GAMING

MATLAB is an important data analysis tool for many TeraGrid users and, as a parallel resource, it has the potential to expand the high performance computing user community.

“MATLAB on the TeraGrid will help enable a broader class of researchers who are well-versed in MATLAB to reduce the time to solution in a scalable manner without having to become parallel programming experts,” said Cornell University’s senior vice provost for research Robert Buhrman. “It will serve as a complementary experimental component to NSF’s large-scale eXtreme Digital vision and TeraGrid Science Gateways, and be a valuable tool to researchers focused on solving complex problems in the environment, healthcare, and many other science and engineering disciplines,” he added. 

 “This project is designed to advance the understanding of how to best deploy a software utility as a transparent and responsive user service,” said CAC director and principal investigator David Lifka. “It will demonstrate an important working model for high-performance utility computing, which may encourage other software vendors to explore and develop similar research capabilities.”

Access to MATLAB on the TeraGrid will not require knowledge of a specific operating system, batch scheduler, or message passing library. Interactive ease-of-use will be facilitated by leveraging Parallel Computing Toolbox and MATLAB Distributed Computing Server to seamlessly access the distributed computing resource through remote desktops and TeraGrid Science Gateways.

 “MATLAB on the TeraGrid will be made available in its initial configuration as a 512-core experimental computing resource to researchers with TeraGrid certificates and through Science Gateways such as nanoHUB.org,” said Gerhard Klimeck, director of the Network for Computational Nanotechnology at Purdue University who is co-PI on the project along with Michael McLennan, a senior research scientist at Purdue. “nanoHUB users will benefit from this new TeraGrid resource through a transparent and instantaneous access for several applications,” added McLennan.

The project team plans to enable two usage paradigms on the TeraGrid:

·        Enable real-time, interactive scaling from desktop systems running MATLAB to the experimental TeraGrid resource. Researchers will connect from remote desktops running MATLAB Parallel Computing Toolbox to the TeraGrid 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 will be able to use the same utility cluster at Cornell.

·        Provide a parallel MATLAB engine for Science Gateways such as nanoHUB.org, a web-based resource for research, education and collaboration in nanotechnology created by the NSF-funded Network for Computational Nanotechnology. Science Gateway users will access discipline-specific web portals and through simulation inputs via a web form launch MATLAB simulations and get timely results without knowledge of the underlying software or hardware infrastructure. This will allow researchers to leverage the TeraGrid as research tool without first overcoming a platform-specific learning curve.

MATLAB on the TeraGrid is being deployed at the Ithaca, NY campus of Cornell University on a Dell PowerEdge high performance computing cluster. Current and next-generation scientists will be educated on how to use the resource as an extension of their desktops and how to integrate it with Science Gateways. Annual classroom workshops will be held at Cornell and Purdue; virtual workshops will be available 24x7.

MATLAB on the TeraGrid is funded through the National Science Foundation Office of Cyberinfrastructure (OCI) and grants from Dell, Microsoft Corporation, and The MathWorks.

The TeraGrid is an NSF-sponsored open scientific discovery infrastructure that unites people, resources, and services to enable discovery in U.S. science and engineering.

For more information on MATLAB on the TeraGrid, visit www.cac.cornell.edu/matlab.