Green Data Center Rackspace Image
»
Special GPU Hybrid Supercomputer Installed in UA Research Data Center
Special GPU Hybrid Supercomputer Installed in UA Research Data Center
Dr. Brant Robertson, and detailed scientific visualization from the new system

University of Arizona assistant professor of Astronomy Brant Robertson is propelling UA’s computational facilities to the forefront of contemporary computer technology. Robertson and collaborators obtained a $1.3 million grant from the National Science Foundation (NSF) Division of Astronomical Sciences through the Major Research Instrumentation program to help fund the implementation of a new supercomputer on UA’s campus. Support from the Office of the Senior Vice President for Research (SVPR) provided vital funding to complete the project. The Extremely LarGe Advanced TechnOlogy system, known as “El Gato,” ranks among the world’s top 500 fastest supercomputers and places 7th in the world for “green” (energy efficient) supercomputing systems on the Green500 list.

The El Gato project is led by Robertson and Co-Principal Investigators professors of Astronomy Feryal Ozel and Dimitrios Psaltis; Paul Cohen, professor and director of the School of Information: Technology, Science, and Arts (SISTA); and Derek Masseth, senior director at University Information Technology Services (UITS). Additional personnel from the College of Science, SISTA, and UITS contributed significantly to the project. El Gato was installed in December 2013 in the UA Research Data Center, a facility available to all UA researchers, supported by the Office of the CIO and the Office of the SVPR and managed and operated by UITS.

El Gato joins the previously existing CPU-based (central processing unit) high performance computers (HPCs) in the Research Data Center. Equipped with GPUs, or graphics processing units—commonly used in smart phones, game consoles, and personal computers—El Gato was created specifically for UA. This unique GPU-CPU hybrid system is designed to be used for its calculation power rather than its ability to display graphics.

While the scientific focal points of El Gato include answering questions in theoretical astrophysics and computer science, any faculty or students performing research at the University can use El Gato through the “windfall” usage policy in place for all the HPCs in the UA Research Data Center—in which their programs run when reserved time is going unused. El Gato account holders may additionally request reserved time for special projects that make use of the system’s unique architecture.

El Gato’s processing power is over 13 times greater than the previous generation of UA HPC systems, which is beneficial to researchers whose work involves advanced computation where multiple calculations must be run simultaneously. The increased processing power of the new computer system allows researchers to run more complex computations at a faster rate and receive more highly detailed results.

Ordinarily, research groups with high performance computer systems use only CPUs, and do not have the accelerator power of the GPUs being used in El Gato. Advanced computational research is often conducted by scientists and researchers through government-funded research facilities called XSEDE resources. However Robertson says, “Many people can’t make full use of these large systems because they are based on technologies that many people don’t have and don’t know how to use.”

The team of Co-PIs, along with their graduate students and postdocs, have developed codes that optimally utilize the new system and address problems ranging from direct imaging of black holes to dynamics of clusters and the formation of galaxies. Robertson hopes El Gato will be used by other UA students, faculty, and researchers to develop code for GPU-based processors—skills which they can then als