My primary research these days focuses on the use of graphics processing units (GPUs) in scientific computations. Specifically the application of high performance computing for structural biology applications. Technologies such as CUDA, Stream and OpenCL have made accessing the processing power of the GPU much easier. OpenCL, an open specification, in particular, has been established as a standard for accessing not only GPUs but any hardware specific feature that possesses a programmable pipeline.
In addition to being an open standard, OpenCL has been designed, from the ground up, to be compatible with OpenGL (the open standard graphics programming API). The integration with of OpenCL with OpenGL allows for the sharing of data between the two technologies (i.e. computation and display) in a seamless and highly efficient manner.
I’ve created a series of tutorials on OpenCL for MacResearch.org. If you are interested in the technology or how to get started using it, you may find these presentations interesting. You can watch the entire series below.
OpenCL Tutorials
Episode 1 - Introduction to OpenCL
Episode 2 - OpenCL Fundamentals
Episode 3 - Building an OpenCL Project
Episode 4 - Memory Layout and Access
Episode 5 - Questions and Answers
Episode 6 - Shared Memory Kernel Optimization
Tutorial Files
OpenCL Files
File size: 16 MB