High Performance Computing (HPC) Services at Pace University
What is High Performance Computing?
High Performance Computing (HPC) describes a variety of computing services typically used for scientific research or computational problems that are either too complex for standard computers or would take too long to run. Rather than a single processor or CPU, the HPC system is comprised of multiple or clustered systems, each with one or more processors and its own memory. Programs developed for HPC are divided into smaller pieces so that each piece can be executed simultaneously by separate processors, allowing HPC to harness the power of parallelism or the ability to work on many computational tasks at the same time.
What hardware is available at Pace for High Performance Computing?
In September 2010, Pace received a donated high performance computer cluster from Blue Sky Studios. It is comprised of 54 blade servers (39 active), each with dual AMD 2.4 processors, 12 GB of RAM and 72 GB of disk space. The hardware resides in a secured, climate controlled environment in the University’s server farm.
Who oversees the High Performance Computing Cluster?
ITS staff members are responsible for all system software and hardware maintenance, troubleshooting and repair of the HPC cluster. There is a HPC team comprised of ITS staff, and faculty members from the Dyson College of Arts and Sciences and the Seidenberg School of CSIS. The team meets regularly to discuss possible enhancements to or expansion of HPC resources, as well as the use of HPC to promote educational initiatives and support research at the University.
What are the future plans for HPC at Pace?
The team recently submitted and received approval for a grant to fund a three dimensional visualization and passive stereographic projection system. (Need to add the newly approved grant here, too.) We are also exploring the possibility of harnessing idle CPU cycles on desktops throughout the University to run HPC compute jobs. We continue to look for opportunities to introduce HPC to students, faculty and staff and welcome suggestions from the Pace community.