MRI: Developing a High Performance Computing Center Through Acquisition of a PC Cluster for Cross-Disciplinary Research and Education

MRI：通过收购 PC 集群来开发高性能计算中心，用于跨学科研究和教育

• 批准号: 0619810
• 负责人: Selahattin Ozcelik
• 金额: --
• 依托单位: Texas A&M University-Kingsville
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2006
• 资助国家: 美国
• 起止时间: 2006-08-01 至 2009-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0619810&HistoricalAwards=false
• 关键词: MRI; Developing; Performance; Computing; Center

This project, acquiring a high-performance PC cluster, will enable computationally aggressive techniques to be applied in a variety of disciplines, with a focus on environmental disciplines that relate to the environment and economy of south Texas. This cluster is designed to support the following projects:Integrated windmill and utility system simulation, Modeling/simulation of linked and braided electroactive polymers, Monte Carlo and molecular dynamics simulations of gas hydrates, Modeling subsurface microbial competition,Modeling coupled transport of colloids and contaminants, and their biological effects in river systems,Decision support tools to estimate groundwater availability,Modeling flow and transport of contaminants,GIS-based flood and storm surge simulation and damage assessment,Air quality forecasting,Instrument and measurement research on estimate precision using ratio indicator, andUsing simulation-based genetic algorithms for dynamic signal control optimization in networks with stochastic route choice and time-variant demand.These applications and the general computational methods to be employed have been planned. The system design includes a gigabit-ethernet-based interconnect and 128 processors with 2 GBytes of local memory, connected to a 1.2 TByte storage-area network. Management will include an Advisory Board and a Management Board. TAMUK is providing power/environmental support and system administrator personnel.

这个项目，获得一个高性能的PC集群，将使计算积极的技术应用于各种学科，重点是环境学科，涉及到南德克萨斯州的环境和经济。这一专题组旨在支持下列项目：综合风车和公用事业系统模拟，链接和编织电活性聚合物的建模/模拟，天然气水合物的蒙特卡罗和分子动力学模拟，地下微生物竞争建模，胶体和污染物的耦合运输及其在河流系统中的生物效应建模，估计地下水可用性的决策支持工具，污染物的流动和运输建模，基于GIS的洪水和风暴潮模拟和损害评估，空气质量预测，使用比率指标进行估计精度的仪器和测量研究，以及基于模拟的遗传算法在具有随机路径选择和时变需求的网络中的动态信号控制优化中的应用，这些应用和一般的计算方法已经被计划使用。该系统设计包括一个基于千兆以太网的互连和128个处理器，具有2千兆字节的本地存储器，连接到一个1.2 TB的存储区域网络。管理层将包括一个咨询委员会和一个管理委员会。TAMUK提供电力/环境支持和系统管理人员。