MRI: Acquisition of a Big Data Compute Cluster for Interdisciplinary Research

美国国家科学基金会（NSF）授予加州大学河滨分校（UCR）一个高可扩展的大数据计算集群，用于长期支持数据密集型研究。通过释放和扩大目前超额预订的研究计算资源，拟议的计算集群将对研究生和本科生的培训和教育产生重大影响，包括来自传统上在STEM学科中代表性不足的人群的高比例。根据URM的入学情况，UCR被教育部指定为西班牙裔服务机构（HSI）。足够的计算机的可用性及其对研究项目的有益影响将有助于吸引优秀的学生参加大量的本科和研究生项目，包括成功的nsf资助的REU和IGERT项目。与系统获取一起计划的新课程将训练学生并行计算概念，并提供将大数据计算集成到他们的研究项目中的扩展访问。该项目的大数据计算集群还将支持加州几家已经使用IIGB计算机设施的小型生物技术公司。未来将积极招募来自外部研究机构、少数民族院校和工业合作伙伴（尤其是初创企业）的用户，以获得IIGB的计算资源。结合UCR的多元化种族和研究使命，这项投资将有利于UCR广泛的研究方向和以技术为基础的经济发展计划，UCR是加州内陆帝国经济发展的重要驱动力。该项目的目标是使UCR的大数据驱动研究能够在高度跨学科的环境中解决重大挑战问题。问题包括：不同的生物群体如何适应和保护自己免受极端环境条件或病原体的侵害？如何开发更高效、更有选择性的小分子来加速以发现为导向的化学生物学和基因组学研究？如何将这些知识转化为提高作物对压力和病原体的耐受性，例如在应对全球气候变化和养活不断增长的世界人口方面？高通量和监测技术的最新进展首次为系统、全面和前所未有的解决方案解决这些挑战提供了新方法。新的大数据计算集群大大加强了UCR的高性能计算基础设施，并为UCR的研究人员提供了一个关键的支持资源，这些研究领域包括环境科学、化学基因组学、进化、统计学、计算生物学和多生物群体的基因组生物学。由于这项研究依赖于高通量和计算建模方法，其成功和未来的增长严重依赖于高性能计算机资源来管理和处理大型和快速增长的数据集。该系统将由综合基因组生物学研究所（IIGB）研究计算设施的经验丰富的人员管理。IIGB设施为分布在50多个研究小组的部门和160个活跃用户提供广泛的用户群。因此，所要求的计算系统将达到最大数量的美国国家科学基金会资助的UCR研究人员，并构成美国国家科学基金会和UCR资金的成本效益投资。

An award is made to the University of California, Riverside (UCR) to acquire a highly scalable Big Data compute cluster dedicated to long-term support for data-intensive NSF research. By freeing and expanding currently overbooked research compute resources, the proposed compute cluster will have a significant impact on training and educating graduate and undergraduate students, including a high percentage from populations traditionally underrepresented in STEM disciplines. Based on URM enrollment, UCR has been designated by the Dept. of Education as an Hispanic Serving Institution (HSI). The availability of adequate computing and its beneficial impact on research programs will serve to attract outstanding students to a large number of undergraduate and graduate programs, including successful NSF-Funded REU and IGERT programs. New courses planned in conjunction with system acquisition will train students in parallel computing concepts and offer expanded access to integrate Big Data computing into their research projects. The Big Data compute cluster of this project also will support several smaller biotechnology companies in California that already use IIGB's computer facility. Future users from external research institutions, minority-serving colleges and industrial partners, particularly start-ups, will be actively recruited to gain access to IIGB's computing resources. Combined with UCR's diverse ethnicity and research mission, this investment will benefit a wide array of research directions and technology-based economic development initiatives at UCR, an institution that serves as an important driver of economic development in California's Inland Empire.The goal of this project is to enable Big Data driven research at UCR to address grand challenge questions in a highly interdisciplinary environment. Questions include: How do different organism groups adapt to, and defend themselves against extreme environmental conditions or pathogens? How can more efficient and selective small molecules be developed to accelerate discovery-oriented chemical biology and genomics research? How can this knowledge translate into improved stress and pathogen tolerance, for example in crops to respond to global climate change and feed a growing world population? Recent advances in high-throughput and monitoring technologies offer, for the first time, novel methods to address these challenges systematically, comprehensively, and with unprecedented resolution. The new Big Data compute cluster substantially strengthens UCR's high-performance compute infrastructure and provides a critical enabling resource for UCR researchers from a broad spectrum of research specializations, including environmental science, chemical genomics, evolution, statistics, computational biology, and genome biology of multiple organism groups. Since this research relies on high-throughput and computational modeling approaches, its success and future growth is critically dependent on high-performance computer resources to manage and process large and rapidly increasing data sets. The system will be managed by experienced personnel in the Research Compute Facility of the Institute for Integrative Genome Biology (IIGB). The IIGB facility serves a broad user population distributed across departments from more than 50 research groups with 160 active users. The requested computing system therefore will reach a maximum number of NSF-funded investigators at UCR and constitutes a cost-effective investment of NSF and UCR funds.