SamsungLabs / FastFlow
FastFlow is a system that automatically detects CPU bottlenecks in deep learning training pipelines and resolves the bottlenecks with data pipeline offloading to remote resources .
☆25Updated last year
Related projects ⓘ
Alternatives and complementary repositories for FastFlow
- Fast and Efficient Model Serving Using Multi-GPUs with Direct-Host-Access (ACM EuroSys '23)☆54Updated 7 months ago
- ☆23Updated 2 years ago
- ☆41Updated last year
- ☆14Updated 5 months ago
- ☆23Updated last year
- DISB is a new DNN inference serving benchmark with diverse workloads and models, as well as real-world traces.☆54Updated 3 months ago
- ☆23Updated last year
- LLM serving cluster simulator☆81Updated 6 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
- Compiler for Dynamic Neural Networks☆43Updated last year
- ☆35Updated 3 years ago
- ☆37Updated 3 years ago
- 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
- Hi-Speed DNN Training with Espresso: Unleashing the Full Potential of Gradient Compression with Near-Optimal Usage Strategies (EuroSys '2…☆15Updated last year
- Paella: Low-latency Model Serving with Virtualized GPU Scheduling☆57Updated 6 months ago
- SHADE: Enable Fundamental Cacheability for Distributed Deep Learning Training☆29Updated last year
- ☆73Updated last year
- Artifacts for our ASPLOS'23 paper ElasticFlow☆52Updated 6 months ago
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆133Updated last month
- Proteus: A High-Throughput Inference-Serving System with Accuracy Scaling☆8Updated 8 months ago
- A GPU-accelerated DNN inference serving system that supports instant kernel preemption and biased concurrent execution in GPU scheduling.☆39Updated 2 years ago
- ☆51Updated 3 years ago
- A resilient distributed training framework☆85Updated 7 months ago
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆57Updated 5 months ago
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆78Updated last year
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆47Updated last year
- ☆33Updated last year
- ☆48Updated last year
- ☆16Updated last year
- An interference-aware scheduler for fine-grained GPU sharing☆111Updated 6 months ago