SCI: TeraGrid Resource Partners

可扩展兆级设施（ETF）是NSF大规模网络基础设施发展的下一个阶段，旨在实现高端计算研究。ETF使研究人员能够利用9个资源合作伙伴的综合资源、数据收集、仪器和可视化能力来解决最具挑战性的计算问题。2004年10月1日，ETF结束了为期三年的建设工作，创建了这个称为TeraGrid （TG）的分布式环境，我们现在正在进入生产操作阶段。TeraGrid资源合作伙伴包括：芝加哥大学/阿贡国家实验室、加州大学圣地亚哥分校的圣地亚哥超级计算机中心、德克萨斯大学奥斯汀分校的德克萨斯高级计算中心、UIUC的国家超级计算应用中心、印第安纳大学、普渡大学、橡树岭国家实验室和匹兹堡超级计算中心。2004年10月19日，美国国家科学基金会（NSF）为TeraGrid电网基础设施集团（GIG）提交了一份单独的提案。在UC/ANL的Charlie Catlett的指导下，GIG将负责协调TeraGrid的开发活动，并将分包合同分包给合作伙伴。资源合作伙伴（RP）将分别与NSF签订独立的合作协议，但将与GIG密切合作，以实现TeraGrid的愿景。该提案概述了SDSC作为TeraGrid团队的资源合作伙伴参与的计划，为不断扩大的用户社区提供对该计算科学设施的持续访问。该提案涵盖2004年11月1日至2009年10月31日期间。SDSC将为TeraGrid带来一套强大的计算和存储资源，并将专注于TeraGrid科学家的数据密集型计算需求。SDSC目前的TeraGrid资源与其他RP站点的资源进行了总结，包括3.1 Tflop IA-64系统，1.1 Tflop Power4系统（更大的DataStar系统的一部分）和实验性1.3 Tflop IA-64系统。每个SDSC计算节点都可以以200mbps或更高的速度访问数据，并且可以使用其DB2客户机通过前沿服务器对结构化数据库进行寻址。中央存储区域网络（SAN）拥有超过0.5 PB的磁盘和6 PB的归档磁带容量，可供所有节点和服务器使用，SDSC计划使用真正的全局文件系统将其功能导出到TeraGrid广域网（WAN）[16,17]。SDSC的540tb SAN在线磁盘将使跨TeraGrid处理数据的新模式成为可能。在本地生成数据，与他人竞争磁盘空间，通过通常很慢的接口将结果保存到存档存储，这些都是对科学家的限制。对科学家来说，直接写入一个非常大的文件系统的能力是一项重大的能力改进，在这个文件系统中可以进行后处理，并将结果自动存档。例如，南加州地震中心（SCEC）[6]进行了一次地震模拟，分析了南加州地区潜在地震造成的破坏。

The Extensible Terascale Facility (ETF) is the next stage in the evolution of NSF large-scale cyberinfrastructure for enabling high-end computational research. The ETF enables researchers to address the most challenging computational problems by utilizing the integrated resources, data collections, instruments and visualization capabilities of nine resource partners. On October 1, 2004, the ETF concluded a three-year construction effort to create this distributed environment called the TeraGrid (TG) and we are now entering the production operations phase.The TeraGrid resource partners include: the University of Chicago/Argonne National Laboratory, the San Diego Supercomputer Center at UCSD, the Texas Advanced Computing Center at UT-Austin, the National Center for Supercomputing Applications at UIUC, Indiana University, Purdue University, Oak Ridge National Laboratory, and the Pittsburgh Supercomputing Center.A separate proposal was submitted to NSF on October 19, 2004 for the TeraGrid Grid Infrastructure Group (GIG). Under the direction of Charlie Catlett at UC/ANL, in general, the GIG will be responsible for coordination of development activities for the TeraGrid with subcontracts to the partner sites. The resource partners (RP) will each have independent cooperative agreements with NSF, but will work closely with the GIG to implement the vision of the TeraGrid.This proposal outlines the plan at SDSC to participate as a resource partner in the TeraGrid team to provide the expanding user community with ongoing access to this computational science facility. This proposal covers the period November 1, 2004 through October 31, 2009.SDSC will bring a powerful set of compute and storage resources to the TeraGrid and will focus on the data-intensive computing needs of TeraGrid scientists. SDSC's current TeraGrid resources are summarized with those of other RP sites and include a 3.1 Tflop IA- 64 system, a 1.1 Tflop Power4 system (part of the larger DataStar system), and an experimental 1.3 Tflop IA-64 system. Every SDSC compute node can access data at 200 MBps or more and can use its DB2 client to address structured databases via leading edge servers. The central Storage Area Network (SAN) has over 0.5 PB of disk and a 6 PB archival tape capacity available to all nodes and servers, and SDSC plans to export its capabilities across the TeraGrid Wide Area Network (WAN) [16, 17] using a true global file system. SDSC's 540 TB SAN of online disk will make possible new modes of handling data across the TeraGrid. Generating data locally, competing with others for disk space and saving results to archival storage through an often slow interface all serve as restrictions to scientists. The ability to write directly to a very large file system, where post-processing can occur and where results are automatically archived, is a significant improvement in capability for scientists. For example, the Southern California Earthquake Center (SCEC) [6] has performed an earthquake simulation analyzing the resulting damage from a potential earthquake in the Southern California region.