Unix Computer Cluster

Unix计算机集群

• 批准号: 6441094
• 负责人: BENNETT DYKE
• 金额: $ 22.82万
• 依托单位: TEXAS BIOMEDICAL RESEARCH INSTITUTE
• 依托单位国家: 美国
• 财政年份: 2002
• 资助国家: 美国
• 起止时间: 2002-04-01 至 2003-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6441094
• 关键词: biomedical equipment purchase; computers

The major research of the Department of Genetics at the Southwest Foundation for Biomedical research (SFBR) is aimed at finding genes that contribute to susceptibility to common diseases (heart disease, diabetes, obesity, osteoporosis, etc.) and their risk factors. Unlike much of the traditional work in these areas which analyzes disease patterns in sibships and nuclear families, we have pioneered the use of extended families (pedigrees) that may include many dozens of relatives. This approach greatly increases the amount of information about genetic transmission available for analysis. To carry out the highly compute- intensive analyses that are the consequence of this approach, we also pioneered in the development and application of parallel processing in statistical genetics. In 1995 we assembled a cluster of Unix processors to run production analyses written specifically for the purpose in our Department. The success of this approach led to a regular series of upgrades to this cluster, and ultimately to construction of a second cluster in 2000. As the number of our projects has increased and processing demand grows, we are faced with a need for additional computing power at a time when our original cluster is increasingly obsolete. Thus, we are requesting funds to purchase a new 100-node dual processor Unix cluster to augment our current capabilities.

西南生物医学研究基金会（SFBR）遗传学系的主要研究旨在寻找对常见疾病（心脏病，糖尿病，肥胖，骨质疏松症等）敏感的基因及其危险因素。与这些领域的许多传统作品不同，这些领域分析了sibships和核心家庭的疾病模式，我们开创了可能包括数十个亲戚的大家庭（家族）的使用。这种方法大大增加了有关可用于分析的遗传传播的信息量。为了进行这种方法的高度计算分析，我们还率先开发了并行处理在统计遗传学中的开发和应用。 1995年，我们组建了一组UNIX处理器，以运行专门为我们部门的目的而编写的生产分析。这种方法的成功导致了对该集群的常规升级，并最终在2000年建造了第二个集群。随着我们项目的数量增加并增加了处理需求，我们面临着额外的计算能力在我们的原始集群越来越过时的时候。因此，我们要求资金购买新的100节点双处理器UNIX群集，以增强我们当前的功能。