google / nccl-fastsocketLinks
NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.
☆121Updated last year
Alternatives and similar repositories for nccl-fastsocket
Users that are interested in nccl-fastsocket are comparing it to the libraries listed below
Sorting:
- RDMA and SHARP plugins for nccl library☆201Updated this week
- NCCL Profiling Kit☆143Updated last year
- Microsoft Collective Communication Library☆360Updated last year
- PArametrized Recommendation and Ai Model benchmark is a repository for development of numerous uBenchmarks as well as end to end nets for…☆149Updated last week
- Fine-grained GPU sharing primitives☆144Updated last month
- ☆371Updated last year
- Synthesizer for optimal collective communication algorithms☆116Updated last year
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆412Updated last week
- This is a plugin which lets EC2 developers use libfabric as network provider while running NCCL applications.☆184Updated last week
- NVIDIA Resiliency Extension is a python package for framework developers and users to implement fault-tolerant features. It improves the …☆216Updated this week
- Ultra and Unified CCL☆530Updated this week
- GPU-scheduler-for-deep-learning☆210Updated 4 years ago
- ☆46Updated 8 months ago
- Microsoft Collective Communication Library☆66Updated 9 months ago
- An Efficient Pipelined Data Parallel Approach for Training Large Model☆77Updated 4 years ago
- NVIDIA NCCL Tests for Distributed Training☆110Updated last week
- NVIDIA Inference Xfer Library (NIXL)☆603Updated this week
- pytorch ucc plugin☆23Updated 4 years ago
- PipeSwitch: Fast Pipelined Context Switching for Deep Learning Applications☆126Updated 3 years ago
- Paella: Low-latency Model Serving with Virtualized GPU Scheduling☆62Updated last year
- A validation and profiling tool for AI infrastructure☆331Updated this week
- HierarchicalKV is a part of NVIDIA Merlin and provides hierarchical key-value storage to meet RecSys requirements. The key capability of…☆169Updated this week
- ☆84Updated 2 months ago
- Thunder Research Group's Collective Communication Library☆41Updated 2 months ago
- GPUDirect Async support for IB Verbs☆130Updated 2 years ago
- A tool for bandwidth measurements on NVIDIA GPUs.☆527Updated 4 months ago
- ☆81Updated 2 years ago
- Splits single Nvidia GPU into multiple partitions with complete compute and memory isolation (wrt to performace) between the partitions☆160Updated 6 years ago
- example code for using DC QP for providing RDMA READ and WRITE operations to remote GPU memory☆142Updated last year
- TACCL: Guiding Collective Algorithm Synthesis using Communication Sketches☆74Updated 2 years ago