AI 中文总结
本文针对科学家在多样硬件上做研究的情况，提出一种软件生态系统，通过模块化、统一命令行界面，让用户在不同资源上有相同开发环境运行工作流，已在项目中成功测试且可扩展至其他项目。
AI 中文摘要
科学家在从笔记本电脑、工作站到超级计算机和云资源等各种硬件上进行研究很常见，标准场景需要多种资源组合。本文描述了一种软件生态系统，能让用户在不同计算资源上运行工作流时依赖相同开发环境。还介绍了模块化、统一的命令行界面，可跨不同硬件平台与用户工作流交互。该软件生态系统已在欧盟H2020项目plan4res中成功测试，可扩展到其他有类似需求的项目。
英文摘要
It is normal for scientists to perform their research on a diverse set of hardware, ranging from laptops and workstations to supercomputers and cloud resources. The standard scenario requires a mix of these resources. In this paper we describe a software ecosystem that enables users to rely on the same development environment for running their workflows across the different computational resources. We describe a modular, unified command-line interface that allows for the interaction with a user-workflow across diverse hardware platform. The software ecosystem has been successfully tested as part of the plan4res EU H2020 project. It can be extended to other projects with similar requirements, so that they can benefit from the same approach for executing computational workflows.
Comments13 pages, 4 figures
Journal refHigh Performance Computing. ISC High Performance 2024 International Workshops. 402-414, Springer Nature Switzerland (2025)
DOI：10.1007/978-3-031-73716-9_28
