Enabling Productive, High-Performance Data Analytics

实现高效、高性能的数据分析

• 批准号: 1234749
• 负责人: Michael Levine
• 金额: $ 111.4万
• 依托单位: Carnegie-Mellon University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2012
• 资助国家: 美国
• 起止时间: 2012-09-01 至 2015-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1234749&HistoricalAwards=false
• 关键词: Enabling; Productive; Performance; Data; Analytics

This award provides funding for Sherlock, a Cray YarcData uRiKA data appliance consisting of a next-generation Cray XMT supercomputer (NG-XMT) running the uRiKA application architecture and augmented by Cray XT5 compute nodes to broaden the range of relevant applications to address the challenges of graph-based data analytics. The system will be deployed at the Pittsburgh Supercomputing Center (PSC). The Cray NG-XMT is a massively multithreaded supercomputer, based on the Cray XT5 infrastructure and specialized for analytics. Its Cray-proprietary Threadstorm 4.0 processors implement multiple, powerful features to support lightweight multithreading, latency hiding, and advanced memory interfaces with AMD HyperTransport-attached SeaStar2 interconnect chips to provide a flat, globally-addressable memory. This will be the first system of its kind to be available to the NSF research community.

该合同为Sherlock提供资金，Sherlock是一款Cray YarcData uRiKA数据设备，由运行uRiKA应用程序架构的下一代Cray XMT超级计算机（NG-XMT）组成，并由Cray XT5计算节点增强，以扩大相关应用的范围，以应对基于图形的数据分析的挑战。该系统将部署在匹兹堡超级计算中心（PSC）。Cray NG-XMT是基于Cray XT5基础架构的大型多线程超级计算机，专门用于分析。其cray专有的Threadstorm 4.0处理器实现了多种强大的功能，支持轻量级多线程，延迟隐藏，以及与AMD hypertransport附加的SeaStar2互连芯片的高级内存接口，以提供扁平的全球可寻址内存。这将是同类系统中第一个提供给NSF研究界的系统。