Issue No. 08 - August (1993 vol. 4)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/71.238620
<p>The Semantic Network Array Processor (SNAP) is a parallel architecture for knowledgerepresentation and reasoning that uses the marker-propagation paradigm. The primaryapplication areas of SNAP are natural language understanding and speech processing. Afirst-generation SNAP-1 system has been designed and constructed using an array of 144digital signal processors organized as 32 multiprocessing clusters with dedicatedcommunication units, a tiered synchronization scheme, and multiported memory network.Issues in the design, performance, and scalability of a marker-propagation architectureare addressed.</p>
Index TermsSNAP-1 parallel AI prototype; Semantic Network Array Processor; parallel architecture;knowledge representation; reasoning; marker-propagation paradigm; natural languageunderstanding; speech processing; digital signal processors; multiprocessing clusters;dedicated communication units; tiered synchronization scheme; multiported memorynetwork; inference mechanisms; natural languages; parallel architectures; parallelmachines; semantic networks; special purpose computers; speech analysis andprocessing
R. DeMara and D. Moldovan, "The SNAP-1 Parallel AI Prototype," in IEEE Transactions on Parallel & Distributed Systems, vol. 4, no. , pp. 841-854, 1993.