Acquisition of Parallel Computing Cluster for Large-Scale Computational Problems in the Mathematical Sciences

数学科学中大规模计算问题并行计算集群的获取

• 批准号: 0532265
• 负责人: Matthew Saltzman
• 金额: $ 14.06万
• 依托单位: Clemson University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2005
• 资助国家: 美国
• 起止时间: 2005-09-15 至 2006-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0532265&HistoricalAwards=false
• 关键词: Acquisition; Parallel; Computing; Cluster; Large

Acquisition of Parallel Computing Cluster forLarge-Scale Computational Probelms in theMathematical SciencesThe investigators will construct a 64-processor parallel computing clus-ter. The computational power of this machine will be applied to solve prob-lems that rank among the most important in computational mathematics.This research will have applications in a variety of important areas, in-cluding: reliability and security of communication networks, cryptography,transportation and logistics, management of medical treatments, and man-ufacturing systems design and operation. Each of these projects requiresmassively parallel computations with high precision or a very large memoryspace on each processing node. This cluster will make possible computationsthat were previously impractical or impossible.The investigators will involve undergraduate and graduate students inthe development and implementation of parallel algorithms related to theirresearch in statistics, operations research, number theory, combinatorics,coding theory and numerical analysis. Through the Research Experience forUndergraduates program in Clemson's Mathematical Sciences Department,they will be training some of the best and brightest young mathematiciansof the future in these techniques. Thus this project will have the effectof training a new generation of researchers in parallel computation in themathematical sciences.

面向数学科学中大规模计算问题的并行计算集群的研制研究人员将建造一个64处理器的并行计算集群。这台机器的计算能力将用于解决计算数学中最重要的问题，这项研究将在各种重要领域中得到应用，包括：通信网络的可靠性和安全性，密码学，运输和物流，医疗管理以及制造系统的设计和操作。这些项目中的每一个都需要在每个处理节点上进行高精度的大规模并行计算或非常大的内存空间。这个集群将使以前不切实际或不可能的计算成为可能。研究人员将让本科生和研究生参与开发和实现与他们在统计学、运筹学、数论、组合学、编码理论和数值分析方面的研究有关的并行算法。通过克莱姆森的数学科学系本科生项目的研究经验，他们将在这些技术中培养一些最优秀和最聪明的年轻数学家。因此，该项目将具有培养新一代数学科学并行计算研究人员的效果。