MRI: Acquisition of a Primary Cluster for the SIO COMPAS Shared Computer Facility

MRI：为 SIO COMPAS 共享计算机设施购买主集群

• 批准号: 0723024
• 负责人: Bruce Cornuelle
• 金额: $ 62.72万
• 依托单位: University of California-San Diego Scripps Inst of Oceanography
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2007
• 资助国家: 美国
• 起止时间: 2007-08-15 至 2010-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0723024&HistoricalAwards=false
• 关键词: MRI; Acquisition; Primary; Cluster; SIO

The proposal requests funding for the acquisition of a high performance cluster supercomputer to support numerical modeling aimed at understanding and predicting physical processes and phenomena in the ocean, atmosphere and land surface. The requested instrumentation consists of 128 dual-socket/dual-core nodes, a front-end node, upgraded storage servers, larger memory (16Gb) postprocessing servers and on-site spares. The cluster, which will serve to update the existing Center for Observations, Modeling, and Prediction at Scripps (COMPAS) cluster, will help support regional to global scale ocean-atmosphere-land modeling, process-oriented modeling of small-scale ocean physics, and process-oriented ocean modeling and state estimation. Broader ImpactsThe existing and proposed instrumentation will provide post-docs, graduate students and undergraduates with hands-on computing skills to enable current and future research efforts. Additionally, the PIs anticipate that the capabilities afforded by the requested instrumentation will be useful in attracting new faculty and supporting collaborative efforts, past examples of which are described in the proposal. Additionally, collaboration with the San Diego Supercomputer Center allows for improvements to cluster management software to be distributed back to the wider cluster community.

该提案要求为购置一台高性能集群超级计算机提供资金，以支持旨在了解和预测海洋、大气和陆地表面的物理过程和现象的数值模拟。所请求的工具包括128个双插槽/双核节点、一个前端节点、升级的存储服务器、更大的内存(16 GB)后处理服务器和现场备件。该群组将用于更新现有的斯克里普斯观测、建模和预测中心(COMPAS)群组，将有助于支持区域到全球尺度的海洋-大气-陆地模型、面向过程的小规模海洋物理模型以及面向过程的海洋模型和状态估计。更广泛的影响现有的和拟议的仪器将为博士后、研究生和本科生提供实际操作的计算技能，以支持当前和未来的研究工作。此外，PIS预计，所要求的仪器所提供的能力将有助于吸引新的教员和支持合作努力，提案中描述了过去的例子。此外，通过与圣地亚哥超级计算机中心的合作，可以改进集群管理软件，并将其分发回更广泛的集群社区。