Preliminary Design Planning for the Leadership-Class Computing Facility

领先级计算设施的初步设计规划

• 批准号: 2033468
• 负责人: Daniel Stanzione
• 金额: $ 350万
• 依托单位: University of Texas at Austin
• 依托单位国家: 美国
• 项目类别: Cooperative Agreement
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2033468&HistoricalAwards=false
• 关键词: Preliminary; Design; Planning; Leadership; Class

Over the past decades, the role of computational and data science in advancing discovery has increased tremendously. NSF has played a central role in this transformation of scientific research as a funder of pioneering research activities and as a leader in enabling the use of High-Performance Computing (HPC), with Computation-based methods now becoming central to research in diverse areas of Science and Engineering (S&E). The NSF leadership-class computing facility (LCCF) program is intended to provide unique services and resources to enable discoveries in the largest and most computationally intensive S&E research frontiers that could not advance otherwise. This award supports the design and planning activities associated with the Preliminary Design (PD) of LCCF after successful completion of Conceptual Design (CD). The LCCF project proposes an advanced computing facility to enable discoveries in all fields of S&E requiring leadership computing capabilities, delivering a facility that will not only deal with the very largest challenges in S&E simulation, but also in Artificial Intelligence (AI), as well as the analysis of data from experimental and observational facilities. To deliver this goal, the LCCF project will provide a large-scale computing platform, called Horizon, with a ten-fold increase in the time-to-solution capability of the NSF's current largest system (Frontera), as well as a complete environment for managing data throughout the research lifecycle. The LCCF will also provide a software environment, supporting services, and training to ensure that researchers can effectively use the facility for maximum impact to enhance the scientific discovery process. Extensive plans for Education and Public Outreach are also planned which includes developing existing high impact formal learning programs for students, teachers, and parents, as well as construction for an informal learning environment that will be part of LCCF called the Computarium. The PD process will continue many of the activities from CD, but with a focus on improving the project plans and estimates sufficiently to make important decisions moving forward. Important PD goals are to make the final site selection for the facility, refine the science use cases into specific application targets with defined Figure of Merit performance increases, and choose the general direction of the Horizon system architecture.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

在过去的几十年里，计算和数据科学在推动发现方面的作用已经大大增加。作为开创性研究活动的资助者和高性能计算（HPC）使用的领导者，NSF在科学研究的转变中发挥了核心作用，基于计算的方法现在成为科学和工程各个领域研究的核心（S&amp；E）。美国国家科学基金会领导级计算设施（LCCF）计划旨在提供独特的服务和资源，以实现在规模最大、计算最密集的科学与电子研究领域的发现，否则这些领域将无法取得进展。该奖项支持LCCF在成功完成概念设计（CD）后与初步设计（PD）相关的设计和规划活动。LCCF项目提出了一个先进的计算设施，以支持在S&amp；E需要领先计算能力的所有领域的发现，提供一个设施，不仅可以处理S&amp；E模拟中最大的挑战，还可以处理人工智能（AI），以及分析来自实验和观测设施的数据。为了实现这一目标，LCCF项目将提供一个名为Horizon的大型计算平台，其解决方案的时间比NSF目前最大的系统（Frontera）增加了十倍，同时还提供了一个完整的环境，用于管理整个研究生命周期的数据。LCCF还将提供软件环境、支持服务和培训，以确保研究人员能够有效地利用该设施，最大限度地提高科学发现过程。广泛的教育和公共宣传计划也在计划中，包括为学生、教师和家长开发现有的高影响力的正式学习项目，以及建设一个非正式的学习环境，这将是LCCF的一部分，称为Computarium。PD过程将继续CD中的许多活动，但重点是充分改进项目计划和评估，以便做出重要的决策。重要的PD目标是为设施做出最终的选址，将科学用例细化为特定的应用目标，具有定义的性能提升图，并选择Horizon系统架构的总体方向。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。