icloud-ecnu / Prophet
Prophet is a predictable communication scheduling strategy to schedule the gradient transfer in an adequate order, with the aim of maximizing the GPU and network resource utilization.
☆16Updated last year
Alternatives and similar repositories for Prophet:
Users that are interested in Prophet are comparing it to the libraries listed below
- spotDNN is a heterogeneity-aware spot instance provisioning framework to provide predictable performance for DDNN training workloads in t…☆15Updated last year
- ebrowser, an energy-efficient and lightweight human interaction framework without degrading the user experience in mobile Web browsers.☆12Updated last year
- ☆12Updated last year
- iSpot is a lightweight and cost-effective instance provisioning framework for Directed Acyclic Graph (DAG)-style big data analytics, in …☆11Updated last year
- DelayStage is a simple yet effective stage delay scheduling strategy to interleave the cluster resources across the parallel stages, so a…☆14Updated last year
- Reading paper list for iCloud group☆13Updated last month
- ☆8Updated 3 years ago
- Tetris, a model predictive control (MPC)-based container scheduling strategy to judiciously make migration decisions for long-running con…☆24Updated 4 months ago
- Opara is a lightweight and resource-aware DNN Operator parallel scheduling framework to accelerate the execution of DNN inference on GPUs…☆20Updated 4 months ago
- λDNN is a cost-efficient function resource provisioning framework to minimize the monetary cost and guarantee the performance for DDNN tr…☆23Updated last year
- iGniter, an interference-aware GPU resource provisioning framework for achieving predictable performance of DNN inference in the cloud.☆37Updated 10 months ago
- ☆9Updated last year
- HeliosArtifact☆20Updated 2 years ago
- Lucid: A Non-Intrusive, Scalable and Interpretable Scheduler for Deep Learning Training Jobs☆53Updated last year
- ☆48Updated 4 months ago
- Helios Traces from SenseTime☆54Updated 2 years ago
- This repository is established to store personal notes and annotated papers during daily research.☆120Updated 2 weeks ago
- ☆16Updated last year
- ☆37Updated 3 years ago
- Artifacts for our ASPLOS'23 paper ElasticFlow☆51Updated 11 months ago
- Proteus: A High-Throughput Inference-Serving System with Accuracy Scaling☆11Updated last year
- ☆49Updated 2 years ago
- PipeEdge: Pipeline Parallelism for Large-Scale Model Inference on Heterogeneous Edge Devices☆31Updated last year
- ☆17Updated 2 years ago
- ☆42Updated 10 months ago
- AQUATOPE: QoS-and-Uncertainty-Aware Resource Management for Multi-Stage Serverless Workflows (ASPLOS'23)☆23Updated last year
- ☆10Updated 3 months ago
- ☆148Updated last year
- Compiler for Dynamic Neural Networks☆46Updated last year
- Artifact for "Shockwave: Fair and Efficient Cluster Scheduling for Dynamic Adaptation in Machine Learning" [NSDI '23]☆43Updated 2 years ago