Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

Zstandard - Fast real-time compression algorithm

Thunder Research Group's Collective Communication Library

LLM training in simple, raw C/CUDA

C macros for hash tables and more

Ongoing maintenance for the original UI Info Suite mod for Stardew Valley.

Collective communications library with various primitives for multi-machine training.

Extremely fast non-cryptographic hash algorithm

Handy PyTorch implementation of Federated Learning (for your painless research)

Optimized primitives for collective multi-GPU communication

【冲破内核瓶颈，让I/O性能飙升】DPDK工程师手册，官方文档，最新视频，开源项目，实战案例，论文，大厂内部ppt，知名工程师一览表

First Challenge & Tutorial for HPC Student Cluster Competition Beginner

🎉 Desktop Spotify Web Player Instant Synchronised Lyrics

A fast in-memory key-value store

Data Plane Development Kit

✨ This pack of 100+ gopher pictures and elements will help you to build own design of almost anything related to Go Programming Language: presentations, posts in blogs or social media, courses, vid…

✉️ Learn Network Protocol and Network Programming

NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.

The most customisable and low-latency cross platform/shell prompt renderer

A curated list of awesome smartnic tutorials, papers and projects.

Infiniswap enables unmodified applications to efficiently use disaggregated memory.

Unified Communication X (mailing list - https://elist.ornl.gov/mailman/listinfo/ucx-group)

NVIDIA DPU OPs collection

htop - an interactive process viewer

Official MPICH Repository

OpenSHMEM Implementation on MPI

A simple logging library implemented in C99

Open-source benchmark suite for cloud microservices

Mellanox userland tools and scripts