Alternative Technologies for Numerical Relativity and LIGO Data-Analysis

数值相对论和 LIGO 数据分析的替代技术

• 批准号: 0902026
• 负责人: Gaurav Khanna
• 金额: $ 16.06万
• 依托单位: University of Massachusetts, Dartmouth
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-09-15 至 2011-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0902026&HistoricalAwards=false
• 关键词: Alternative; Technologies; Numerical; Relativity; LIGO

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5). It supports a project to explore the use of alternative technologies, such as the Sony Playstation 3 gaming console, multimedia workstations and similar hardware for scientific computing. The key aspect of these technologies is that they use a processor that is based on a hybrid approach --- using elements from both traditional CPUs and GPUs. If successful, the hybrid approach would yield a significantly higher cost effectiveness (measured as performance-per-dollar and performance-per-watt) compared with traditional Beowulf compute clusters. More specifically, the project will focus on investigating the performance of such hardware for projects in numerical relativity and (LIGO) data analysis. Preliminary work that has been performed in this direction by the PI has yielded very promising results. If such clusters yield these benefits, searches of LIGO data and simulations of gravitational wave sources could move into computer intensive regimes not now possible and also yield benefits for cluster computing and scientific computing, in general. Undergraduate and graduate students will participate in the project. The PI will participate in outreach activities aimed at promoting science awareness and public knowledge on current scientific research.

该奖项是根据2009年美国复苏和再投资法案（公法111-5）资助的。它支持一个探索使用替代技术的项目，如索尼Playstation 3游戏机、多媒体工作站和类似的科学计算硬件。这些技术的关键方面是它们使用基于混合方法的处理器-使用传统CPU和GPU的元素。如果成功，与传统的Beowulf计算集群相比，混合方法将产生更高的成本效益（以每美元的性能和每瓦的性能来衡量）。更具体地说，该项目将侧重于研究这些硬件在数值相对论和（LIGO）数据分析项目中的性能。PI在这方面开展的初步工作取得了非常有希望的成果。如果这样的集群产生这些好处，那么LIGO数据的搜索和引力波源的模拟可以进入现在不可能的计算机密集型领域，并且通常也会为集群计算和科学计算带来好处。本科生和研究生将参与该项目。 主要研究者将参加旨在提高科学意识和公众对当前科学研究的了解的外联活动。