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:
- ☆62Updated 11 months ago
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆171Updated 7 months ago
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆121Updated last year
- Stateful LLM Serving☆70Updated 2 months ago
- ☆99Updated 6 months ago
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆81Updated last year
- PyTorch implementation of paper "Response Length Perception and Sequence Scheduling: An LLM-Empowered LLM Inference Pipeline".☆86Updated 2 years ago
- Dynamic resources changes for multi-dimensional parallelism training☆25Updated 6 months ago
- nnScaler: Compiling DNN models for Parallel Training☆113Updated last month
- An interference-aware scheduler for fine-grained GPU sharing☆137Updated 4 months ago
- LLM serving cluster simulator☆100Updated last year
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆61Updated 11 months ago
- ☆83Updated 3 years ago
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆157Updated 8 months ago
- An experimental parallel training platform☆54Updated last year
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆50Updated 2 years ago
- ☆23Updated 10 months ago
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆39Updated 3 weeks ago
- Ultra | Ultimate | Unified CCL☆87Updated this week
- Microsoft Collective Communication Library☆65Updated 6 months ago
- [ICML 2024] Serving LLMs on heterogeneous decentralized clusters.☆25Updated last year
- High performance Transformer implementation in C++.☆124Updated 4 months ago
- A lightweight design for computation-communication overlap.☆132Updated 3 weeks ago
- Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.☆63Updated 2 months ago
- ☆46Updated 11 months ago
- ☆73Updated 4 years ago
- [ATC '24] Metis: Fast automatic distributed training on heterogeneous GPUs (https://www.usenix.org/conference/atc24/presentation/um)☆26Updated 6 months ago
- Automated Parallelization System and Infrastructure for Multiple Ecosystems☆79Updated 6 months ago
- [NeurIPS 2024] Efficient LLM Scheduling by Learning to Rank☆48Updated 6 months ago
- Official repository for the paper DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines☆19Updated last year