Tech-X Corporation Announces a Library of Mathematical Kernels of Vector Operations for NVIDIA GPUs
[ Back ]   [ More News ]   [ Home ]
Tech-X Corporation Announces a Library of Mathematical Kernels of Vector Operations for NVIDIA GPUs

BOULDER, Colo.—(BUSINESS WIRE)—November 18, 2008— Tech-X Corporation of Boulder, Colorado, today announced the release of GPULib v1.0. This software library executes vectorized mathematical functions on graphics processing units (GPUs) from NVIDIA, bringing high-performance numerical operations to everyday desktop computers.

High-Performance Computing for the Desktop

GPULib allows users to access high-performance computing with minimal modification to their existing programs. By providing bindings for a number of Very High Level Languages (VHLLs) including MATLAB and IDL from ITT Visual Information Solutions, GPULib can accelerate new applications or be incorporated into existing applications with minimal effort. No knowledge of GPU programming and memory management is required. Speedup factors of 10x to 40x are typically realized in IDL and MATLAB applications.

To leverage the compute power of modern, high-performance GPUs, GPULib provides mathematical operations with an easy-to-use interface, allowing researchers to incorporate accelerated numerical computations into their applications. “Many tasks in scientific computing, data mining or image processing involve identical mathematical operations on a large set of data elements,” stated Peter Messmer, Ph.D., vice president of the Space Applications Group at Tech-X Corporation. “Until recently, high-performance computing meant a significant investment in large, specialized computer systems. Now, we can perform these operations quickly on GPUs costing just a couple of thousand dollars that may already be part of your desktop computer.”

"GPULib from Tech-X Corporation is a must have tool for engineers and scientists that need to accelerate their work with parallel computing,” said Andy Keane, general manager of the Tesla computing business at NVIDIA. "Mathematical libraries simplify access to parallel, GPU-based computations to solve complex problems in many areas of technical computing. In addition to C, developers now have the choice of other high level environments like IDL, MATLAB, and other very high level languages for parallel computing.”

About GPULib

GPULib provides a library of mathematical functions that facilitate the use of high performance computing resources available on modern graphics processing units (GPUs) by engineers, scientists, analysts, and other technical professionals.

GPULib is available under two licensing arrangements, depending on your development and use scenario. One supports software development within the Open Source community. The second option provides an avenue for the development of commercial applications based on GPULib. Pricing for the commercial version of GPULib is US$950 per platform per developer.

Development of GPULib was funded in part under NASA SBIR Phase II Award No. NNG06CA13C.

For more information regarding GPULib, please visit

About Tech-X Corporation

Tech-X Corporation of Boulder, Colorado develops software solutions for research, engineering and education to aid with specific scientific challenges. Tech-X Corporation offers commercial products as well as collaborates with government and private institutions on research projects. Products and services from Tech-X help solve difficult technical problems, increasing design and development productivity and accelerating project deployment for scientists and engineers. For more information, visit

© 2008 Tech-X Corporation


Tech-X Corporation
Ed Kase, 303-444-2416
Email Contact