IT News & Events

News about IT at Indiana University and the world

New HPC at Slippery Rock University

Slippery Rock University begins high performance scientific computing with help from IU

IU and Dell team up to build a new high performance computing cluster at Slippery Rock University.

In early October, the XSEDE Cyberinfrastructure Resource Integration team assisted in building a new HPC cluster (the Lava cluster) at Slippery Rock University (SRU) in Slippery Rock, Pennsylvania. SRU was founded in 1889 with the primary purpose of training teachers. It wasn’t until 1983 that SRU was granted status as a university due to an increased number of degree programs. Since that time, Slippery Rock University has increased its offerings and now will begin offering high performance scientific computing!

The new HPC cluster at SRU, the Lava cluster, will allow the growth of programs based around parallel and high-performance scientific computing. “We are very thrilled here as this marks the start of the high performance computing research and education era at Slippery Rock,” said Dr. Sukhija. While the current stage will focus primarily on the needs of the Computer Science department, the Lava cluster will enable greater collaboration with CS and other scientific departments as well.

HPC build team at Slippery Rock University

From left to right: Henry Magusiak, Director of Enterprise Systems Applications at SRU, Dr. Nitin Sukhija, Assistant Professor of Computer Science at SRU, and Eric Coulter, XCRI Engineer, in front of the Lava cluster.

How did the process begin? After an extended consultation with Dr. Nitin Sukhija, during which XCRI provided feedback on hardware options for a new HPC system, new Dell hardware for a small HPC system was installed at SRU on October 10, 2017. XCRI engineer Eric Coulter, from Indiana University, arrived on the 11th, and began the installation of the XSEDE Compatible Basic Cluster (XCBC) toolkit, while Dell support staff were still on site for the day to assist with any hardware issues.  A week later, Eric, Dell, and the SRU staff had implemented a fully functioning HPC system, for Dr. Sukhija to use for both research and educational purposes. A prime goal of the visit was to provide Dr. Sukhija and the SRU Enterprise IT staff the knowledge needed for efficient HPC administration. Thanks to the hardware support provided by Dell, Eric was able to spend a majority of his time educating site administrators, Dr. Sukhija, and students on HPC system administration. The cluster will enable future curricula at SRU based on parallel computing, which will also leverage toolkits produced by the XCRI team. XCRI will continue to provide remote support and consultation for the growing HPC program at SRU.

The Lava cluster is theoretically capable of 9.8 TFLOPS, has approximately 2.25TB of RAM and 30TB of storage space, not counting backups. Lava is also the first example of a power-managed XCBC, which takes a step towards green computing by automatically powering down compute nodes while they are not being used.