instance scale AI-agent sandboxing demands.
Also known as: beyond Kubernetes, decentralized scheduling, horizontally scalable by default, load-balancing-style scheduling, execution plane vs coordination plane, per-worker source of truth, walking around Kubernetes
How authors use it
Oldest first (extraction date when no publication date is known).
| Author | Their term | Their meaning |
|---|---|---|
| Beyond Kubernetes at Modal: How to Scale 1 Million Concurrent Sandboxes in Seconds Sergio De Simone · 2026-09-23 | horizontally scalable by default | Modal's governing design constraint: any operation whose load is O(sandboxes) or O(nodes) must be able to scale out across independent servers rather than route through a single coordinating component. |
Documents (1)
- Beyond Kubernetes at Modal: How to Scale 1 Million Concurrent Sandboxes in Seconds Sergio De Simone · 2026-09-23 Reader 🖥️
Modal replaced Kubernetes-style centralized, strongly-consistent orchestration with a decentralized architecture — per-worker sources of truth, a parallel fleet of scheduling servers acting like load balancers, and a single Redis stream as the only shared bottleneck — to run over a million concurrent AI sandboxes, demonstrating that extreme-scale, millisecond-latency AI infrastructure requires walking away from, rather than extending, Kubernetes' coordination model.
Propositions touching this concept (1)
By publication month, oldest first (extraction month when no publication date is known).
2026-09
- explicitModal's core design principle was that everything with O(sandboxes) or O(nodes) load must be horizontally scalable by default, the sandbox-creation path should be as simple as possible, and everything else is secondary.
Open gaps (1)
- incompleteThe article surfaces Alex Jones's prediction of an execution-plane/coordination-plane split in AI infrastructure but doesn't examine what integration mechanism would let a Modal-like execution layer interoperate with a Kubernetes-like coordination layer — a concrete architecture question other corpus pieces (GKE Agent Sandbox, virtual-kubelet-based integrations) only partially address.Beyond Kubernetes at Modal: How to Scale 1 Million Concurrent Sandboxes in Seconds — read next: Read the GKE Agent Sandbox hypercluster extraction and the Unikraft/1M-sandbox extraction's virtual-kubelet section for how others are gluing decentralized execution back into Kubernetes' coordination plane.