论文信息 - The Power of a One-Dimensional Vector of Processors

The Power of a One-Dimensional Vector of Processors

Kung [1979b] has recently enunciated a set of principles for designing algorithms for implementation in Very Large Scale Integrated circuitry (VLSI), and supported these principles by displaying a number of particular algorithms based on various "communication geometries". In this paper we will examine a communication geometry which Kung calls the "one-dimensional array of processors", and which we call a "processor vector" or "PV". We will see that this simple structure can efficiently solve the rather difficult problems of multiplying matrices and of constructing minimum spanning trees.

Thomas Ottmann | Jon Louis Bentley

[1] C. Thomborson,et al. Area-time complexity for VLSI , 1979, STOC.

[2] Franco P. Preparata,et al. The cube-connected-cycles: A versatile network for parallel computation , 1979, 20th Annual Symposium on Foundations of Computer Science (sfcs 1979).

[3] Lynn Conway,et al. Introduction to VLSI systems , 1978 .

[4] C. Savage. Parallel Algorithms for Graph Theoretic Problems , 1977 .

[5] H. T. Kung. Let's Design Algorithms for VLSI Systems , 1979 .

[6] Robert E. Tarjan,et al. Finding Minimum Spanning Trees , 1976, SIAM J. Comput..

[7] Jon Louis Bentley,et al. A Parallel Algorithm for Constructing Minimum Spanning Trees , 1980, J. Algorithms.

[8] H. T. Kung,et al. Systolic Arrays for (VLSI). , 1978 .

[9] Seymour E. Goodman,et al. Introduction to the Design and Analysis of Algorithms , 1977 .

[10] Andrew Chi-Chih Yao,et al. On Constructing Minimum Spanning Trees in k-Dimensional Spaces and Related Problems , 1977, SIAM J. Comput..

[11] H. T. Kung. The Structure of Parallel Algorithms , 1980, Adv. Comput..