Welcome to the REANA Performance Evaluation project! This openlab summer student project focuses on assessing the performance of contemporary Kubernetes batch scheduling systems, with a particular emphasis on Kueue, a Kubernetes-native scheduling solution actively developed by the Kubernetes community.
REANA is an open-source platform designed for reproducible declarative data analyses on containerized compute clouds. With Kubernetes as its primary backend, it efficiently manages computational workflows through the inherent Kubernetes Job API, streamlining user job scheduling.
The central goal of this project is to evaluate the suitability of the Kueue scheduler within the REANA ecosystem. We aim to assess its performance, focusing on performance benchmarks and supplementary capabilities, such as equitable resource allocation, dynamic resource adaptability, and scalability under the load of thousands of workloads. Our evaluation actively contributes to the development of Kueue for reproducible computational research.
-
Performance Evaluation: We conduct rigorous performance testing to measure Kueue's efficiency and effectiveness for REANA workflows.
-
Resource Allocation: We assess Kueue's ability to allocate resources fairly among multiple concurrent workflows.
-
Dynamic Adaptability: We investigate how Kueue adapts to changing resource requirements during workflow execution.
We welcome contributions and collaboration from the open-source community. If you're interested in participating or have ideas to share, please feel free to reach out and get involved in this exciting project.
This project is licensed under the MIT License.