{"id":3014,"date":"2009-10-08T18:07:03","date_gmt":"2009-10-08T22:07:03","guid":{"rendered":"https:\/\/www.bu.edu\/tech\/research\/scv_import\/about\/mission\/"},"modified":"2018-11-30T11:52:29","modified_gmt":"2018-11-30T16:52:29","slug":"mission","status":"publish","type":"page","link":"https:\/\/www.bu.edu\/tech\/support\/research\/rcs\/mission\/","title":{"rendered":"Mission"},"content":{"rendered":"

The Research Computing Services (RCS) group within Information Services & Technology at Boston University provides computing and visualization resources and services to support research that has specialized or highly intensive computation, storage, bandwidth, or graphics requirements. Typical applications -include scientific and engineering simulation, data analysis, and visualization.<\/p>\n

Resources are managed in close consultation with the Research Computing Governance Committee, the BU Center for Computational Science (ccs.bu.edu), and the Rafik B. Hariri Institute for Computing and Computational Science & Engineering (bu.edu\/hic).<\/p>\n

MGHPCC<\/h2>\n

Boston University is a founding member of the Massachusetts Green High Performance Computing Center (MGHPCC), a collaboration of universities, industry, and Massachusetts state government. The MGHPCC operates a state-of-the-art data center in Holyoke, Massachusetts, specifically designed to address the research computing needs of its members. Utilizing the abundant, clean, renewable energy from Holyoke Gas & Electric\u2019s hydroelectric power plant on the Connecticut River, the data center features a 33,000-square-foot computer room, a 19MW power feed, and a high-efficiency cooling plant that uses outside air to provide 70 percent of the cooling. The MGHPCC has been awarded LEED Platinum certification.<\/p>\n

The partners include university members Boston University, Harvard, MIT, Northeastern, and the University of Massachusetts; industry partners Cisco and EMC; and the Commonwealth of Massachusetts.<\/p>\n

Service Models<\/h2>\n

RCS offers several service models\u2014including shared, buy-in\/coop, dedicated, and co–located systems\u2014to centrally support research computing and offer researchers a full spectrum of computing options ranging from -university-wide, fully shared resources to dedicated, -individually owned-and-operated machines. In most cases, the systems are centrally managed by Research Computing staff; in all cases the physical infrastructure\u2014space, power, cooling, and core networking at the MGHPCC\u2014is provided by the University without charge-back to individual researchers or departments.<\/p>\n

The shared<\/b> service model applies to equipment acquired with a significant University contribution, either fully funded centrally or under an institutional-level infrastructure grant leveraged by substantial matching funds. These computing resources are offered without charge to all faculty and research staff on a fair-share allocation basis; allocations are reviewed by a committee of faculty and staff. The University currently maintains the Shared Computing Cluster (SCC) as a shared service.<\/p>\n

The buy-in<\/b> program allows researchers to acquire additional, standardized compute servers to support their individual research projects. The additional resources are integrated into the shared facility and managed centrally by the RCS group, and the owners of the equipment are given priority access while any excess compute capacity is returned to the pool for general, shared use. Owners determine the access and queuing policies for their portion of the facility. Storage may also be purchased under the buy-in program. All other standard services are -provided without charge.<\/p>\n

Computation Facilities<\/h2>\n

Located at the MGHPCC, the Shared Computing Cluster is a heterogeneous Linux cluster composed of both shared and buy-in components. The system currently includes over 2500 shared processors, 4400 buy-in processors, a combined 236 GPUs, and over two petabytes of storage for research data.<\/p>\n

A wide range of programming languages, parallelizing compilers, mathematical and scientific libraries, graphics and visualization software, and discipline-specific application packages is maintained on the Cluster.<\/p>\n

In addition to the high-performance storage system for use in computation, Information Services & Technology offers a Data Archiving Service for secure, long-term storage of infrequently accessed data.<\/p>\n

Networking<\/h2>\n

Boston University has a long history of running high-performance networks both within the campus and in connecting with the New England region and the world.<\/p>\n

Boston University is a member of Internet2, which maintains national networks (100 Gb\/s) in support of research and education. The University is also a founding member of Northern Crossroads (NoX), which maintains regional networks (10 Gb\/s) and serves as a high-performance exchange for New England participants and commodity Internet service providers. The NoX operates a node at the MGHPCC and provides a Meet-Me interconnect for member equipment installed at the MGHPCC.<\/p>\n

The University is connected to the Metro Ring\u2014144 optical fibers over a 7.4-mile radius through Boston and Cambridge.<\/p>\n

The campus core utilizes 10-Gigabit Ethernet with multiple 10-Gigabit links to the Boston\/Cambridge NoX node and a pair of 10-Gigabit connections to the MGHPCC. RCS uses InfiniBand, 10-Gigabit Ethernet, and Gigabit Ethernet to interconnect its computing systems.<\/p>\n

Visualization<\/h2>\n

RCS provides graphics and virtual reality facilities as well as staff expertise to support a broad range of research application areas.<\/p>\n

The Computer Graphics Lab houses Windows and Linux workstations running scientific visualization, molecular design, graphics, modeling, and animation software. <\/p>\n

Consulting<\/h2>\n

Research Computing’s broadly trained staff assist members of the research and education community to advance research and education.<\/p>\n

Scientific programming consultants assist in program parallelization, performance
\nmeasurement, and code tuning, as well as with numerical methods and computational techniques.<\/p>\n

System consultants help in the use of the supercomputing systems and their various programming tools and packages, as well as data management.<\/p>\n

Graphics consultants assist researchers in developing visualizations, virtual environments, and graphics problem solving. They cover the use of graphics and visualization packages, the development of custom computer graphics tools, the application of graphics algorithms, the conversion of data into effective graphical representations, and the creation of virtual environments.<\/p>\n

Training<\/h2>\n

RCS offers live and online tutorials customized for the specifics of BU\u2019s installation. A live tutorial series is offered each semester covering programming languages, high-performance computing, mathematics and data analysis, and scientific visualization. On request, Research Computing staff will customize presentations for courses and seminars and host workshops, user groups, symposia, and vendor training sessions.<\/p>\n

Software<\/h2>\n

Software on the SCC<\/a> supports the development of parallel programs using data-parallel, message-passing, and shared-memory paradigms. Parallelizing compilers for FORTRAN and C are available, as are mathematical and scientific subroutine libraries. Other packages are available for distributed computing, statistics, scientific visualization, modeling and rendering, animation, image manipulation, and plotting. Comprehensive help pages on this software are written specifically for getting started on the SCC.<\/p>\n

Programming Languages<\/h5>\n