Phase-I IUCRC Virginia Tech: Center for Space, High-performance, and Resilient Computing (SHREC)

通过这一努力，弗吉尼亚理工大学成为现有的空间、高性能和弹性计算中心（SHREC）的研究基地，该中心是美国国家科学基金会的一个产学研合作研究中心。SHREC由四所大学组成：匹兹堡大学、杨百翰大学、佛罗里达大学和弗吉尼亚理工大学。该中心致力于协助美国工业合作伙伴、政府机构和研究组织在三个领域的关键任务计算：地球科学、空间科学和国防的空间计算；在各种具有重大挑战的应用中实现高性能计算；弹性计算在恶劣或关键环境下的可靠性。该中心还寻求解决关键任务计算劳动力短缺的问题，通过培训许多学生的知识和技能来解决这个不断发展的行业面临的许多挑战。由于每个站点的专业知识具有互补性，该中心将通过利用各种现有和新兴的计算技术，包括现场可编程门阵列（fpga）、图形处理单元（gpu）和混合处理器，解决上述三个关键任务计算领域面临的研究挑战。对于空间计算，该中心将开发、评估和部署新型空间架构、系统和应用，同时利用商业和抗辐射技术。对于高性能计算，该中心将探索异构计算技术和架构的应用和生产使用，以支持高速、关键任务计算。对于弹性计算，该中心将利用其在节能嵌入式计算弹性、故障注入和缓解以及辐射测试方面的专业知识，展示可靠性概念和解决方案，包括自适应硬件冗余、故障屏蔽和软件容错。由于行业合作伙伴采用的新思想和新技术的进步以及中心出版物的特色，SHREC的研究项目将在经济影响方面显着造福社会。此外，新的IUCRC将以学生为中心，每个工业联盟理事会（IAB）成员资助一名研究生，每个项目旨在支持一篇研究生论文，并将雇用不同的本科生和研究生。除了研究活动外，该中心还将成为支持和扩大其大学合作伙伴和成员的教育和外展任务的催化剂。世界自然与自然保护委员会将在中心范围内的存储库（http://www.chrec.org/）中保存由SHREC产生的工件，并逐渐转移到https://nsf-shrec.org/。此外，如果合适，代码和数据将托管在开源GIT存储库（如github、gitlab或gitea）上。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。

With this effort, Virginia Tech becomes a research Site of the existing Center for Space, High-performance, and Resilient Computing (SHREC), an NSF Industry-University Collaborative Research Center. SHREC comprises four university Sites: University of Pittsburgh, Brigham Young University, University of Florida, and Virginia Tech. The Center is dedicated to assisting U.S. industrial partners, government agencies, and research organizations in mission-critical computing across three domains: space computing for Earth science, space science, and defense; high-performance computing across a broad range of grand-challenge applications; and resilient computing for dependability in harsh or critical environments. The Center also seeks to address the shortage in the mission-critical computing workforce by training many students with the knowledge and skills necessary to solve the many challenges facing this growing industry.With the complementary nature of expertise at each Site, the Center will address research challenges facing the three aforementioned domains of mission-critical computing by exploiting a variety of existing and emerging computing technologies, including field-programmable gate arrays (FPGAs), graphical processing units (GPUs), and hybrid processors. For space computing, the Center will develop, evaluate, and deploy novel forms of space architectures, systems, and applications while leveraging commercial and radiation-hardened technologies. For high-performance computing, the Center will explore the application and productive use of heterogeneous computing technologies and architectures in support of high-speed, mission-critical computing. For resilient computing, the Center will exploit its expertise in energy-efficient embedded computing for resilience and in fault injection and mitigation and radiation testing to demonstrate reliability concepts and solutions, including adaptive hardware redundancy, fault masking, and software fault tolerance.Research projects in the SHREC IUCRC will significantly benefit society in terms of economic impact, due to the advancement of new ideas and technologies adopted by industry partners and featured in Center publications. Moreover, the new IUCRC will be student-centric, where each industrial affiliate board (IAB) membership funds a graduate student, and each project aims to support a graduate thesis and will employ a diverse body of undergraduate and graduate students. In addition to its research activities, the Center will serve as a catalyst for supporting and broadening the educational and outreach missions of its university partners and members. Artifacts that result from this SHREC IUCRC will be maintained at the Center-wide repository at http://www.chrec.org/ and gradually transitioned to https://nsf-shrec.org/. In addition, as appropriate, code and data will be hosted at an open-source GIT repository such as github, gitlab, or gitea.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.