Computer cluster for computational biology

计算生物学计算机集群

基本信息

批准号：
6877645
负责人：
Ron Elber
金额：
$ 50万
依托单位：
CORNELL UNIVERSITY
依托单位国家：
美国
项目类别：
财政年份：
2005
资助国家：
美国
起止时间：
2005-06-01 至 2006-05-31
项目状态：
已结题

项目摘要

DESCRIPTION (provided by applicant): The present application aims to establish a computer Cluster for Computational Biology and Bioinformatic (CCBB). The cluster will consists of 256 dual nodes connected with Giganet switches to enable rapid communication between the processors. The cluster will enable the integration of the two approaches and make it possible to effectively address the highly demanding computational tasks of the field. It will serve a small group of investigators, supported by the NIH, and their close collaborators. The hardware needs of computational biology and bioinformatic applications, and of the team of investigators listed in this application can be summarized as follows: 1. Significant computer power for complex and expensive simulations. 2. Large storage capacity for the whole cluster (shared) and (separately) for the individual nodes. 3. Large and rapidly accessible memory for effective statistical analysis, application of machine learning techniques, and biological discovery. 4. Fast network for information updates across the network. In addition CCBB will have high level of databases and software integration including 1. Updates of important "mirrors" of shared databases (such as NR, swissprot, human EST, human genome, protein databank, etc.) 2. Local installation and frequent upgrade of widely used software packages (e.g. BLAST, Pfam, CHARMm etc.) 3. Help in porting novel software for optimal use on the CCBB hardware platform. The combined unification of optimal hardware and software for computational biology and bioinformatic will make the new cluster; an outstanding resource for NIH related research

描述（由申请人提供）：本应用程序旨在建立用于计算生物学和生物信息学（CCBB）的计算机群集。该群集将由256个与Giganet开关连接的双节点组成，以实现处理器之间的快速通信。该集群将使两种方法的集成能够集成，并可以有效解决该领域的高度苛刻的计算任务。它将为NIH支持的一小群调查人员及其密切合作者提供服务。计算生物学和生物信息学应用程序的硬件需求以及本应用程序中列出的调查人员团队的硬件需求可以总结如下： 1。复杂且昂贵的模拟的重要计算机功率。 2。整个群集（共享）和（分别）单个节点的大存储容量。 3。用于有效的统计分析，机器学习技术的应用和生物学发现的大型且迅速可访问的内存。 4。快速网络以跨网络的信息更新。另外，CCBB将具有高水平的数据库和软件集成，包括 1。共享数据库重要的“镜像”的更新（例如NR，SwissProt，Human EST，人类基因组，蛋白质数据库等） 2。本地安装和频繁升级使用的软件包（例如BLAST，PFAM，CHARMM等） 3。帮助移植新的软件，以在CCBB硬件平台上进行最佳使用。用于计算生物学和生物信息学的最佳硬件和软件的合并统一将使新群集成为新集群； NIH相关研究的杰出资源