Extending the Work of the International Exascale Software Project

扩展国际百亿亿次级软件项目的工作

• 批准号: 1136509
• 负责人: Jack Dongarra
• 金额: $ 9.98万
• 依托单位: University of Tennessee Knoxville
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2011
• 资助国家: 美国
• 起止时间: 2011-08-01 至 2015-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1136509&HistoricalAwards=false
• 关键词: Extending; Work; International; Exascale; Software

The International Exascale Software Project (IESP) brings together representatives from the major segments of the global high performance computing (HPC) community -- software research and development, leading edge scientific applications, leadership-class facilities, major system vendors and relevant government funding agencies -- in a proactive effort to plan and create a new software infrastructure for the extreme scale systems that represent the future of computational science. Through a series of workshops, begun in the spring of 2009 and rotating between the US, Europe, and Asia, the IESP and its international partners have endeavored to provide leadership in formulatung a long-term roadmap to an exascale-capable software infrastructure, and in initializing the corresponding research and development process necessary to produce it. As expressed in the IESP Roadmap version 1.0 (referenced below as "Roadmap" and published on May 30, 2010), the project's mission commits to a vision of the future of scientific computing that involves to two key objectives: The guiding purpose of the IESP is to empower ultrahigh resolution and data intensive science and engineering research through the year 2020 by developing a plan for 1) a common, high quality computational environment for peta/exascale systems and for 2) catalyzing, coordinating, and sustaining the effort of the international open source software community to create that environment as quickly as possible. (Roadmap, p. 3)In a series of workshops, rotating between the United States, Europe and Asia, IESP participants from all segments of the HPC community have begun to forge a working consensus on a plan to address the momentous challenge of creating a software stack for extreme-scale Computational Science, which we call an "X-stack." Building on this success, our proposed IESP workshop for the spring of 2012 in Chicago will focus on further developing and expanding upon the strategies identified in the plan in order to make more rapid progress toward the project's objectives.

国际兆级软件项目（IESP）汇集了来自全球高性能计算（HPC）社区主要领域的代表--软件研发、前沿科学应用、领先级设施，主要系统供应商和相关政府资助机构--积极努力为代表计算科学未来的极端规模系统规划和创建新的软件基础设施。通过一系列的研讨会，从2009年春天开始，在美国，欧洲和亚洲之间轮流，IESP及其国际合作伙伴已经承诺提供领导，制定一个长期的路线图，以一个具有千兆级能力的软件基础设施，以及启动生产该文件所需的相应研究和开发进程。（以下称为“路线图”，并于2010年5月30日发布），该项目的使命致力于科学计算的未来愿景，涉及两个关键目标：IESP的指导目的是通过制定一项计划，在2020年之前为高分辨率和数据密集型科学和工程研究提供支持，PETA/Exascale系统的高质量计算环境，以及2）催化，协调和维持国际开源软件社区的努力，以尽快创建该环境。（路线图，第3页）在美国、欧洲和亚洲轮流举办的一系列研讨会上，来自HPC社区各个部分的IESP参与者已经开始就一项计划达成工作共识，以解决为极端规模计算科学创建软件堆栈的重大挑战，我们称之为“X堆栈”。“在这一成功的基础上，我们提议于2012年春季在芝加哥举行的IESP研讨会将侧重于进一步发展和扩大计划中确定的战略，以便更快地实现项目的目标。