Performance monitoring and benchmarking suite
-
Updated
Aug 19, 2026 - C
Performance monitoring and benchmarking suite
POWER9 gzip engine documentation and code samples
Information on setting up an HPC cluster on CentOS using xCAT, CUDA and Slurm on an IBM POWER9 system.
Supporting GATK4 on IBM Power architecture, including enablement, performance tuning and optimization.
SGLang fork for IBM POWER9 (ppc64le): Tesla V100 sm70, CUDA 12.4, Granite LLM inference. Triton attention, float16, OpenAI-compatible API.
Which 16 of 24? Recovering per-die CPU core-harvest maps from out-of-band BMC telemetry — 1,962 POWER9 sockets, reproducible from public M100 ExaData
PyTorch 2.12 fork for IBM POWER9/POWER10 (ppc64le) with CUDA 12.4 and Triton. Tesla V100 sm70, GPU training and LLM inference.
OpenAI Triton compiler fork for IBM POWER9/POWER10 (ppc64le) with CUDA 12.4. Tesla V100 sm70 GPU kernels for PyTorch and SGLang.
Add a description, image, and links to the power9 topic page so that developers can more easily learn about it.
To associate your repository with the power9 topic, visit your repo's landing page and select "manage topics."