Tech-X Corporation announced the release of GPULib v1.0, software to enable more rapid numerical calculations. The product allows users to easily access the computing power of their computer’s graphic processing unit (GPUs) made by NVIDIA to execute vectorized mathematical functions for use with programs including MATLAB and IDL from ITT Visual Information. Speedup factors of 10X to 40X are typical for GPULib use with these applications. The product has potential utility for researchers across a variety of scientific arenas including structural and fluid mechanics, biosciences, medical imaging, and financial engineering. The software is available under two licenses: one for the Open Source community and a second for commercial application development ($950 per platform per developer).
GPULib’s value is the ease of access it provides to the computational power of GPUs. GPUs have a highly parallel structure which makes them more efficient at performing math-intensive tasks than CPUs. Previously, however, the GPU’s power was relatively difficult to access for calculations and integration with other programs as users would have to create customized code, which often fell outside the expertise and focus of scientific researchers. Easier access to GPUs offers a more affordable alternative to increase computation power compared to specialized computer systems.
Tech-X Corporation, based in Boulder, Colorado, develops software solutions for research, engineering and education to solve difficult technical challenges. Visit the company’s website for more information.
