Parallelisation of algorithms

dc.contributor.advisorGledhill, I M Aen_ZA
dc.contributor.advisorKritzinger, Pieter Sen_ZA
dc.contributor.authorSchuilenburg, Alexander Mariusen_ZA
dc.date.accessioned2016-10-19T13:39:15Z
dc.date.available2016-10-19T13:39:15Z
dc.date.issued1990en_ZA
dc.description.abstractMost numerical software involves performing an extremely large volume of algebraic computations. This is both costly and time consuming in respect of computer resources and, for large problems, often super-computer power is required in order for results to be obtained in a reasonable amount of time. One method whereby both the cost and time can be reduced is to use the principle "Many hands make light work", or rather, allow several computers to operate simultaneously on the code, working towards a common goal, and hopefully obtaining the required results in a fraction of the time and cost normally used. This can be achieved through the modification of the costly, time consuming code, breaking it up into separate individual code segments which may be executed concurrently on different processors. This is termed parallelisation of code. This document describes communication between sequential processes, protocols, message routing and parallelisation of algorithms. In particular, it deals with these aspects with reference to the Transputer as developed by INMOS and includes two parallelisation examples, namely parallelisation of code to study airflow and of code to determine far field patterns of antennas. This document also reports on the practical experiences with programming in parallel.en_ZA
dc.identifier.apacitationSchuilenburg, A. M. (1990). <i>Parallelisation of algorithms</i>. (Thesis). University of Cape Town ,Faculty of Science ,Department of Computer Science. Retrieved from http://hdl.handle.net/11427/22211en_ZA
dc.identifier.chicagocitationSchuilenburg, Alexander Marius. <i>"Parallelisation of algorithms."</i> Thesis., University of Cape Town ,Faculty of Science ,Department of Computer Science, 1990. http://hdl.handle.net/11427/22211en_ZA
dc.identifier.citationSchuilenburg, A. 1990. Parallelisation of algorithms. University of Cape Town.en_ZA
dc.identifier.ris TY - Thesis / Dissertation AU - Schuilenburg, Alexander Marius AB - Most numerical software involves performing an extremely large volume of algebraic computations. This is both costly and time consuming in respect of computer resources and, for large problems, often super-computer power is required in order for results to be obtained in a reasonable amount of time. One method whereby both the cost and time can be reduced is to use the principle "Many hands make light work", or rather, allow several computers to operate simultaneously on the code, working towards a common goal, and hopefully obtaining the required results in a fraction of the time and cost normally used. This can be achieved through the modification of the costly, time consuming code, breaking it up into separate individual code segments which may be executed concurrently on different processors. This is termed parallelisation of code. This document describes communication between sequential processes, protocols, message routing and parallelisation of algorithms. In particular, it deals with these aspects with reference to the Transputer as developed by INMOS and includes two parallelisation examples, namely parallelisation of code to study airflow and of code to determine far field patterns of antennas. This document also reports on the practical experiences with programming in parallel. DA - 1990 DB - OpenUCT DP - University of Cape Town LK - https://open.uct.ac.za PB - University of Cape Town PY - 1990 T1 - Parallelisation of algorithms TI - Parallelisation of algorithms UR - http://hdl.handle.net/11427/22211 ER - en_ZA
dc.identifier.urihttp://hdl.handle.net/11427/22211
dc.identifier.vancouvercitationSchuilenburg AM. Parallelisation of algorithms. [Thesis]. University of Cape Town ,Faculty of Science ,Department of Computer Science, 1990 [cited yyyy month dd]. Available from: http://hdl.handle.net/11427/22211en_ZA
dc.language.isoengen_ZA
dc.publisher.departmentDepartment of Computer Scienceen_ZA
dc.publisher.facultyFaculty of Scienceen_ZA
dc.publisher.institutionUniversity of Cape Town
dc.subject.otherComputer Scienceen_ZA
dc.titleParallelisation of algorithmsen_ZA
dc.typeMaster Thesis
dc.type.qualificationlevelMasters
dc.type.qualificationnameMScen_ZA
uct.type.filetypeText
uct.type.filetypeImage
uct.type.publicationResearchen_ZA
uct.type.resourceThesisen_ZA
Files
Original bundle
Now showing 1 - 2 of 2
Loading...
Thumbnail Image
Name:
thesis_sci_1990_schuilenburg_alexander_marius_1.pdf
Size:
6.69 MB
Format:
Adobe Portable Document Format
Description:
Loading...
Thumbnail Image
Name:
thesis_sci_1990_schuilenburg_alexander_marius_2.pdf
Size:
5.98 MB
Format:
Adobe Portable Document Format
Description:
Collections