HuaizhengZhang / MIGProfiler
Multi-Instance-GPU profiling tool
☆56Updated last year
Alternatives and similar repositories for MIGProfiler:
Users that are interested in MIGProfiler are comparing it to the libraries listed below
- SOTA Learning-augmented Systems☆34Updated 2 years ago
- ☆24Updated last year
- Fast and Efficient Model Serving Using Multi-GPUs with Direct-Host-Access (ACM EuroSys '23)☆55Updated 10 months ago
- An interference-aware scheduler for fine-grained GPU sharing☆122Updated 2 weeks ago
- DISB is a new DNN inference serving benchmark with diverse workloads and models, as well as real-world traces.☆53Updated 5 months ago
- ☆44Updated last month
- Fine-grained GPU sharing primitives☆141Updated 4 years ago
- A GPU-accelerated DNN inference serving system that supports instant kernel preemption and biased concurrent execution in GPU scheduling.☆40Updated 2 years ago
- A Cluster-Wide Model Manager to Accelerate DNN Training via Automated Training Warmup☆34Updated 2 years ago
- Artifacts for our ASPLOS'23 paper ElasticFlow☆53Updated 9 months ago
- ☆37Updated 3 years ago
- PipeSwitch: Fast Pipelined Context Switching for Deep Learning Applications☆127Updated 2 years ago
- ☆35Updated 4 years ago
- LLM serving cluster simulator☆92Updated 9 months ago
- ☆53Updated 4 years ago
- Model-less Inference Serving☆84Updated last year
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆81Updated last year
- A resilient distributed training framework☆88Updated 10 months ago
- Code for "Heterogenity-Aware Cluster Scheduling Policies for Deep Learning Workloads", which appeared at OSDI 2020☆126Updated 6 months ago
- A Deep Learning Cluster Scheduler☆37Updated 4 years ago
- SpotServe: Serving Generative Large Language Models on Preemptible Instances☆111Updated 11 months ago
- Paella: Low-latency Model Serving with Virtualized GPU Scheduling☆58Updated 9 months ago
- Lucid: A Non-Intrusive, Scalable and Interpretable Scheduler for Deep Learning Training Jobs☆52Updated last year
- ☆48Updated 2 years ago
- Bamboo is a system for running large pipeline-parallel DNNs affordably, reliably, and efficiently using spot instances.☆48Updated 2 years ago
- ☆35Updated 7 months ago
- A ChatGPT(GPT-3.5) & GPT-4 Workload Trace to Optimize LLM Serving Systems☆145Updated 4 months ago
- (NeurIPS 2022) Automatically finding good model-parallel strategies, especially for complex models and clusters.☆37Updated 2 years ago
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆60Updated 8 months ago
- ☆16Updated 2 years ago