Parallel computing algorithm and architecture design with case study on parallel matrix multiplication and pipelined data transmission
Date of Publication
1990
Document Type
Master's Thesis
Degree Name
Master of Science in Computer Science
College
College of Computer Studies
Department/Unit
Computer Science
Thesis Adviser
Nopporn Luangprasert
Defense Panel Chair
Alphonse Chung
Defense Panel Member
Patricia Maria Claudio
Reynaldo Villafuerte
Abstract/Summary
A tool activity diagram is presented. The tool facilitates parallel program development by providing pre-coding visualization of the program. It is also useful in depicting and evaluating parallel algorithms.
A theory, pipelined data transmission, categorizes modes of bulk data transmission to and from large numbers of processing elements in parallel computer systems. New terms for the concepts were coined. The theory is developed to an extent beyond the immediate need of the thesis. The theory can readily be used by other works. The activity diagram found extensive use in illustrating the concepts.
Several algorithms for parallel matrix multiplication were developed and analyzed. The Amdahl's law was tested to ascertain validity.
A new parallel computer architecture was developed that resulted from the studies made on the algorithms.
Abstract Format
html
Language
English
Format
Accession Number
TG01796
Shelf Location
Archives, The Learning Commons, 12F Henry Sy Sr. Hall
Physical Description
iii, 181 p., 28 cm.
Keywords
Parallel programs (Computer programs); Computer algorithms
Recommended Citation
Sheng, P. (1990). Parallel computing algorithm and architecture design with case study on parallel matrix multiplication and pipelined data transmission. Retrieved from https://animorepository.dlsu.edu.ph/etd_masteral/1285