SRU acquires high performance computer cluster

Group with cluster

Alex Gessinger, a junior computing major from Murrysville, joins Nitin Sukhija, Slippery Rock University assistant professor of computer science, and Zach Miller, a graduate student majoring in computing from Beaver, in working on SRU’s first high performance computer cluster. The HPC cluster will allow students across SRU’s scientific departments to conduct computational research.

Dec. 18, 2017

SLIPPERY ROCK, Pa. - Slippery Rock University has obtained its first high performance computer cluster that will allow students across the University's scientific departments to conduct advanced computational research.

The new HPC cluster, called the Lava Cluster, was recently installed in the Maltby Center, thanks to a collaboration with XSEDE, a virtual organization funded by the National Science Foundation that integrates and shares infrastructure and advanced digital services with national researchers and educators to support science.

"We are thrilled as this marks the start of the high performance computing research and education era at SRU," said Nitin Sukhija, SRU assistant professor of computer science, who initiated the project and serves as the University's liaison to the XSEDE team. "The cluster will not only benefit the students in learning parallel and distributed programming models but also will help faculty, staff and students get involved in interdisciplinary research, thus enabling discoveries in diverse domains through high performance computing."

The HPC cluster is theoretically capable of computing 9.8 trillion floating-point operations per second, or TFLOPS; it has 2.25 terabytes of random-access memory, which is about 280 times as much as a typical laptop computer; and it has 30 terabytes of storage space, not counting backups.

According to Sukhija, the supercomputing resources will enable SRU students to solve and optimize the performance of scientific and big data applications pertaining to a variety of domains, such as fluid dynamics, earthquake simulation, cyber resilience and community health.

Several individuals from SRU worked on the project proposal and implementation, which involved a collaborative effort with a XSEDE Cyberinfrastructure Resource Integration team from Indiana University Bloomington and an engineer from Dell. In addition to Sukhija, SRU contributors included Jerry Chmielewski, dean of the College of Health, Environment and Science; Deborah Whitfield, professor of computer science; and Henry Magusiak, director of enterprise technologies.

The immediate goal is to incorporate the parallel programming education and research in the computer science department. Training will be available to other departments and staff beginning in 2018.

Prior to having the high performance-computing cluster, curriculum and resources at SRU were based on single core, or serial computing.

"Without parallel computing, keeping the complexity and resource consumption of serial computing in mind, we cannot take the modern computing era even one step ahead," Sukhija said. "In the current era, where everyone uses multiple cores that exist in their phones and laptops, it is of paramount importance that the students learn the power of parallel computing in their classroom settings, where they will not only learn the benefits of parallel programming but also the application of parallel computing in diverse domains, thus enhancing their interdisciplinary research and education."

The XSEDE group will be assisting SRU with the installed cluster management software toolkit for the next four years at no cost and will aid in managing the cluster remotely.

MEDIA CONTACT: Justin Zackal | 724.738.4854 | justin.zackal@sru.edu