Silicon Informatics and UTSA to develop, commercialize scalable computing apps

Photo: freedigitalphotos.net

By Christi Fish
Associate Director of Media Relations

(March 5, 2013) -- Edina, Minn.-based start-up Silicon Informatics Inc. has been awarded a Small Business Technology Transfer (STTR) contract by the U.S. Army Research Office to transition recent university research in highly scalable parallel random number generation into products for high-performance computing applications.

Scholars from The University of Texas at San Antonio (UTSA) and Florida State University (FSU) will participate in the research, which will ultimately lead to the development and commercialization of software tools that can help software applications realistically mimic complex phenomena.

The STTR program, established by Congress in 1992, supports the commercialization of basic scientific research by awarding contracts to collaborating small businesses and nonprofit research institutions for research and development projects. This is the first Phase II STTR grant for Silicon Informatics and UTSA.

Random number generation has long been crucial for the realistic computer modeling of complex phenomena in science and industry. Its use extends from cryptography, astrophysics and weather phenomena to the simulation of turbulent behavior in internal combustion engines, risk in financial instruments and portfolios, animation for films and video games, and more.

Nature is highly random: No two snowflakes or trees are exactly alike. Hence, the ability to numerically model natural and derivative artificial phenomena (e.g., engines, animated landscapes and biological processes) realistically can be limited by the scale at which randomness (random numbers) can be generated.

In recent years, the availability of highly parallel, extremely powerful computers with many thousands of processors has created the potential to dramatically increase the realism of modeled phenomena. To help realize this potential, random number generators need to become much more parallel so that they can exploit many more processors and "compute threads" than is possible today.

Phase I of the STTR project led to two breakthroughs. First, UTSA researchers developed a method to evaluate the quality of random numbers among the thousands of parallel streams.  Second, UTSA demonstrated a new, context aware pseudo random number generator that was shown to retain its integrity through billions of parallel streams. Current random number generators only retain their integrity through 1.5 million streams.

Led by Professor Raj Boppana in the Department of Computer Science and UTSA scholars including professors Ram Tripathi and Ravi Sandhu will develop new statistical tests to evaluate the quality of the random number generators that produce billions and trillions of high-quality individual random number streams. They also will develop random number generators that can be incorporated securely into new computer modeling software that can be used by developers, multicore processors, cluster computers and GPU-based parallel computers. Silicon Informatics will optimize and market the software for broad commercial use.

"UTSA and FSU are the two universities in the United States with the expertise to undertake this project with us, and I am thankful that both are part of our team," said Bob Keller, Silicon Informatics president and CEO. "Once the remaining technical goals have been achieved, we plan to offer the solution in a range of commercial markets."

"The extent to which computer modeling can reflect reality is often limited by the quality and scalability of the random number generation methods. The random number generator and the quality evaluation tool developed in this project will help remove this limitation," said Boppana. "We feel very privileged to be selected by Silicon Informatics for this research and expect the methods we create to be applicable to a wide range of industries that model complex behaviors, from entertainment and finance to science and engineering."

------------------------------

About Silicon Informatics

Silicon Informatics is a Minnesota-based company, formed in 2003 by successful entrepreneurs. Since its inception, SI's vision has been to fully exploit emerging, off-the-shelf processors and coprocessors for high-performance computing, parallel data processing and secure communications.

About UTSA

The University of Texas at San Antonio is one of the largest of nine academic universities and six health institutions in the UT System. As a multicultural institution of access and excellence, UTSA aims to be a national research university providing access to educational excellence and preparing citizen leaders for the global environment. UTSA serves nearly 31,000 students in more than 140 degree programs in the colleges of Architecture, Business, Education and Human Development, Engineering, Liberal and Fine Arts, Public Policy and Sciences as well as University College, the Honors College and the Graduate School. Founded in 1969, UTSA is an intellectual and creative resource center and a socioeconomic development catalyst for Texas and beyond.

About the U.S. Army Research Office

This STTR project is supported by the U.S. Army Research Office. This announcement does not necessarily reflect the position or the policy of the government, and no official endorsement should be inferred.