gpudirect / libgdsync
GPUDirect Async support for IB Verbs
☆110Updated 2 years ago
Alternatives and similar repositories for libgdsync:
Users that are interested in libgdsync are comparing it to the libraries listed below
- ☆23Updated 3 years ago
- Magnum IO community repo☆89Updated 3 months ago
- GPUDirect example☆59Updated 3 years ago
- Provides a set of benchmarks that can be used to measure the memory bandwidth performance of CPU's☆89Updated last year
- oneAPI Collective Communications Library (oneCCL)☆232Updated 2 weeks ago
- NCCL Profiling Kit☆130Updated 9 months ago
- Sandia OpenSHMEM is an implementation of the OpenSHMEM specification over multiple Networking APIs, including Portals 4, the Open Fabric …☆68Updated last week
- pytorch ucc plugin☆21Updated 3 years ago
- Pytorch process group third-party plugin for UCC☆20Updated last year
- rocSHMEM intra-kernel networking runtime for AMD dGPUs on the ROCm platform.☆76Updated last week
- RDMA and SHARP plugins for nccl library☆189Updated 2 weeks ago
- Mellanox libibverbs☆64Updated 5 years ago
- Unified Collective Communication Library☆248Updated last week
- example code for using DC QP for providing RDMA READ and WRITE operations to remote GPU memory☆127Updated 8 months ago
- verbs profiling library☆22Updated last year
- Simple message passing library☆23Updated 6 years ago
- A hierarchical collective communications library with portable optimizations☆33Updated 4 months ago
- oneAPI Level Zero Conformance & Performance test content☆49Updated this week
- RCCL Performance Benchmark Tests☆64Updated last week
- ☆339Updated last year
- Linux Cross-Memory Attach☆92Updated 7 months ago
- TransferBench is a utility capable of benchmarking simultaneous copies between user-specified devices (CPUs/GPUs)☆39Updated last week
- Synthesizer for optimal collective communication algorithms☆105Updated last year
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆116Updated last year
- OFI Programmer's Guide☆52Updated 2 years ago
- ROC profiler library. Profiling with perf-counters and derived metrics.☆141Updated last week
- This is a plugin which lets EC2 developers use libfabric as network provider while running NCCL applications.☆168Updated this week
- A GPU-Accelerated In-Memory Key-Value Store (AWS-focused fork)☆28Updated 7 years ago
- NVIDIA GPUDirect Storage Driver☆240Updated 4 months ago
- Automatic virtualization of (general) accelerators.☆42Updated 2 years ago