POSE: Phase I: Constellation: A Pathway to Establish the STE||AR Open-Source Organization

POSE：第一阶段：Constellation：建立 STE||AR 开源组织的途径

• 批准号: 2229751
• 负责人: Mohammad Tohid
• 金额: $ 30万
• 依托单位: Louisiana State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-15 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2229751&HistoricalAwards=false
• 关键词: POSE; Phase; Constellation; Pathway; Establish

To further advance research in Artificial Intelligence (AI) and combatting climate change, society requires the efficient use of high-performance computing (HPC) resources. This project will lay out a plan to establish the STEllAR open-source organization that aims at making the use of HPC as accessible and efficient as possible. With the help of the community, STEllAR will develop a software ecosystem of users and developers by coordinating the development of tools and technologies targeting multiple science domains, delivering educational services and technical support for faster adoption of such technologies, and providing consultancy services to the community. The CONSTEllATION project will develop a roadmap to establish the STEllAR non-profit open-source organization. The STEllAR organization will strategize and oversee the development of a secure, standardized, and industrial strength HPX (High Performance ParalleX, a standard C++ library for concurrency and parallelism) runtime to overcome the challenges that arise in the ExaScale era and beyond. STEllAR will also help the community to build the application program interface (API), infrastructure, tools, and ecosystem around HPX. In addition, the organization will provide (1) application assurance through building developer tools and documentation, technical support, continuous testing and integration, and standardization; (2) professional support including training, custom packaging, and technology architecture; and (3) community services such as education, workforce development, and outreach. Rapid adoption of Asynchronous Many-Task runtime systems (AMTs) will help to increase the parallel efficiency and portability of modern high-performance applications and improve the energy efficiency of computing resources as society enters the ExaScale era. These capabilities will enable the scientific community to overcome the shortcomings of the commonly practiced Bulk Synchronization model and make way for developing a next generation of applications in AI, climate change, bio-informatics, and many other application domains of high societal impact. In addition, STEllAR will be a hub for outreach and education in the field of high-performance computing and related application areas.See:https://github.com/STEllAR-GROUP/hpxhttps://hpx-docs.stellar-group.org/latest/html/index.htmlhttps://stellar-group.orgThis award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

为了进一步推进人工智能（AI）和对抗气候变化的研究，社会需要有效利用高性能计算（HPC）资源。该项目将制定一项计划，以建立出色的开源组织，旨在使HPC尽可能易于访问和高效。在社区的帮助下，Stellar将通过协调针对多个科学领域的工具和技术的开发，提供教育服务和技术支持，以更快地采用此类技术，并向社区提供咨询服务，从而开发一个用户和开发人员的软件生态系统。 。星座项目将开发一个路线图，以建立出色的非营利开源组织。恒星组织将制定并监督安全，标准化和工业实力HPX（高性能拼图，标准的C ++库的并发和并行性）运行时的运行时，以克服Exascale时代及以后出现的挑战。 Stellar还将帮助社区围绕HPX构建应用程序界面（API），基础架构，工具和生态系统。此外，组织将通过构建开发人员工具和文档，技术支持，连续测试和集成以及标准化来提供（1）应用程序保证； （2）专业支持，包括培训，定制包装和技术体系结构； （3）教育，劳动力发展和外展等社区服务。 快速采用异步多任务运行时系统（AMT）将有助于提高现代高性能应用的平行效率和可移植性，并在社会进入Exascale时代时提高计算资源的能源效率。这些功能将使科学界能够克服通常实践的批量同步模型的缺点，并为在AI，气候变化，生物信息学以及许多其他高社会影响的应用领域中开发下一代应用。此外，恒星将是高性能计算领域的外展和教育的枢纽。