EM Photonics, UD develop advanced algorithms for Air Force
Michela Taufer, assistant professor in the Department of Computer and Information Sciences at UD, and John Humphrey of EM Photonics.
Pictured are, from left, Boyu Zhang, Narayan Ganesan, Louis Fogel, John Humphrey, Michela Taufer and Kyle Benson.

ADVERTISEMENT

UDaily is produced by Communications and Public Affairs
The Academy Building
105 East Main Street
University of Delaware
Newark, DE 19716 • USA
Phone: (302) 831-2792
email: publicaffairs@udel.edu
www.udel.edu/cpa

10:55 a.m., Aug. 23, 2010----EM Photonics and the Department of Computer and Information Sciences at the University of Delaware have been selected by the Air Force Office of Scientific Research (AFOSR) to develop innovative algorithms for scientific computing, modeling and simulation for a multi-GPU environment.

THIS STORY
Email E-mail
Delicious Print
Twitter

Numerous Air Force applications will benefit from this research effort, including electromagnetic modeling, computational fluid dynamics, structural mechanics, and radiation transport, to name a few.

The EM Photonics' GPU computing team is led by John Humphrey, who developed the GPU-accelerated math library CULAtools, and will be working closely with Michela Taufer's group, the Global Computing Laboratory at UD. Taufer is an assistant professor in the Department of Computer and Information Sciences.

The new team has tremendous experience in the massive parallelization of computationally intense algorithms on GPUs and can rely on the deployment of the University's largest supercomputer, code-named “Geronimo,” which is based on a custom GPGPU design utilizing NVIDIA Tesla and Fermi GPU computing technology.

“This is a valuable opportunity for EM Photonics. We have successfully collaborated in the past with Dr. Taufer's group and their familiarity with our CULA library adds great value to this project. We look forward to extending our work in dense matrix solvers on multiple GPUs, as well as researching the feasibility of multi-GPU sparse solvers. With access to a set of GPU libraries that can address the computational needs of so many applications, Air Force engineers can stay focused on their specific projects without having to write software code for advanced GPU architectures,” said Humphrey, head of the GPU computing team at EM Photonics.

“There is a tremendous need today for skilled engineers and computer scientists in this research area. Collaborating with a vibrant and successful company such as EM Photonics opens new opportunities for our students at UD to embrace research in concurrent programming and multi-core architectures. This project provides us the unique opportunity to train our students for future challenges in high performance computing by combining academic and industrial experience,” Taufer said.

The AFOSR invests in long-term, broad-based research into aerospace-related science and engineering. For more information about their research areas, programs and accomplishments, see the website.

About EM Photonics

Headquartered in Newark, Del., EM Photonics is a recognized leader in implementing computationally intense algorithms on commodity hardware platforms. Using specialized computer architectures such as GPUs and FPGAs, EM Photonics accelerates their clients' applications to achieve better, faster results. EM Photonics also offers consulting services, training, and custom-designed tools to commercial, government, and academic organizations seeking to optimize their scientific computing, image processing, and numerical analysis applications.

About CULAtools

CULAtools is EM Photonics' product family comprised of CULABasic, Premium, and Commercial. CULA is a GPU-accelerated implementation of LAPACK - a collection of commonly used linear algebra functions used by millions of developers in the scientific and engineering community. After developing accelerated linear algebra solvers since 2004, EM Photonics partnered with NASA Ames Research Center in 2007 to extend and unify these libraries into a single, GPU-accelerated package. Through a partnership with NVIDIA, the EM Photonics GPU Group focused on developing a commercially available implementation of accelerated linear algebra routines. By leveraging NVIDIA's CUDA architecture, CULA provides linear algebra functions with unsurpassed performance to more than 5,000 users today. CULA 2.0 was released earlier this summer to support NVIDIA's FERMI-based GPUs and is available for download at this website.

About the University of Delaware

The University of Delaware has a great tradition of excellence, from our founding as a small private academy in 1743, to the research-intensive, technologically advanced institution of today. The University received its charter from the state of Delaware in 1833 and was designated one of the nation's historic Land Grant colleges in 1867. Today, UD is a Land Grant, Sea Grant and Space Grant institution. UD also is classified by the Carnegie Foundation for the Advancement of Teaching as a research university with very high research activity -- a designation accorded to less than three percent of U.S. colleges and universities.

Photos by Evan Krape

close
University of Delaware • Newark, DE 19716
publicaffairs@udel.edu • (302) 831-2792 • ©2012
University of Delaware • Newark, DE 19716 • USA • Phone: (302) 831-2792 • © 2013
Comments|Contact Us|Legal Notices