facebookresearch / fairring
Fairring (FAIR + Herring) is a plug-in for PyTorch that provides a process group for distributed training that outperforms NCCL at large scales
☆63Updated 2 years ago
Related projects ⓘ
Alternatives and complementary repositories for fairring
- ☆44Updated last year
- System for automated integration of deep learning backends.☆48Updated 2 years ago
- A Python library transfers PyTorch tensors between CPU and NVMe☆98Updated last week
- ☆140Updated last year
- pytorch-profiler☆50Updated last year
- Benchmark scripts for TVM☆73Updated 2 years ago
- ☆48Updated 8 months ago
- Memory Optimizations for Deep Learning (ICML 2023)☆60Updated 8 months ago
- PipeTransformer: Automated Elastic Pipelining for Distributed Training of Large-scale Models. ICML 2021☆55Updated 3 years ago
- FTPipe and related pipeline model parallelism research.☆41Updated last year
- Training neural networks in TensorFlow 2.0 with 5x less memory☆129Updated 2 years ago
- extensible collectives library in triton☆71Updated last month
- A performant, memory-efficient checkpointing library for PyTorch applications, designed with large, complex distributed workloads in mind…☆146Updated 2 weeks ago
- Research and development for optimizing transformers☆125Updated 3 years ago
- PET: Optimizing Tensor Programs with Partially Equivalent Transformations and Automated Corrections☆114Updated 2 years ago
- A library for syntactically rewriting Python programs, pronounced (sinner).☆70Updated 2 years ago
- (NeurIPS 2022) Automatically finding good model-parallel strategies, especially for complex models and clusters.☆34Updated 2 years ago
- ☆55Updated 5 months ago
- Benchmark code for the "Online normalizer calculation for softmax" paper☆59Updated 6 years ago
- ☆167Updated 4 months ago
- ☆23Updated 10 months ago
- ☆11Updated 3 years ago
- DLPack for Tensorflow☆36Updated 4 years ago
- This repository contains the results and code for the MLPerf™ Training v1.0 benchmark.☆37Updated 8 months ago
- Home for OctoML PyTorch Profiler☆107Updated last year
- ☆65Updated 3 years ago
- A tensor-aware point-to-point communication primitive for machine learning☆249Updated last year
- Odysseus: Playground of LLM Sequence Parallelism☆57Updated 5 months ago
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆90Updated 4 months ago
- Automated Parallelization System and Infrastructure for Multiple Ecosystems☆75Updated this week