Biorthogonally accessed three-dimensional two-photon memory for relational database operations.

Memory bandwidth is a bottleneck for very large database machines. Parallel-access three-dimensional two-photon memories have the potential of achieving enormous throughput (>100 Gbit/s) and capacity (1 Tbit/cm(3)) [Appl. Opt. 29, 2058 (1990)] and, consequently, are well suited for this application. Our analysis shows that some operations can be completed more than 2 orders of magnitude faster with this type of memory than with a system based on serial-access storage. These particular memories have a further feature of being accessible in orthogonal directions. We show that this property, used in conjunction with a three-dimensional data-organization scheme designed for this approach, leads to improved performance by permitting the user a choice of accessing strategies for a given operation.

Parallel algorithms based on expander graphs for optical computing.

We consider the task of interconnecting processors to realize efficient parallel algorithms. We propose interconnecting processors using certain graphs called expander graphs, which can provide fast communication from any group of processors to the rest of the network. We show that these interconnections would result in a number of efficient parallel algorithms for sorting, routing, associative memory, and fault-tolerance networks. As the interconnections based on expander graphs are global and irregular, we reason that optical interconnections are preferred to electronic and propose implementation of these interconnections using the programmable optoelectronic multiprocessor architecture.