SymbioticLab / SalusLinks
Fine-grained GPU sharing primitives
☆143Updated 3 weeks ago
Alternatives and similar repositories for Salus
Users that are interested in Salus are comparing it to the libraries listed below
Sorting:
- PipeSwitch: Fast Pipelined Context Switching for Deep Learning Applications☆126Updated 3 years ago
- Tiresias is a GPU cluster manager for distributed deep learning training.☆158Updated 5 years ago
- Code for "Heterogenity-Aware Cluster Scheduling Policies for Deep Learning Workloads", which appeared at OSDI 2020☆128Updated last year
- Model-less Inference Serving☆91Updated last year
- ☆83Updated 2 months ago
- ☆192Updated 5 years ago
- An Efficient Dynamic Resource Scheduler for Deep Learning Clusters☆42Updated 7 years ago
- GPU-scheduler-for-deep-learning☆210Updated 4 years ago
- ☆38Updated 4 years ago
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆120Updated last year
- Synthesizer for optimal collective communication algorithms☆116Updated last year
- ☆50Updated 8 months ago
- ☆37Updated last month
- ☆51Updated 2 years ago
- Helios Traces from SenseTime☆57Updated 2 years ago
- RDMA and SHARP plugins for nccl library☆201Updated 2 months ago
- ☆55Updated 4 years ago
- An interference-aware scheduler for fine-grained GPU sharing☆143Updated 6 months ago
- A GPU-accelerated DNN inference serving system that supports instant kernel preemption and biased concurrent execution in GPU scheduling.☆42Updated 3 years ago
- A Deep Learning Cluster Scheduler☆39Updated 4 years ago
- An Efficient Pipelined Data Parallel Approach for Training Large Model☆77Updated 4 years ago
- Analyze network performance in distributed training☆18Updated 4 years ago
- NCCL Profiling Kit☆141Updated last year
- Artifacts for our NSDI'23 paper TGS☆84Updated last year
- Artifacts for our ASPLOS'23 paper ElasticFlow☆52Updated last year
- ☆24Updated 2 years ago
- 🔮 Execution time predictions for deep neural network training iterations across different GPUs.☆63Updated 2 years ago
- Paella: Low-latency Model Serving with Virtualized GPU Scheduling☆61Updated last year
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆84Updated 2 years ago
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆51Updated 2 years ago