Numa Python Packages

tensortorrent

PyTorch compiler and Rust runtime for multi-CPU/GPU inference, NUMA-aware scheduling, model serving, parameter streaming, and activation spill.

knor

A repo to allow validation of performance results in the knor paper and provide a fast, scalable k-means implementation.

pyexpool

Python Multi-Process Execution Pool: concurrent asynchronous execution pool with custom resource constraints (memory, timeouts, affinity, CPU cores and caching), load balancing and profiling capabilities of the external apps on NUMA architecture

lltune

LLTune tunes Linux servers for ultra-low latency workloads. It scans system state, generates a config, audits for risks, applies safe CPU/ memory/network/kernel/IRQ tuning with backups and rollback, and stages persistence services for reboot.