Workshop: PetaFLOPS Architecture & Software Workshops Held in Oxnard, CA April 22-26, 1996 & Bodega Bay, CA June 17-21, 1996

研讨会：PetaFLOPS 架构

• 批准号: 9634437
• 负责人: Paul Messina
• 金额: $ 4.52万
• 依托单位: California Institute of Technology
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1996
• 资助国家: 美国
• 起止时间: 1996-04-01 至 1999-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9634437&HistoricalAwards=false
• 关键词: Workshop; PetaFLOPS; Architecture; & Software

The successful federal HPCC program is in its last phase, establishing the basis for Teraflops scale computing systems and applications. An important finding of this historical collaborative program is that many mission critical objectives of key Federal agencies require computing capability well beyond even that provided by the Teraflops systems of the near future. Furthermore, the challenges to programming and resource management of the largest contemporary MPPs has revealed the need for alternative system structures and methodologies if very high performance computing is to become an important and wide spread enabler of industrial, commercial, government and science objectives. In preparation to establishing future policy and planning directions for next generation research programs in advanced computing, a number of Federal agencies with compute critical requirements have joined to initiate a set of preliminary inquires in the form of workshops, panels, and studies. This grant, made in cooperation with ARPA, DOE, DOD, and NASA, will provide funding for two workshops on Petaflop-scale computing. The first workshop will focus on architectures, the second on system software issues. Both will be held in California in summer 1996.

成功的联邦HPCC项目已进入最后阶段，为Teraflops规模的计算系统和应用奠定了基础。这一历史性合作项目的一个重要发现是，关键联邦机构的许多关键任务目标需要的计算能力远远超过不久将来的Teraflops系统所能提供的能力。此外，当代最大的mpp在编程和资源管理方面面临的挑战表明，如果要使高性能计算成为工业、商业、政府和科学目标的重要和广泛的推动者，就需要替代的系统结构和方法。在为下一代高级计算研究计划制定未来政策和规划方向的准备工作中，许多具有计算关键需求的联邦机构已经联合起来，以研讨会、小组和研究的形式发起了一系列初步调查。这笔拨款是与ARPA、DOE、DOD和NASA合作的，将为两个千万亿次计算研讨会提供资金。第一个研讨会将关注架构，第二个研讨会将关注系统软件问题。两者都将于1996年夏季在加利福尼亚举行。