A multi-wavelength optical content-addressable parallel processor (MW-OCAPP) for high-speed parallel relational database processing: Architectural concepts and preliminary experimental system

Abstract

This paper presents a novel architecture for parallel database processing called Multi-Wavelength Optical Content Addressable Parallel Processor (MW-OCAPP). MW-OCAPP is designed to provide efficient parallel retrieval and processing of data by moving the bulk of database operations from electronics to optics. It combines a parallel model of computation with the many degrees of processing freedom that light provides. MW-OCAPP uses a polarization and wavelength-encoding scheme to achieve a high level of parallelism. Distinctive features of the proposed architecture include (1) the use of a multiple-wavelength encoding scheme to enhance processing parallelism, (2) multiple-comparand word-parallel and bit-parallel magnitude comparison with an execution-time independent of the data size or word size, (3) the implementation of a suite of eleven database primitives, and (4) multi-comparand two-dimensional data processing. The MW-OCAPP architecture realizes eleven relational database primitives: difference, intersection, union, conditional selection, maximum, minimum, join, product, projection, division and update. Most of these operations execute in constant time independent of the data size. This paper outlines the architectural concepts and motivation behind MW-OCAPP's design, as well as describes the architecture required for implementing the equality and magnitude comparison processing cores. Additionally, a physical demonstration of the multiwavelength equality operation is presented.