SymbioticLab / OobleckLinks
A resilient distributed training framework
☆95Updated last year
Alternatives and similar repositories for Oobleck
Users that are interested in Oobleck are comparing it to the libraries listed below
Sorting:
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆213Updated 2 months ago
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆129Updated last year
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆88Updated 2 years ago
- ☆74Updated this week
- Stateful LLM Serving☆86Updated 7 months ago
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆62Updated last year
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆51Updated 2 years ago
- An interference-aware scheduler for fine-grained GPU sharing☆147Updated 8 months ago
- [ICML 2024] Serving LLMs on heterogeneous decentralized clusters.☆30Updated last year
- ☆26Updated last year
- An experimental parallel training platform☆54Updated last year
- Dynamic resources changes for multi-dimensional parallelism training☆29Updated 2 months ago
- A framework for generating realistic LLM serving workloads☆71Updated 2 weeks ago
- LLM serving cluster simulator☆116Updated last year
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆186Updated last year
- Compiler for Dynamic Neural Networks☆46Updated last year
- ☆124Updated 11 months ago
- nnScaler: Compiling DNN models for Parallel Training☆118Updated 3 weeks ago
- (NeurIPS 2022) Automatically finding good model-parallel strategies, especially for complex models and clusters.☆41Updated 2 years ago
- Microsoft Collective Communication Library☆66Updated 11 months ago
- PyTorch implementation of paper "Response Length Perception and Sequence Scheduling: An LLM-Empowered LLM Inference Pipeline".☆92Updated 2 years ago
- Official repository for the paper DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines☆20Updated last year
- Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.☆67Updated 7 months ago
- A Cluster-Wide Model Manager to Accelerate DNN Training via Automated Training Warmup☆35Updated 2 years ago
- ☆77Updated 3 years ago
- ☆75Updated 4 years ago
- Artifacts for our ASPLOS'23 paper ElasticFlow☆53Updated last year
- ☆25Updated 2 years ago
- ☆23Updated last year
- ☆21Updated last year