zhuangwang93 / Espresso
Hi-Speed DNN Training with Espresso: Unleashing the Full Potential of Gradient Compression with Near-Optimal Usage Strategies (EuroSys '23)
☆15Updated last year
Related projects ⓘ
Alternatives and complementary repositories for Espresso
- Cupcake: A Compression Scheduler for Scalable Communication-Efficient Distributed Training (MLSys '23)☆9Updated last year
- Ok-Topk is a scheme for distributed training with sparse gradients. Ok-Topk integrates a novel sparse allreduce algorithm (less than 6k c…☆23Updated last year
- Artifacts for our SIGCOMM'22 paper Muri☆40Updated 10 months ago
- ☆41Updated last year
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆47Updated last year
- ☆14Updated 5 months ago
- ☆23Updated last year
- Proteus: A High-Throughput Inference-Serving System with Accuracy Scaling☆8Updated 8 months ago
- ☆51Updated 3 years ago
- ☆48Updated last year
- THC: Accelerating Distributed Deep Learning Using Tensor Homomorphic Compression☆14Updated 3 months ago
- Aequitas enables RPC-level QoS in datacenter networks.☆16Updated 2 years ago
- ☆23Updated 2 years ago
- Official repository for the paper DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines☆14Updated 11 months ago
- Artifacts for our ASPLOS'23 paper ElasticFlow☆52Updated 6 months ago
- Analyze network performance in distributed training☆16Updated 4 years ago
- A Cluster-Wide Model Manager to Accelerate DNN Training via Automated Training Warmup☆34Updated last year
- Code for "Shockwave: Fair and Efficient Cluster Scheduling for Dynamic Adaptation in Machine Learning" [NSDI '23]☆38Updated last year
- Helios Traces from SenseTime☆48Updated 2 years ago
- ☆33Updated 4 months ago
- TACCL: Guiding Collective Algorithm Synthesis using Communication Sketches☆63Updated last year
- ☆69Updated last year
- ☆37Updated 3 years ago
- Efficient Interactive LLM Serving with Proxy Model-based Sequence Length Prediction | A tiny model can tell you the verbosity of an LLM (…☆22Updated 5 months ago
- REEF is a GPU-accelerated DNN inference serving system that enables instant kernel preemption and biased concurrent execution in GPU sche…☆85Updated last year
- Justitia provides RDMA isolation between applications with diverse requirements.☆38Updated 2 years ago
- ☆13Updated 2 years ago
- ☆12Updated last year
- [NSDI 2023] TopoOpt: Optimizing the Network Topology for Distributed DNN Training☆26Updated 2 months ago
- ☆15Updated 4 months ago