Issue No. 03 - March (1990 vol. 16)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/32.48936
<p>Much effort has been expended on developing special architectures dedicated to the efficient execution of production systems. While data-flow principles of execution offer the promise of high programmability for numerical computations, it is shown that the data-driven principles can also be applied to symbolic computations. In particular, a mapping of the RETE match algorithm along the line of production systems is considered. Bottlenecks of the RETE match algorithm in a multiprocessor environment are identified and possible solutions are suggested. The modifications to the actor set as well as the program graph design are shown for execution on the tagged data-flow computer. The results of a deterministic simulation of this multiprocessor architecture demonstrate that artificial intelligence production systems can be efficiently mapped on data-driven architectures.</p>
parallel production systems; data-flow principles; high programmability; numerical computations; data-driven principles; symbolic computations; RETE match algorithm; multiprocessor environment; actor set; program graph design; tagged data-flow computer; deterministic simulation; multiprocessor architecture; artificial intelligence production systems; data-driven architectures; expert systems; parallel architectures; parallel programming; symbol manipulation.
J. Gaudiot and A. Sohn, "Data-Driven Parallel Production Systems," in IEEE Transactions on Software Engineering, vol. 16, no. , pp. 281-293, 1990.