REU Site: Interdisciplinary Program in High Performance Computing

REU网站：巴尔的摩县（UMBC）马里兰州大学数学与统计系的高性能计算跨学科课程是一个为期八周的暑期住宿课程，面向数学、统计或计算机科学专业的本科生。 本REU网站使用可分为三个主要组成部分的活动来培养未来劳动力的技能：（1）证书课程中的结构化培训通过讲座和TA支持的计算机实验室教授高性能计算的基本技能，使用Linux，C，MPI，MATLAB和R等特定工具。 (2)科学，统计和并行计算的技能适用于由来自工业，政府机构或数学和统计以外的部门的科学家提出的跨学科研究项目。 (3)教学和研究是由一套完整的高度综合的专业发展活动，包括团队工作经验，演讲技巧，（海报，口头，项目更新，和'电梯间距'），写作技巧（摘要，网页和技术报告），科学专业诚信的讨论，到当地雇主或计算中心（如NSA，NASA或NIH）的专题实地考察，有关研究生院申请和GRE准备课程的信息，以及与教师，研究生和该计划的访客进行非正式接触的机会。 该计划的所有方面都由学生团队进行，这是参与者从该计划中获得的最重要的经验之一。 科学、统计和并行计算技能对21世纪的劳动力至关重要。 同样重要的是在团队中互动和沟通的技能，以及跨学科项目的工作。 这个REU网站将提供一个综合的计划，在教师导师和研究生助理的指导下整合这些元素，其中许多人通过跨学科研究和咨询中心（circ.umbc.edu）拥有以客户为中心的跨学科咨询的丰富经验。 高性能计算环境将由NSF支持的UMBC高性能计算设施（hpcf.umbc.edu）中的最先进的分布式内存集群提供。 其目前的集群maya拥有240个节点，配备现代多核CPU，并包括混合节点，配备为科学计算而设计的最先进的大规模并行GPU以及尖端的60核Intel Phi加速器。 REU网站的参与者将受益于一个完整的专业发展经验，涉及科学，统计和并行计算，跨学科的研究项目，以及专业技能的综合发展。 这些技能对未来的劳动力和美国的国家竞争力至关重要。 该网站由国防部与NSF REU计划合作支持。

The REU Site: Interdisciplinary Program in High Performance Computing in the Department of Mathematics and Statistics at the University of Maryland, Baltimore County (UMBC) is an eight-week residential summer program for undergraduate students majoring in mathematics, statistics, or computer science. This REU Site uses activities that can be grouped into three principal components to develop skills of the future workforce: (1) Structured training in a certificate program teaches fundamental skills in high performance computing through lectures and TA-supported computer labs, with specific tools of Linux, C, MPI, MATLAB, and R. (2) The skills in scientific, statistical, and parallel computing are applied to interdisciplinary research projects posed by scientists from industry, government agencies, or departments outside of mathematics and statistics. (3) The teaching and research are supported by a complete set of highly integrated professional development activities, including team work experience, presentation skills (poster, oral, project update, and 'elevator pitch'), writing skills (abstracts, webpage, and technical report), discussion of professional integrity in the sciences, topical field trips to local employers or computing centers such as the NSA, NASA, or NIH, information on graduate school application and a GRE preparation course, and opportunities for informal contact with faculty, graduate students, and visitors to the program. All aspects of the program are conducted by student teams, which is one of the most important experiences participants take away from the program. Scientific, statistical, and parallel computing skills are critical to the 21st century workforce. Equally crucial are skills in interacting and communicating in teams, and working on interdisciplinary projects. This REU Site will provide a comprehensive program integrating these elements under the guidance of the faculty mentors and graduate assistants, many of whom have extensive experience with client-focused interdisciplinary consulting through the Center for Interdisciplinary Research and Consulting (circ.umbc.edu). The high performance computing environment will be provided by a state-of-the-art distributed-memory cluster in the NSF-supported UMBC High Performance Computing Facility (hpcf.umbc.edu). Its current cluster maya has 240 nodes with modern multi-core CPUs and includes hybrid nodes with state-of-the-art massively parallel GPUs designed for scientific computing as well as cutting-edge 60-core Intel Phi accelerators. The REU Site participants will benefit from a complete professional development experience involving scientific, statistical, and parallel computing, interdisciplinary research projects, and integrated development of professional skills. These skills are of vital importance to the workforce of tomorrow and for the national competitiveness of the United States. The site is supported by the Department of Defense in partnership with the NSF REU program.