Solution
To answer this challenge, Techniscan ported its proprietary inverse scattering algorithm from a traditional FORTRAN and MPI system to NVIDIA® CUDA™. After testing this configuration, the developers were sold as they realized that they were able to run the system much faster with no sacrifice in the quality of the results.
The Tesla computing system gives the WBU system the horsepower it needs to meet the healthcare industry’s pace. A configuration of four Tesla GPU processors is able to run through Techniscan’s algorithm in less than 20 minutes. By comparison, the traditional 16-processor computer system took more than twice the time.
Techniscan has now replaced every one of the clusters built into the scanners with two massively parallel NVIDIA® Tesla™ D870 deskside GPU Computing systems, which race through the inverse scattering code.
“CUDA has allowed us to tap into the processing power of the GPU with tremendous ease, saving us time and money” says Jim Hardwick, Senior Software Engineer at Techniscan. “A single host system and two Tesla D870s are considerably cheaper than the 16-core cluster. This is significant not only for production and sales, but also reduces research and development costs in engineering.”
|