Very informative article about Heterogeneous Programming by Michael Wolfe, Compiler Engineer, The Portland Group, Inc.
The heterogeneous systems of interest to HPC use an attached coprocessor or accelerator that is optimized for certain types of computation. These devices typically exhibit internal parallelism, and execute asynchronously and concurrently with the host processor. Programming a heterogeneous system is then even more complex than “traditional” parallel programming (if any parallel programming can be called traditional), because in addition to the complexity of parallel programming on the attached device, the program must manage the concurrent activities between the host and device, and manage data locality between the host and device.
- concurrency and parallelism (nullptr.me)
- Parallel Computing Terms (nullptr.me)
- Parallel Computing (nullptr.me)
- PGI’s Michael Wolfe on Heterogeneous Programming (insidehpc.com)
- Microsoft publishes fancy-pants heterogeneous parallel GPGPU C++ AMP specification (arstechnica.com)
- Video: AMD – The Future of Heterogeneous Processing (insidehpc.com)