Fast 3D ZO CRS Stack – An FPGA Implementation of an Optimization Based on the Simultaneous Estimate of Eight Parameters

We show over 200x speedup of our proprietary data intensive application, 3D-ZO-CRS, utilizing a FPGA-based solution, compared to conventional x86_64 CPU cores. As well as performance a second, but no less important, target was to maintain the portability of the basic C/FORTRAN source code for other platforms. A third important point considered in this paper regards maintaining the capability to “recompile” the code in case of small modifications and/or discovery of bugs. In this paper we present our approach and some general considerations regarding the use of such acceleration technology on the data grid environment previously developed for this application.