Comput Sci Res Dev (2011) 26: 317–324 DOI 10.1007/s00450-011-0172-2 SPECIAL ISSUE PAPER A system architecture supporting high-performance and cloud computing in an academic consortium environment Michael Oberg · Matthew Woitaszek · Theron Voran · Henry M. Tufo Published online: 6 May 2011 © Springer-Verlag 2011 Abstract The University of Colorado (CU) and the Na- tional Center for Atmospheric Research (NCAR) have been deploying complimentary and federated resources support- ing computational science in the Western United States since 2004. This activity has expanded to include other partners in the area, forming the basis for a broader Front Range Computing Consortium (FRCC). This paper describes the development of the Consortium’s current architecture for federated high-performance resources, including a new 184 teraflop/s (TF) computational system at CU and prototype data-centric computing resources at NCAR. CU’s new Dell- based computational plant is housed in a co-designed pre- fabricated data center facility that allowed the university to install a top-tier academic resource without major capital facility investments or renovations. We describe integration of features such as virtualization, dynamic configuration of high-throughput networks, and Grid and cloud technologies, into an architecture that supports collaboration among re- gional computational science participants. M. Oberg ( ) · M. Woitaszek · H.M. Tufo National Center for Atmospheric Research, 1850 Table Mesa Drive, Boulder, CO 80305, USA e-mail: oberg@ucar.edu M. Woitaszek e-mail: mattheww@ucar.edu H.M. Tufo e-mail: tufo@cs.colorado.edu T. Voran · H.M. Tufo University of Colorado, Boulder, UCB 430, Boulder, CO 80309, USA T. Voran e-mail: theron.voran@colorado.edu Keywords High-performance computing · Data-centric computing · Regional and federated supercomputing initiatives 1 Introduction Access to state-of-the-art computational facilities is essen- tial for a wide range of computation-driven science disci- plines and computational science research and education programs. Often, the demands for high-performance com- puting (HPC) resources quickly outstrip the ability of a sin- gle project, group, or even organization to satisfy indepen- dently. Moreover, as the resources, software applications, and collaborative projects increase in size and complexity, the ability for batch scheduling and manual data manage- ment techniques to meet the diverse requirements dimin- ishes, and advanced workflow technologies are needed to appropriately map computational requirements to the avail- able systems and infrastructure. The development of computing consortiums among peer institutions allows each institution to better support its re- searchers through an increase in the diversity of available resources and technical capabilities on those resources. By dynamically coupling distinct resources, and then support- ing data-centric and multi-resource workflows, the consor- tium provides the foundation for large-scale computational science and collaborative research. Consortium participants can augment each other’s resources and technical expertise while still retaining control over their individual resources, thus establishing a continuum of resource availability and infrastructure development and growth. The consortium en- vironment also lays a common substrate for addressing the technical hurdles common in running large computer sys- tems in cross-organization collaborations. Additionally, a