eniac / paellaLinks
Paella: Low-latency Model Serving with Virtualized GPU Scheduling
☆65Updated last year
Alternatives and similar repositories for paella
Users that are interested in paella are comparing it to the libraries listed below
Sorting:
- REEF is a GPU-accelerated DNN inference serving system that enables instant kernel preemption and biased concurrent execution in GPU sche…☆103Updated 2 years ago
- ☆83Updated 3 years ago
- ☆53Updated 11 months ago
- A GPU-accelerated DNN inference serving system that supports instant kernel preemption and biased concurrent execution in GPU scheduling.☆43Updated 3 years ago
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆63Updated last year
- An interference-aware scheduler for fine-grained GPU sharing☆153Updated this week
- ☆23Updated 2 years ago
- ☆38Updated 5 months ago
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆91Updated 2 years ago
- Thunder Research Group's Collective Communication Library☆42Updated 4 months ago
- NCCL Profiling Kit☆149Updated last year
- [ACM EuroSys 2023] Fast and Efficient Model Serving Using Multi-GPUs with Direct-Host-Access☆57Updated 3 months ago
- Compiler for Dynamic Neural Networks☆46Updated 2 years ago
- A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.☆116Updated 6 months ago
- PipeSwitch: Fast Pipelined Context Switching for Deep Learning Applications☆126Updated 3 years ago
- Artifacts for our ASPLOS'23 paper ElasticFlow☆55Updated last year
- A lightweight design for computation-communication overlap.☆188Updated last month
- DISB is a new DNN inference serving benchmark with diverse workloads and models, as well as real-world traces.☆57Updated last year
- Fine-grained GPU sharing primitives☆147Updated 4 months ago
- A tool for examining GPU scheduling behavior.☆89Updated last year
- Artifacts for our NSDI'23 paper TGS☆90Updated last year
- ☆51Updated 2 years ago
- ☆26Updated 9 months ago
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆54Updated 2 years ago
- Synthesizer for optimal collective communication algorithms☆121Updated last year
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆144Updated 2 months ago
- Microsoft Collective Communication Library☆66Updated last year
- High performance Transformer implementation in C++.☆142Updated 10 months ago
- SOTA Learning-augmented Systems☆37Updated 3 years ago
- ☆47Updated 11 months ago