HPC · High-Performance Computing
The scientific supercomputing tradition AI clusters descend from — gang scheduling, low-latency fabrics, parallel filesystems — now repurposed for training at scale.
Current numbers
~70% / ~20% / ~10%AI/HPC scheduler share: Slurm / Kubernetes / in-house (rule of thumb)
500–2,000 AVery-high-current delivery range for modern AI/HPC processors at sub-1V core voltages; peak requirements reach approximately 2,000 A
~41% / ~17% / ~6%parallel-FS deployment share: Lustre / IBM Storage Scale (GPFS) / WekaFS in HPC surveys