google / nccl-fastsocketLinks
NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.
☆120Updated 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☆208Updated 3 weeks ago
- NCCL Profiling Kit☆145Updated last year
- Fine-grained GPU sharing primitives☆144Updated 2 months ago
- Microsoft Collective Communication Library☆360Updated 2 years ago
- PArametrized Recommendation and Ai Model benchmark is a repository for development of numerous uBenchmarks as well as end to end nets for…☆151Updated 3 weeks ago
- This is a plugin which lets EC2 developers use libfabric as network provider while running NCCL applications.☆186Updated this week
- Synthesizer for optimal collective communication algorithms☆117Updated last year
- ☆373Updated last year
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆418Updated this week
- NVIDIA Resiliency Extension is a python package for framework developers and users to implement fault-tolerant features. It improves the …☆223Updated last week
- Ultra and Unified CCL☆567Updated this week
- Microsoft Collective Communication Library☆66Updated 10 months ago
- pytorch ucc plugin☆23Updated 4 years ago
- ☆46Updated 9 months ago
- GPU-scheduler-for-deep-learning☆210Updated 4 years ago
- NVIDIA Inference Xfer Library (NIXL)☆648Updated this week
- PipeSwitch: Fast Pipelined Context Switching for Deep Learning Applications☆127Updated 3 years ago
- NVIDIA NCCL Tests for Distributed Training☆111Updated last week
- An Efficient Pipelined Data Parallel Approach for Training Large Model☆76Updated 4 years ago
- ☆84Updated 3 months ago
- ☆60Updated 8 months ago
- GPUDirect Async support for IB Verbs☆130Updated 2 years ago
- HierarchicalKV is a part of NVIDIA Merlin and provides hierarchical key-value storage to meet RecSys requirements. The key capability of…☆171Updated this week
- Paella: Low-latency Model Serving with Virtualized GPU Scheduling☆62Updated last year
- Unified Collective Communication Library☆277Updated this week
- A tool for bandwidth measurements on NVIDIA GPUs.☆536Updated 5 months ago
- Offline optimization of your disaggregated Dynamo graph☆70Updated this week
- TACCL: Guiding Collective Algorithm Synthesis using Communication Sketches☆75Updated 2 years ago
- ☆83Updated 2 years ago
- Tiresias is a GPU cluster manager for distributed deep learning training.☆160Updated 5 years ago