MRI: Acquisition of a High-Performance Parallel Computing Cluster for Astrophysics Research at the University of California, Santa Cruz

MRI：在加州大学圣克鲁斯分校购买用于天体物理学研究的高性能并行计算集群

• 批准号: 1229745
• 负责人: Piero Madau
• 金额: $ 91万
• 依托单位: University of California-Santa Cruz
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2012
• 资助国家: 美国
• 起止时间: 2012-09-01 至 2014-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1229745&HistoricalAwards=false
• 关键词: MRI; Acquisition; Performance; Parallel; Computing

Astronomers don't just use telescopes in their attempts to unravel the mysteries of the universe. Many problems require heavy analytic computation and the use of computer simulations to understand how planets, stars, and galaxies form, grow, evolve, and die. The simulations can include a huge number of objects, embedded in a complex environment of dust, gases, magnetic fields as well as the unknown dark matter, and follow these objects as they interact, coalesce, or disrupt. Today's computers have become massively parallel - that is to say that they have many "processing units" working in parallel to solve these highly complex problems. One noted and successful team at the University of California Santa Cruz proposes to acquire a new computer "cluster" to attack difficult unsolved problems in astrophysics. The Principal Investigator, Dr. Piero Madau, and his team will use their new computing power to try to answer many questions in astrophysics from the very early days of the universe up to the present time and beyond. This new computer cluster will consist of hundreds of multiple processors, both conventional computer processors as well as graphical processors. It will have one petabyte of disk storage capacity (that is one million gigabytes!). The acquisition of this important research tool is funded by NSF's Division of Astronomical Sciences through the Major Research Instrumentation program.

天文学家不仅仅是用望远镜来解开宇宙的奥秘。许多问题需要大量的分析计算和使用计算机模拟来理解行星、恒星和星系如何形成、生长、演化和死亡。模拟可以包括大量的物体，嵌入在一个由尘埃、气体、磁场以及未知暗物质组成的复杂环境中，并跟踪这些物体相互作用、合并或破坏的过程。今天的计算机已经变得高度并行——也就是说，它们有许多“处理单元”并行工作，以解决这些高度复杂的问题。加州大学圣克鲁斯分校（University of California Santa Cruz）的一个著名且成功的团队提议收购一个新的计算机“集群”，以解决天体物理学中尚未解决的难题。首席研究员Piero Madau博士和他的团队将利用他们新的计算能力，试图回答从宇宙早期到现在以及以后的天体物理学中的许多问题。这个新的计算机集群将由数百个多个处理器组成，既有传统的计算机处理器，也有图形处理器。它将具有1拍字节的磁盘存储容量（即100万千兆字节！）。这个重要的研究工具是由美国国家科学基金会天文科学部通过主要研究仪器项目资助的。