Resource scheduling and cluster management for AI

Best practices & guides on how to write distributed pytorch training code

AI动力(AI Power) GPU云平台使用指南

GPU management System on Kubernetes For AI, Deep-Learning, Machine-Learning Researcher.

qihoo360 xlearning with GPU support; AI on Hadoop

Example code for deploying GPU workloads on ECS

Project "Springfield"

Docker Images for the GPU Cluster

Detailed Analysis Traces for GPU-Disaggregated Deep Learning Recommendation Models

Detailed Analysis Traces for AI jobs leveraging spot GPU resources

Root repo for ACM GPU Cluster

A simple tool to expose only specified number of GPUs with desired memory to Tensorflow

Template for custom docker files on the gpu cluster

NCCL pairwise communication benchmarking and topology visualization on multi‑node GPU clusters.

ProcPlan is a dependency-free GPU resource planner built entirely on the Python standard library and SQLite

Slack bot for monitoring GPU usage on a server.