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:
- PipeSwitch: Fast Pipelined Context Switching for Deep Learning Applications☆126Updated 3 years ago
- ☆77Updated 4 years ago
- ☆83Updated 2 years ago
- ☆80Updated 6 months ago
- A baseline repository of Auto-Parallelism in Training Neural Networks☆147Updated 3 years ago
- Fine-grained GPU sharing primitives☆147Updated 3 months ago
- PET: Optimizing Tensor Programs with Partially Equivalent Transformations and Automated Corrections☆121Updated 3 years ago
- FTPipe and related pipeline model parallelism research.☆43Updated 2 years ago
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆91Updated 2 years ago
- An experimental parallel training platform☆56Updated last year
- Synthesizer for optimal collective communication algorithms☆119Updated last year
- ☆82Updated 4 months ago
- ☆25Updated 2 years ago
- this is the release repository of superneurons☆54Updated 4 years ago
- Microsoft Collective Communication Library☆372Updated 2 years ago
- Compiler for Dynamic Neural Networks☆46Updated 2 years ago
- nnScaler: Compiling DNN models for Parallel Training☆118Updated last month
- Model-less Inference Serving☆91Updated 2 years ago
- ☆38Updated 4 months ago
- Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.☆67Updated 7 months ago
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆53Updated 2 years ago
- Paella: Low-latency Model Serving with Virtualized GPU Scheduling☆62Updated last year
- ☆145Updated 9 months ago
- 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
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆218Updated 3 months ago
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆122Updated last year
- ☆393Updated 3 years ago
- Artifacts for our ASPLOS'23 paper ElasticFlow☆55Updated last year
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆63Updated last year