ACADEMIA

The number of users of the NSF-funded “MATLAB on the TeraGrid” experimental computing resource has been steadily increasing since it went into production at SC09. The Cornell University Center for Advanced Computing and Purdue University team have been working closely with these users to continually improve the software interface from the user’s MATLAB client machines to the cluster and to support Science Gateway access.

“We are now able and anxious to allow access to users at a faster rate,” said David A. Lifka, principal investigator on the project and Cornell CAC director. New users interested in using the resource or connecting a Science Gateway to it may request access at www.cac.cornell.edu/matlab/status/interest.aspx.

“Scientists from across the nation are excited about seamlessly accessing parallel MATLAB computational services at Cornell right from their desktops,” said Susan Mehringer, assistant director of consulting. “This resource has generated significant speed-ups on a wide range of applications including earth mantle dynamics, environmental remediation, space systems design, and Hepatitis C virus (HCV) modeling, a major cause of liver disease worldwide.”

The resource was also recently connected to a Science Gateway. Use of the Cornell MATLAB resource is completely transparent to users of the nanoHUB.org Science Gateway, a resource for nanoscience created by the Network for Computational Nanotechnology at Purdue University. Communication protocols have been implemented to enable secure and authenticated data transmission between nanoHUB.org and Cornell. A tool called “NanoNet,” the first of many parallel applications on nanoHUB.org, has been converted to take advantage of the MATLAB resource.

“MATLAB is an important research tool for U.S. scientists and engineers and as a seamlessly-accessible parallel resource it has the potential to broaden the high performance user community,” noted Lifka. “The MATLAB experimental resource is proving valuable not only as a tool for computational research and as a backend for Science Gateways, but also for teaching undergraduate and graduate students how to do computational science,” he added.

Students and researchers connect from remote desktops running MATLAB Parallel Computing Toolbox to the experimental resource running MATLAB Distributed Computing Server. One important feature of the Parallel Computing Toolbox is that it doesn’t matter what operating system the client is using; researchers with Microsoft Windows, Apple Macintosh, or Linux-based clients are able to use the same utility cluster at Cornell.

A two-day workshop on using the resource will be held at Cornell on June 15-16, 2010. In addition, Cornell and Purdue will present a half-day tutorial titled “Running Parallel Simulations and Enabling Science Gateways with the NSF MATLAB Experimental Resource” and Cornell will lead a Birds of a Feather session at the 5th annual TeraGrid Conference, TG’10, which will be held August 2-5, 2010 at the Sheraton Station Square in Pittsburgh, PA.

The experimental MATLAB computing project is funded by the National Science Foundation Office of Cyberinfrastructure (OCI) and grants from Dell, Microsoft, and The MathWorks. Information on the 512-core Dell Windows HPC Server 2008 resource, research applications, education and outreach are available at www.cac.cornell.edu/matlab.

The Cornell University Center for Advanced Computing (CAC) is a leader in high-performance computing systems, applications, and data solutions that enable research success. CAC receives support from Cornell University, the National Science Foundation, and other leading public agencies, foundations, and corporations. For more information, visit www.cac.cornell.edu.