SCI: TeraGrid Resource Partners

可扩展太尺度设施（ETF）是NSF大规模网络基础设施发展的下一个阶段，用于支持科学和工程研究。ETF使研究人员能够利用九个资源合作伙伴（RP）的集成计算资源，数据收集，仪器和可视化功能来解决具有挑战性的科学和工程问题。2004年10月1日，ETF完成了为期三年的TeraGrid建设工作，目前正进入运营阶段。TeraGrid资源合作伙伴包括： 芝加哥大学/阿贡国家实验室（UC/ANL）、加州大学圣地亚哥分校（UCSD）的圣地亚哥超级计算机中心、德州大学奥斯汀分校的德克萨斯高级计算中心、印第安纳州大学UIUC的国家超级计算应用中心、普渡大学、橡树岭国家实验室（ORNL）和匹兹堡超级计算中心。2004年10月19日，TeraGrid网格基础设施组（GIG）向NSF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，并将合同分包给合作伙伴。每个资源合作伙伴都将与NSF签订独立的协议，但将与GIG密切合作，以实现TeraGrid愿景。ORNL RP项目愿景和目标ETF中的其他RP合作伙伴提供高端计算和数据存储资源。ORNL的独特贡献是将仪器数据源连接到ETF，并开发了一个中子科学网关，使中子科学家能够使用TeraGrid资源分析他们的数据，包括将实验与模拟进行比较。ORNL将与GIG和其他RP工作密切合作，以实现ETF范围功能的本地实例或利用其他合作伙伴（例如TeraGrid运营中心）提供的ETF范围服务。中子科学应用将通过科学网关部署，并使用ETF广泛的计算和数据存储资源。ORNL RP认为，ETF的合作伙伴关系不是竞争机构之间的合作，而是互补服务和功能的合作，旨在为国家利益创建一个独特的大规模研究设施。该提案概述了ORNL作为RP参与TeraGrid团队的计划，为中子科学界提供访问这个独特的计算科学设施的机会。美国能源部（DOE）正在ORNL建造下一代中子源，价值14亿美元的SNS。SNS将提供比世界上任何其他设施更大的通量，并将在审查的基础上向全国所有科学家提供，独立于资金来源。有了这种新的能力，中子科学界正面临着数量级更大的数据集，并处于数据分析和模拟的关键点。他们认识到需要新的方法来管理和分析数据，以优化射束时间和科学产出。TeraGrid将提供他们所需要的新能力，主要目标群体是中子散射设施的所有用户。每个设施大约有几千名用户（许多科学家使用不止一个设施）。在美国有五（5）家工厂，在世界各地有几十家。ORNL RP不能也不会假设这些用户将完全依赖NSTG和TeraGrid来解决他们出现的问题。他们的忠诚必须赢得。这个RP成功的关键指标之一是它可以说服这个社区使用这些资源的程度。关键是提供一个基础设施，帮助社区超越本来可以实现的目标。其次，将对其他大型用户设施（如光源和显微镜中心）的科学行为产生影响。

The Extensible Terascale Facility (ETF) is the next stage in the evolution of NSF large-scale cyberinfrastructure for enabling scientific and engineering research. The ETF enables researchers to address challenging scientific and engineering problems by utilizing the integrated computing resources, data collections, instruments and visualization capabilities of nine resource partners (RP). On October 1, 2004, the ETF concluded a three-year construction effort to construct the TeraGrid and is now entering the operations phase. The TeraGrid resource partners include; the University of Chicago/Argonne National Laboratory (UC/ANL), the San Diego Supercomputer Center at University of California, San Diego (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 (ORNL), 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, the GIG will be responsible for coordination of development activities for the TeraGrid with subcontracts to the partner sites. The resource partners will each have independent agreements with NSF, but will work closely with the GIG to implement the TeraGrid vision. ORNL RP Project Vision and GoalsOther RP partners in the ETF provide high-end computing and data storage resources. The unique contribution of ORNL is the connection of instrument data sources to the ETF and the development of a neutron science gateway that will allow neutron scientists to use TeraGrid resources to analyze their data, including comparison of experiment with simulation. ORNL will work closely with the GIG and other RP efforts to implement local instances of ETF-wide functionality or utilize ETF-wide services provided by the other partners, for example the TeraGrid Operations Center. Neutron science applications will be deployed through a science gateway and use ETF wide computing and data storage resources. ORNL RP sees the ETF partnership not as a collaboration between competing institutions but as a collaboration of complementary services and functionality to create a unique, large scale research facility for the national interest. This proposal outlines the plan for ORNL to participate as a RP in the TeraGrid team by providing the neutron science community with access to this unique computational science facility. A next-generation neutron source, the$1.4 billion SNS, is being built by the U.S. Department of Energy (DOE) at ORNL. The SNS will provide an order of magnitude greater flux than any other facility in the world and will be available to all of the nation's scientists, independent of funding source, on a reviewed basis. With this new capability, the neutron science community is facing order of magnitude larger data sets and is at a critical point for data analysis and simulation. They recognize the need for new ways to manage and analyze data to optimize both beam time and scientific output. The TeraGrid will provide the new capabilities that they need.The primary target community is all users of neutron scattering facilities. This is on the order of a few thousand users per facility (many scientists use more than one facility). There are five (5) facilities in the U.S. and a few dozen world-wide. The ORNL RP cannot and does not assume that these users will rely exclusively on the NSTG and the TeraGrid as a solution to their emerging problems. Their allegiance must be earned. One of the key metrics of this RP's success is the extent to which it can persuade this community to use these resources. The key will be to provide an infrastructure that helps the community reach beyond what would otherwise have been accomplished. Secondarily, there will be impacts on the conduct of science at other large user facilities such as light sources and microscopy centers.