Parallel Computer

并行计算机

• 批准号: 6803671
• 负责人: Jason H. Moore
• 金额: $ 152.62万
• 依托单位: VANDERBILT UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2004
• 资助国家: 美国
• 起止时间: 2004-07-01 至 2005-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6803671
• 关键词: biomedical equipment purchase; computational biology; parallel processing

DESCRIPTION (provided by applicant): We propose establishing an 824-processor parallel computer system to support computationally intensive, NIH-funded, biomedical research at Vanderbilt University Medical School. The proposed system will use the Beowulf model that provides supercomputer performance at a small fraction of the cost. This system will build upon our successful preliminary studies over the last three years establishing and utilizing a 110-processor pilot cluster. The proposed system will be added to a 270-processor parallel computer that was recently established by the School of Engineering and Department of Physics as part of Vanderbilt's new Scientific Computing Center. Thus, the required infrastructure and expertise is already in place. The proposed parallel computer will be used by a minimum of 16 NIH-funded biomedical researchers and will support at least 100 computational projects from 68 active and 16 pending NIH grants. The faculty users of the system have primary appointments in seven different departments within the Vanderbilt University School of Medicine. Research programs include computational studies in bioinformatics, biophysics, biostatistics, genetic epidemiology, genomics, human genetics, proteomics, and structural biology. This resource will make it possible to carry out computationally intensive biomedical investigations otherwise not feasible thus significantly increasing the rate of scientific discovery in a variety of biomedical disciplines.

描述（由申请人提供）：我们建议在范德比尔特大学医学院建立一个824个处理器并行计算机系统，以支持计算密集型，NIH资助的生物医学研究。提出的系统将使用Beowulf模型，该模型以一小部分成本提供超级计算机性能。该系统将基于我们在过去三年中成功建立和利用110个过程的飞行员集群的成功初步研究。拟议的系统将被添加到270个处理器并行计算机上，该计算机最近由工程学院和物理系建立，作为范德比尔特新科学计算中心的一部分。因此，所需的基础架构和专业知识已经到位。所提出的平行计算机将至少16个由NIH资助的生物医学研究人员使用，并将支持来自68个活跃和16个悬而未决的NIH赠款的至少100个计算项目。该系统的教职用户在范德比尔特大学医学院的七个不同部门中有主要任命。研究计划包括生物信息学，生物物理学，生物统计学，遗传流行病学，基因组学，人类遗传学，蛋白质组学和结构生物学的计算研究。该资源将使进行计算密集型生物医学研究否则不可行是可能的，因此可以显着提高各种生物医学学科的科学发现率。