NVIDIA / nvidia-resiliency-extLinks
NVIDIA Resiliency Extension is a python package for framework developers and users to implement fault-tolerant features. It improves the effective training time by minimizing the downtime due to failures and interruptions.
☆187Updated this week
Alternatives and similar repositories for nvidia-resiliency-ext
Users that are interested in nvidia-resiliency-ext are comparing it to the libraries listed below
Sorting:
- A library to analyze PyTorch traces.☆391Updated this week
- NVIDIA Inference Xfer Library (NIXL)☆459Updated this week
- Ultra and Unified CCL☆390Updated this week
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆117Updated last year
- Perplexity GPU Kernels☆395Updated last month
- ☆94Updated 6 months ago
- Applied AI experiments and examples for PyTorch☆278Updated last month
- QuickReduce is a performant all-reduce library designed for AMD ROCm that supports inline compression.☆29Updated 3 months ago
- ☆51Updated 4 months ago
- ☆310Updated 10 months ago
- Zero Bubble Pipeline Parallelism☆403Updated 2 months ago
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆385Updated this week
- CUDA checkpoint and restore utility☆345Updated 5 months ago
- Fast low-bit matmul kernels in Triton☆327Updated this week
- NCCL Profiling Kit☆139Updated last year
- A tool for bandwidth measurements on NVIDIA GPUs.☆474Updated 2 months ago
- Microsoft Collective Communication Library☆351Updated last year
- NVIDIA NCCL Tests for Distributed Training☆97Updated 2 weeks ago
- 🚀 Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.☆205Updated this week
- PArametrized Recommendation and Ai Model benchmark is a repository for development of numerous uBenchmarks as well as end to end nets for…☆147Updated last week
- This is a plugin which lets EC2 developers use libfabric as network provider while running NCCL applications.☆177Updated this week
- Fault tolerance for PyTorch (HSDP, LocalSGD, DiLoCo, Streaming DiLoCo)☆359Updated last week
- RDMA and SHARP plugins for nccl library☆197Updated 3 weeks ago
- extensible collectives library in triton☆87Updated 3 months ago
- Dynamic Memory Management for Serving LLMs without PagedAttention☆399Updated last month
- Synthesizer for optimal collective communication algorithms☆108Updated last year
- A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")☆343Updated this week
- A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.☆176Updated this week
- ☆214Updated last year
- Genai-bench is a powerful benchmark tool designed for comprehensive token-level performance evaluation of large language model (LLM) serv…☆161Updated this week