Issue No.08 - August (2012 vol.45)
Benedict R. Gaster , Advanced Micro Devices
Lee Howes , Advanced Micro Devices
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MC.2012.257
Heterogeneous parallel primitives (HPP) addresses two major shortcomings in current GPGPU programming models: it supports full composability by defining abstractions and increases flexibility in execution by introducing braided parallelism.
Graphics processing unit, Programming, Performance evaluation, Parellel processing, Indexes, Hardware, Multithreading, hardware, massively threaded computing systems, heterogeneous parallel primitives, braided parallelism, persistent threading, GPGPU programming, data-parallel execution, distributed arrays
Benedict R. Gaster, Lee Howes, "Can GPGPU Programming Be Liberated from the Data-Parallel Bottleneck?", Computer, vol.45, no. 8, pp. 42-52, August 2012, doi:10.1109/MC.2012.257