AlibabaPAI / DAPPLELinks
An Efficient Pipelined Data Parallel Approach for Training Large Model
☆76Updated 4 years ago
Alternatives and similar repositories for DAPPLE
Users that are interested in DAPPLE are comparing it to the libraries listed below
Sorting:
- ☆83Updated 2 years ago
- ☆75Updated 4 years ago
- PipeSwitch: Fast Pipelined Context Switching for Deep Learning Applications☆126Updated 3 years ago
- Fine-grained GPU sharing primitives☆146Updated 2 months ago
- Synthesizer for optimal collective communication algorithms☆118Updated last year
- ☆81Updated 5 months ago
- Microsoft Collective Communication Library☆367Updated 2 years ago
- FTPipe and related pipeline model parallelism research.☆43Updated 2 years ago
- A baseline repository of Auto-Parallelism in Training Neural Networks☆147Updated 3 years ago
- nnScaler: Compiling DNN models for Parallel Training☆118Updated last month
- An experimental parallel training platform☆54Updated last year
- Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.☆67Updated 7 months ago
- PET: Optimizing Tensor Programs with Partially Equivalent Transformations and Automated Corrections☆122Updated 3 years ago
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆88Updated 2 years ago
- this is the release repository of superneurons☆54Updated 4 years ago
- ☆145Updated 8 months ago
- ☆83Updated 4 months ago
- ☆393Updated 2 years ago
- ☆25Updated 2 years ago
- Model-less Inference Serving☆92Updated last year
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆121Updated last year
- Boost hardware utilization for ML training workloads via Inter-model Horizontal Fusion☆32Updated last year
- 🔮 Execution time predictions for deep neural network training iterations across different GPUs.☆62Updated 2 years ago
- ☆42Updated 2 years ago
- [MLSys 2021] IOS: Inter-Operator Scheduler for CNN Acceleration☆199Updated 3 years ago
- ☆38Updated 3 months ago
- Compiler for Dynamic Neural Networks☆46Updated last year
- ☆154Updated last year
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆51Updated 2 years ago
- Dynamic Tensor Rematerialization prototype (modified PyTorch) and simulator. Paper: https://arxiv.org/abs/2006.09616☆132Updated 2 years ago