Computer Science, № 4, 1996

Associative neural memory of the Hopfield type, where each element (neuron) has a restricted amount of connections is considered as a cellular-automaton-like network. The goal of the investigation is to determine the degradation of memory capacity and efficiency of pattern retrieval caused by the decrease of connection number. As a...

Solving of large-scale systems of linear algebraic equations, ill-conditioned and non-square systems requires special algorithms which should be suitable for effective implementation in modern multi-processor systems. One of such methods was proposed by A.A. Abramov. This approach is based on consequential projecting of solution to a system of ortogonal vectors...

Two new cellular pipelined algorithms (2D and 3D) for computing a sum of products with a very short multiplication time (six and three steps, respectively) are proposed. The initial data and results are binary signed-digit integers. The design and investigation tool is an experemental computer simulating system (Animating Language Tools)...

The history of the Siberian researches in parallel processing is surveyed and present project of the Siberian Network Supercomputing Center (SNSC) is considered.

In this paper by means of an abstract model (the STAR-machine) we describe an associative version of the Gabow algorithm for the degree-restricted problem. We have obtained that on an associative parallel processor this algorithm takes the same time as a minimal spanning tree algorithm. In Appendix the associative algorithm...

Finite element non-stationary electromagnetic fields modeling technique is proposed. It permits to decrease computing expenses of three-dimensional problem solving. This technique is implemented in program complex TELMA, its efficiency is confirmed by solving significant number of both modeling tasks, and complicated practical problems.