Two ways to exploit chips with a very large number of transistors are multicore processors and programmable logic chips. Some data parallel algorithms can be executed efficiently on ordinary parallel computers, including multicores. A class of data parallel algorithms is identified which have characteristics that make implementation on multiprocessors inefficient, but they are well suited for direct design as digital circuits. This leads to a programming model called circuit parallelism. The characteristics of circuit parallel algorithms are discussed, and a prototype system for supporting them is described.
@InProceedings{odonnell:DagSemProc.07361.4, author = {O'Donnell, John}, title = {{Parallelism through Digital Circuit Design}}, booktitle = {Programming Models for Ubiquitous Parallelism}, pages = {1--9}, series = {Dagstuhl Seminar Proceedings (DagSemProc)}, ISSN = {1862-4405}, year = {2008}, volume = {7361}, editor = {Albert Cohen and Mar{\'\i}a J. Garzar\'{a}n and Christian Lengauer and Samuel P. Midkiff}, publisher = {Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik}, address = {Dagstuhl, Germany}, URL = {https://drops.dagstuhl.de/entities/document/10.4230/DagSemProc.07361.4}, URN = {urn:nbn:de:0030-drops-13724}, doi = {10.4230/DagSemProc.07361.4}, annote = {Keywords: Circuit parallelism, data parallelism, FPGA} }
Feedback for Dagstuhl Publishing