A Parallel Optical Computer Architecture for Large Database and Knowledge Based Systems

Abstract

We propose a parallel electro-optical computer architecture that uses optical devices to exploit the data-level parallelism available in the database and knowledge base systems. The massive amount of data of the large database and knowledge base are represented in two-dimensional space. The proposed system performs concurrent pattern matching operations using three-dimensional space. The execution speed is theoretically estimated and is shown to be potentially orders of magnitude faster than current electronic systems. © Springer-Verlag 2004.