News

TechTarget: “CNCF’s llm-d 1.0 tackles high AI inference costs”
TechTarget: “CNCF’s llm-d 1.0 tackles high AI inference costs” October 20, 2026
The llm-d project, now ready for production, aims to do for distributed inference on GPUs what Kubernetes did for web apps: make the most of infrastructure resources.

The New Stack: “AI agents turn one request into about 20. Kubernetes is evolving to handle it.”
The New Stack: “AI agents turn one request into about 20. Kubernetes is evolving to handle it.” October 16, 2026
Welcome to another edition of Road to KubeCon, where we’re tracking interesting movements in the cloud-native and Kubernetes ecosystem as we count down the days to KubeCon + CloudNativeCon NA 2026, to be held in Salt Lake City,...

The New Stack: “Kubernetes on cgroup v1 is dead. Here’s what comes next.”
The New Stack: “Kubernetes on cgroup v1 is dead. Here’s what comes next.” October 9, 2026
Edge AI is reshaping Kubernetes: node swap lifts density up to 3x, cgroup v2 takes over, and Edge Day and CiliumCon return ahead of KubeCon NA. 

Virtualization & Cloud Review: “vMotion for Containers? How DevZero Uses CRIU to Cure Kubernetes and AI Sprawl”
Virtualization & Cloud Review: “vMotion for Containers? How DevZero Uses CRIU to Cure Kubernetes and AI Sprawl” October 8, 2026
When KubeCon + CloudNativeCon North America lands in Salt Lake City this November (9–12), its inaugural AI Inference + Agentic track will take center stage. This is a clear indicator that Kubernetes is solidifying its role as the de facto control...

The New Stack: “Kubernetes won the orchestration wars. Its teams are still fighting.”
The New Stack: “Kubernetes won the orchestration wars. Its teams are still fighting.” October 7, 2026
Kubernetes won the orchestration wars, but dev and ops teams still pull in opposite directions. HPE’s David Estes thinks AI may ease the strain.

The New Stack: “What Kubernetes’ “monolith” lesson means for AI agent harnesses”
The New Stack: “What Kubernetes’ “monolith” lesson means for AI agent harnesses” October 2, 2026
In this edition of our Road to KubeCon series, Craig McLuckie wants agent harnesses off the laptop, and Koordinator pushes GPU allocation past 95%.

The New Stack: “A live Kubernetes cluster can still have an ownership gap”
The New Stack: “A live Kubernetes cluster can still have an ownership gap” October 1, 2026
After a Kubernetes cluster goes live, platform and application teams need clear responsibilities for upgrades, access, recovery, and shared workloads.

The New Stack: “OpenTelemetry and Prometheus are getting along. What’s still missing?”
The New Stack: “OpenTelemetry and Prometheus are getting along. What’s still missing?” September 25, 2026
OpenTelemetry and Prometheus are getting easier to use together. Teams are seeing gains, but mismatched data models still complicate observability.

TFiR: “Kubernetes 1.37: Scale to Zero, Checkpoint Restore, and API Server Resilience | Dipesh Rawat”
TFiR: “Kubernetes 1.37: Scale to Zero, Checkpoint Restore, and API Server Resilience | Dipesh Rawat” September 24, 2026
In this interview on TFiR, Dipesh Rawat, Kubernetes v1.37 Release Lead at CNCF, walks through every major enhancement in the 1.37 release, covering HPA scale to zero, Resilient Watch Cache, Pod-level Checkpoint and Restore, key deprecations, and the broader...

The New Stack: “Kubernetes can run AI inference. But can it count the real cost?”
The New Stack: “Kubernetes can run AI inference. But can it count the real cost?” September 18, 2026
One bank cut token-processing costs 60% with a Kubernetes stack. But a warning about its resource model raises questions about AI inference economics.