Kubernetes Adds Gang Scheduling and Finer GPU Control for AI Workloads
Kubernetes 1.38 moves gang scheduling and finer GPU allocation toward stable, while AWS shares control-plane fixes aimed at 100,000-node clusters.
Development operations, CI/CD, and automation
Found 237 articles
Kubernetes 1.38 moves gang scheduling and finer GPU allocation toward stable, while AWS shares control-plane fixes aimed at 100,000-node clusters.
Cosmonic packed 1.42 million Wasm sandboxes for AI agents onto one 64-core server running Kubernetes. Lightweight isolation is the pitch.
West Virginia online casinos set a record October in bet365's second full month. 2026 revenue has passed all of 2025.
Idle AI agents are tying up Kubernetes pods and straining the control plane. Google's Agent Substrate, gVisor, and NVIDIA's OpenShell point to the fixes.
MotorCity was the only Detroit casino to grow table and slot revenue in October, while MGM and Greektown declined. Retail sports betting jumped too.
Threat intel teams are consolidating vendors to sharpen correlation and prove outcomes, not to trim budgets. Here's where the 2027 survey shows gaps persisting.
CoreWeave Database runs managed PostgreSQL inside customers' Kubernetes clusters, cutting cross-cloud hops for AI workloads. It's now generally available.
EDB donated Klio to CloudNativePG and previewed enterprise add-ons for encryption, support and multi-cluster backup management on Kubernetes.
SUSE expands Rancher Prime with task focused AI agents, hosted RKE2 control planes and CLI previews for infrastructure changes.
Replay realistic API traffic in isolated Kubernetes previews to catch behavioral regressions before merge and keep test results easier to trace.
AKS adds managed GPU and network allocation, while Fleet Manager gains placement controls and upgrade checks for multi-cluster operations.
PENN Interactive revenue rose 17% in Q3, but online gaming revenue slipped. Lower costs have narrowed losses as the segment targets fourth-quarter profitability.
FanDuel’s US revenue grew to $1.455 billion, but profit fell sharply as Flutter invested in prediction markets and Alberta expansion.
Kubernetes can keep workloads running through failures, but teams also need independent access to repair the networks and systems beneath them.
Arkansas is challenging VGW’s sweepstakes model and Kalshi’s sports contracts, citing $66 million in VGW sales and a lack of state gambling licenses.
Kubernetes response automation can erase evidence or trigger configuration drift. Agentic SOCs need live cluster context and clear approval limits.
AI is speeding up both applications and screening, creating a hiring cycle that sidelines recruiters. The lesson: design controls around human intent.
A K3s platform uses Prometheus metrics and temporary, node-local logs to investigate incidents while limiting centralized collection of user data.
Kinference helps teams manage distributed vLLM workloads, improve GPU use and route selected inference requests to cloud capacity.
GKE automates much of Kubernetes infrastructure, but low utilization persists. Workload data, smarter capacity choices and gradual changes can help close the gap.