openucx / xcclLinks
☆24Updated 4 years ago
Alternatives and similar repositories for xccl
Users that are interested in xccl are comparing it to the libraries listed below
Sorting:
- GPUDirect Async support for IB Verbs☆120Updated 2 years ago
- pytorch ucc plugin☆22Updated 3 years ago
- Pytorch process group third-party plugin for UCC☆21Updated last year
- verbs profiling library☆22Updated last year
- A hierarchical collective communications library with portable optimizations☆35Updated 6 months ago
- Unified Collective Communication Library☆256Updated this week
- Sandia OpenSHMEM is an implementation of the OpenSHMEM specification over multiple Networking APIs, including Portals 4, the Open Fabric …☆70Updated 2 months ago
- RDMA and SHARP plugins for nccl library☆195Updated last week
- NCCL Profiling Kit☆137Updated 11 months ago
- High-performance, GPU-aware communication library☆87Updated 5 months ago
- A Micro-benchmarking Tool for HPC Networks☆30Updated 5 months ago
- gossip: Efficient Communication Primitives for Multi-GPU Systems☆59Updated 2 years ago
- Simple message passing library☆23Updated 6 years ago
- rocSHMEM intra-kernel networking runtime for AMD dGPUs on the ROCm platform.☆90Updated this week
- Linux Cross-Memory Attach☆94Updated 9 months ago
- Portals is a low-level network API for high-performance networking on high-performance computing systems developed by Sandia National Lab…☆39Updated 9 months ago
- MPI Microbenchmarks☆39Updated 9 years ago
- OpenSHMEM Application Programming Interface☆57Updated 7 months ago
- ☆72Updated 8 years ago
- Reference implementations of MLPerf™ HPC training benchmarks☆48Updated 3 months ago
- CUPTI GPU Profiler☆38Updated 6 years ago
- Magnum IO community repo☆95Updated last month
- Synthesizer for optimal collective communication algorithms☆108Updated last year
- Provides a set of benchmarks that can be used to measure the memory bandwidth performance of CPU's☆90Updated last year
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆116Updated last year
- Tartan: Evaluating Modern GPU Interconnect via a Multi-GPU Benchmark Suite☆65Updated 6 years ago
- TransferBench is a utility capable of benchmarking simultaneous copies between user-specified devices (CPUs/GPUs)☆40Updated last week
- Multi-GPU communication profiler and visualizer☆29Updated last year
- Prototype of OpenSHMEM for NVIDIA GPUs, developed as part of DoE Design Forward☆24Updated 7 years ago
- UnifyFS: A file system for burst buffers☆114Updated 3 months ago