Simplify HPC and Batch workloads on Azure
AMD RAD's multi-GPU Triton-based framework for seamless multi-GPU programming
DINOMO: An Elastic, Scalable, High-Performance Key-Value Store for Disaggregated Persistent Memory (PVLDB 2022, VLDB 2023)
Lumina is a user-friendly tool to test the correctness and performance of hardware network stacks.
A high-fidelity NS-3 simulator for cross–data center RDMA networks, built on High-Precision-Congestion-Control and conweave-ns3. It extends single-DC RDMA load balancing with multi–data center topologies, message-aware FEC with hierarchical interleaving, and EdgeCNP for improved congestion control across wide-area links.
Early-warning bottleneck profiler for GPU training nodes: GPU + RDMA fabric telemetry with job-level classification
AI & Infra 오타쿠를 위한 감성 노트 (AI＆インフラオタクのための感性ノート)
Complete setup guide for a 2-node NVIDIA DGX Spark cluster — distributed training, CUDA inference with EXO, NCCL tuning for Grace Blackwell, NVMe-TCP shared storage, and 200 Gb/s direct fabric networking.
Python RDMA P2P communication library based on OpenUCX.
Evidence-based NCCL failure diagnosis, topology linting, and tuning advice for GPU clusters.
A reproducible benchmark suite for NCCL GPU collective communications over RDMA fabric, targeting AI/HPC workloads.
A pedagogically-structured curriculum covering modern datacenter virtualization, networking, and infrastructure technologies
Packet-level simulator (ns-3 + RoCEv2/DCQCN/PFC/ECN) comparing fat-tree vs rail-optimized topologies for AI training fabrics. Validated against NCCL on real GPUs.
Python interface to the Linux RDMA stack
A hybrid testbed for evaluating top open-source LLMs (like gpt-oss-20b and Llama 3.3) on local, cloud GPUs, and AWS Inferentia2/Trainium instances, focusing on vLLM optimization, capacity management, kernel bypass, hardware-software co-design, as well as supporting infrastructure such as NCCL, RDMA, NVMeoF.
🌊 Distill flow maps from pre-trained models with FreeFlow, enhancing sampling speed and quality without requiring additional data.
Add a description, image, and links to the rdma topic page so that developers can more easily learn about it.
To associate your repository with the rdma topic, visit your repo's landing page and select "manage topics."