SUNK
CoreWeave adds guided self-service to its SUNK AI training system
CoreWeave is extending SUNK, its Slurm on Kubernetes training system, with a guided self-service preview for deploying and managing clusters through the CoreWeave…
CoreWeave CEO: AI's next phase is about production, not possibility
CoreWeave CEO Michael Intrator says the central challenge for AI in the coming year is production, not possibility. In a blog post titled "The Year AI Gets to Work", he…
CoreWeave reports 51-52% MFU and 3.66-day MTTF in 1,024-GPU H100 test
CoreWeave says a six-week benchmark of up to 1,024 NVIDIA H100 GPUs reached 51-52% Model FLOPs Utilization (MFU) and a mean time to failure (MTTF) of 3.66 days. The…
CoreWeave adds Telemetry Relay and GPU Straggler Detection to Mission Control
CoreWeave has expanded its Mission Control operating standard with Telemetry Relay, now in general availability, and previews of GPU Straggler Detection and a…
CoreWeave Adds Self-Service and SUNK Anywhere to Its AI Training System
CoreWeave is expanding SUNK, its Slurm-on-Kubernetes training system, with self-service deployment and a new option called SUNK Anywhere that extends the system to…
CoreWeave's SUNK runs Slurm inside Kubernetes on one shared compute pool
CoreWeave has introduced SUNK ("SlUrm oN Kubernetes"), an implementation that runs the Slurm workload manager on top of Kubernetes so that HPC and AI teams can use a…