This section explains how to use the XLL+ toolkit with the CUDA developer's toolkit.
The CUDA developer's toolkit allows programmers to use an Nvidia Graphical Processing Unit (GPU) to accelerate the performance of mathematics-intensive code, by applying massive parallelization. Using the techniques described in this section, it is possible to create add-ins that achieve huge performance gains: improvements of over 100x have been regularly achieved by using the GPU for parallel processing.
For more information please see the Nvidia developer zone at http://developer.nvidia.com/cuda-toolkit.
A sample project that uses CUDA is included in the XLL+ toolkit: AvgOptCuda.