SOURCE: Interactive Supercomputing

November 13, 2008 09:00 ET

TOMLAB and Star-P Extend Fast and Easy Desktop Optimization Solvers to HPCs

WALTHAM, MA--(Marketwire - November 13, 2008) - Interactive Supercomputing Inc. (ISC) has teamed with Tomlab to provide a next generation of technical computing solutions for tackling hard-to-solve optimization problems by utilizing high performance computing clusters (HPCs).

Tomlab has integrated the TOMLAB Optimization Environment for MATLAB® with ISC's Star-P® software platform. TOMLAB facilitates the development of robust algorithms to solve a wide variety of optimization problems, from stock market predictions or airline flight schedules to optimal trajectory control solutions for mission planning. TOMLAB supplies more than 100 different algorithms for linear, discrete, global and nonlinear optimization, as well as a large number of other solvers, including advanced tools for general optimal control.

When combined with Star-P, TOMLAB users can extend their optimization and optimal control models to powerful parallel processing clusters and servers to gain dramatic performance improvements. Star-P eliminates the need to re-program MATLAB code using complex languages such as C, FORTRAN and MPI (Message Passing Interface) in order for them to run on parallel clusters and servers; a process which would otherwise require arcane programming knowledge and months to complete.

For example, a top 10, $100+ billion money management firm that is also one of the world's premier hedge fund managers, has enabled its analysts to perform portfolio optimization using TOMLAB and Star-P On-Demand, a pay-per-use HPC resource "in the cloud." Star-P On-Demand is an on-line service that combines HPC hardware and the Star-P parallel computing software platform in a single, on-demand solution, allowing the user to easily connect to a high performance system from their desktop.

The TOMLAB integration with Star-P is significant because it allows for large-scale optimization that is otherwise not achievable. TOMLAB brings fast algorithm development and modeling of applied optimization problems, while Star-P automatically parallelizes them to run on multi-processor HPCs. Users who have "hit the wall" performance-wise on their single processor desktops can now solve much larger and more complicated Monte Carlo simulations, forecasting and other optimization applications, boosting their productivity and out-performing their competitors.

"The TOMLAB integration with Star-P is the next logical step and a monumental advantage for our customers," said Prof. Dr. Kenneth Holmström, founder of Tomlab. "Working with the ISC development team has convinced us that our users stand to gain an enormous speed-up for large-scale and distributed models. This is part of Tomlab's future and we will strive to continuously expand our offerings with ISC."

About Interactive Supercomputing

Interactive Supercomputing (ISC) launched in 2004 to commercialize Star-P, an interactive parallel computing platform. With automatic parallelization and interactive execution of existing desktop simulation applications, Star-P merges two previously distinct environments -- desktop computers and high performance servers -- into one. Based in Waltham, Mass., the privately held company markets Star-P for a range of biomedical, financial, and government applications. For additional information, visit http://www.interactivesupercomputing.com.

Star-P is a registered trademark of Interactive Supercomputing Inc. All other trademarks mentioned herein are the property of their respective owners.

Contact Information