microsoft / SuperScaler
An experimental parallel training platform
☆54Updated 11 months ago
Alternatives and similar repositories for SuperScaler:
Users that are interested in SuperScaler are comparing it to the libraries listed below
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆148Updated 4 months ago
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆82Updated last year
- nnScaler: Compiling DNN models for Parallel Training☆101Updated last month
- ☆72Updated 3 years ago
- LLM serving cluster simulator☆93Updated 10 months ago
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆112Updated last year
- Compiler for Dynamic Neural Networks☆45Updated last year
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆60Updated 9 months ago
- ☆75Updated 2 years ago
- ☆52Updated 9 months ago
- Synthesizer for optimal collective communication algorithms☆104Updated 11 months ago
- Stateful LLM Serving☆46Updated this week
- Official repository for the paper DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines☆17Updated last year
- Artifacts for our ASPLOS'23 paper ElasticFlow☆52Updated 10 months ago
- Microsoft Collective Communication Library☆60Updated 3 months ago
- An Efficient Pipelined Data Parallel Approach for Training Large Model☆74Updated 4 years ago
- A resilient distributed training framework☆89Updated 11 months ago
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆49Updated 2 years ago
- ☆87Updated 4 months ago
- ☆23Updated last year
- (NeurIPS 2022) Automatically finding good model-parallel strategies, especially for complex models and clusters.☆38Updated 2 years ago
- Paella: Low-latency Model Serving with Virtualized GPU Scheduling☆58Updated 10 months ago
- TACCL: Guiding Collective Algorithm Synthesis using Communication Sketches☆71Updated last year
- Dynamic resources changes for multi-dimensional parallelism training☆24Updated 4 months ago
- Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines.☆60Updated last year
- REEF is a GPU-accelerated DNN inference serving system that enables instant kernel preemption and biased concurrent execution in GPU sche…☆92Updated 2 years ago
- DISB is a new DNN inference serving benchmark with diverse workloads and models, as well as real-world traces.☆53Updated 6 months ago
- [ICML 2024] Serving LLMs on heterogeneous decentralized clusters.☆22Updated 10 months ago
- ☆20Updated 8 months ago
- ☆53Updated 4 years ago