Funatiq / gossip
gossip: Efficient Communication Primitives for Multi-GPU Systems
☆58Updated 2 years ago
Alternatives and similar repositories for gossip:
Users that are interested in gossip are comparing it to the libraries listed below
- Kernel Fusion and Runtime Compilation Based on NNVM☆70Updated 8 years ago
- pytorch ucc plugin☆18Updated 3 years ago
- HierarchicalKV is a part of NVIDIA Merlin and provides hierarchical key-value storage to meet RecSys requirements. The key capability of…☆138Updated this week
- ☆23Updated 3 years ago
- Repository for SysML19 Artifacts Evaluation☆53Updated 5 years ago
- ☆74Updated 2 years ago
- A Distributed Multi-GPU System for Fast Graph Processing☆65Updated 6 years ago
- High-performance, GPU-aware communication library☆84Updated last month
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆115Updated last year
- Symbolic Expression and Statement Module for new DSLs☆205Updated 4 years ago
- A GPU-Accelerated In-Memory Key-Value Store (AWS-focused fork)☆28Updated 7 years ago
- MLIR-based partitioning system☆60Updated this week
- Synthesizer for optimal collective communication algorithms☆103Updated 10 months ago
- A Library for fast Hash Tables on GPUs☆114Updated 2 years ago
- An experimental ahead of time compiler for Relay.☆50Updated 4 years ago
- NCCL Profiling Kit☆127Updated 7 months ago
- BytePS examples (Vision, NLP, GAN, etc)☆19Updated 2 years ago
- Machine Learning System☆14Updated 4 years ago
- An IR for efficiently simulating distributed ML computation.☆27Updated last year
- CUPTI GPU Profiler☆37Updated 5 years ago
- Enhanced networking support for TensorFlow. Maintained by SIG-networking.☆98Updated 3 years ago
- GPUDirect Async support for IB Verbs☆99Updated 2 years ago
- Documentation for StreamExecutor open source proposal☆83Updated 8 years ago
- this is the release repository of superneurons☆52Updated 4 years ago
- Python bindings for NVTX☆66Updated last year
- Pytorch process group third-party plugin for UCC☆20Updated 10 months ago
- Fairring (FAIR + Herring) is a plug-in for PyTorch that provides a process group for distributed training that outperforms NCCL at large …☆64Updated 2 years ago
- An Efficient Pipelined Data Parallel Approach for Training Large Model☆73Updated 4 years ago
- ☆23Updated 2 months ago
- A framework for pipelined computing on GPU☆29Updated 5 years ago