New Scholar cluster brings supercomputing to Purdue classrooms

August 13, 2013

A portion of the Carter community cluster, which faculty and their students use for computationally demanding research, is now available to give Purdue students in undergraduate courses introductory experience with high-performance computing.

Dubbed the Scholar cluster, the resource is open to Purdue instructors from any field whose classes include assignments that could make use of supercomputing, from high-end graphics rendering and weather modeling to simulating millions of molecules and exploring masses of data to understand the dynamics of social networks.

Faculty can get help using the system for classes and preparing their students to do so from ITaP Research Computing (RCAC) staff. Assistance in integrating supercomputing into instructional plans is likewise available from ITaP Teaching and Learning Technologies staff.

For information, contact ITaP Research Computing at rcac-help@purdue.edu; or Pat Reid, manager of teaching and learning initiatives for ITaP, 4-9371 or patreid@purdue.edu; or Akesha Horton, senior educational technologist for ITaP, 6-3499 or akesha@purdue.edu; or visit the ITaP Teaching and Learning Technologies course redesign website.

The Carter community cluster, including the nodes now dedicated for undergraduate instructional use, was among the top 500 supercomputers in the world when the latest rankings were released in June 2013 and was the fastest system for use on a single U.S. campus when it went online in 2012.

The Scholar cluster, while capable of it, is not intended for larger, longer-running research jobs, says ITaP’s Andy Howard, a high-performance computing specialist.

“The focus is on the instructional aspect,” Howard says, “smaller, shorter jobs for class assignments and projects to give undergraduate students an introduction to using supercomputing in preparation for their careers or graduate school.”

As with Purdue research labs using the Carter cluster, instructors employing Scholar in classes will have access to the system that covers their students as well. Students in the class will then be able to sign on simply by using their Purdue user names and passwords.

Scholar includes a software suite similar to Carter and the other Purdue Community Cluster Program supercomputers, including versatile applications such as MATLAB and the R statistical computing environment.

Originally posted: August 13, 2013