Matrix algorithms on data-flow computers

Matej Žniderič (2017) Matrix algorithms on data-flow computers. MSc thesis.

Preview

Abstract

While the processor frequency has failed to significantly increase in the last decade, the needs of science for computing power continue to increase. Data-flow computer architecture presents a good alternative to traditional control-flow computer. In this work a set of algorithms for matrix-vector multiplication and matrix-matrix multiplication was developed. It was compared with control-flow implementations. With matrix-vector multiplication there was no speedup. Algorithm for multiplying matrix with a set of vectors achieved almost 4-fold speedup. Matrix-matrix multiplication algorithm achieved more than 100-fold speedup. Further matrix exponentiation and all-pairs shortest path search algorithm were implemented. First algorithm achieved 100-times speedup, while the second one was about the same speed as control-flow Floyd-Warshall algorithm.

Item Type:

Thesis (MSc thesis)

Keywords:

dataflow architecture, matrix multiplication, matrix algorithms, graph algorithms, Maxeler Technologies

Number of Pages:

Language of Content:

Slovenian

Mentor / Comentors:

Name and Surname	ID	Function
doc. dr. Jurij Mihelič	305	UNSPECIFIED

Link to COBISS:

http://www.cobiss.si/scripts/cobiss?command=search&base=51012&select=(ID=1537345731)

Institution:

University of Ljubljana

Department:

Faculty of Computer and Information Science

Item ID:

3754

Date Deposited:

18 Jan 2017 15:03

Last Modified:

03 Feb 2017 11:30

URI:

http://eprints.fri.uni-lj.si/id/eprint/3754

Actions (login required)

View Item