University of Minnesota Supercomputing Institute
The University of Minnesota Supercomputing Institute (MSI) in Minneapolis, Minnesota is an interdisciplinary research program that provides hardware and software resources, as well as technical user support, to faculty and researchers at the University of Minnesota and at other institutions of higher education in Minnesota. MSI is located on the University of Minnesota’s Twin Cities campus in Walter Library.
History
In 1981, the University of Minnesota was the first U.S. University to acquire a supercomputer (a Cray-1). The Supercomputing Institute was created in 1984 to provide high-performance computing resources to the University of Minnesota's research community.
Claudia Neuhauser is Interim Director of the Supercomputing Institute. Professor Neuhauser is also Director of the University of Minnesota Informatics Institute.
MSI currently has three HPC systems available for use by researchers. In February 2015, a fourth system began installation and acceptance testing.
Memberships
Currently MSI is a member of the Minnesota High Tech Association, the Great Lakes Consortium, and XSEDE.
Mission
The Minnesota Supercomputing Institute seeks to provide researchers at the University of Minnesota and at other institutions of higher education in the State of Minnesota access to high-performance computing resources and user support to facilitate successful and cutting-edge research in all disciplines, help researchers attract funding, contribute to undergraduate and graduate education, and benefit the broader community.
MSI is committed to expanding and developing the types of service it offers in order to continue to play its key support role across the growing spectrum of scientific fields.
MSI is also committed to facilitating University-industry collaboration and to promoting technology transfer through the interchange of ideas in the field of supercomputing research, including the dissemination of results of research accomplished with MSI resources.
Supercomputing capabilities
HPC resources
- "Mesabi": Mesabi is an HP heterogeneous system. It is a distributed cluster featuring a large number of nodes with leading edge Intel processors that are tightly integrated via a very high speed communication network. It is currently being installed and tested.
- "Cascade": Cascade is a mixed GPU and Phi cluster. There are eight compute nodes with Tesla GPUs and four nodes with dual Kepler GPUs. In addition, there are three nodes that feature a Phi coprocessor.
- Itasca: Itasca is an HP Linux cluster with 1,091 HP ProLiant BL280c G6 blade servers, each with two quad-core 2.8 GHz Intel Xeon X5560 "Nehalem EP" processors sharing 24 GiB of system memory, plus 51 Proliant BL460c G8 blade servers, each with two eight-core 2.6 GHz E5-2670 Xeon "Sandy Bridge EP" processors with 64, 128, or 256 GiB of memory, with a 40-gigabit QDR InfiniBand (IB) interconnect. In total, Itasca consists of 9,712 compute cores and 25 TiB of main memory.
- Red Nodes: Red Nodes is a Hadoop cluster to support Big Data analytics. It is composed of 50 nodes, each with "Sandy Bridge EP" E5-2620 processors with six cores at 2 GHz, QDR InfiniBand, eight GB of memory, and a 500 GB hard drive. MSI has configured 40 of these compute nodes into two Hadoop clusters, each with 20 nodes.
Laboratories
- Basic Sciences Computing (BSCL)
- Computational Genetics (CGL)
- Biomedical Modeling, Simulation, and Design (BMSDL)
- LCSE-MSI Visualization Laboratory (LMVL)
- Scientific Development and Visualization (SDVL)
References
- Moore, Rick. "Blade Runner : UMNews." University of Minnesota. Web. 29 July 2010. http://www1.umn.edu/news/features/2009/UR_CONTENT_148391.html
- Vance, Ashlee. "Minnesota’s Enormous Apples Computer - Bits Blog - NYTimes.com." Technology - Bits Blog - NYTimes.com. Web. 29 July 2010. http://bits.blogs.nytimes.com/2009/12/10/minnesotas-enormous-apples-computer/?smid=pl-share