An Empirical Study on the Energy Efficiency of Matrix Transposition Algorithms

Gonçalo Lopes, João Paulo Fernandes, Luís Paquete

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Energy consumption is becoming a serious concern in the context of software development. Recent works have shown that energy consumption of an algorithm not only depends on its running-time but also on its number of memory accesses. In this work, we empirically analyse several algorithms for matrix transposition with different patterns of low-level cache access, and compare them in terms of energy consumption and running-time with respect to CPU instructions and memory accesses. Our results suggest that different memory access patterns have a strong influence on the energy consumption and on the cache performance of these algorithms.

Original languageEnglish (US)
Title of host publicationComposability, Comprehensibility and Correctness of Working Software - 8th Summer School, CEFP 2019, Revised Selected Papers
EditorsZoltán Porkoláb, Viktória Zsók
PublisherSpringer Science and Business Media Deutschland GmbH
Pages375-391
Number of pages17
ISBN (Print)9783031428326
DOIs
StatePublished - 2023
Event8th Central European Functional Programming School, CEFP 2019 - Budapest, Hungary
Duration: Jun 17 2019Jun 21 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11950 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference8th Central European Functional Programming School, CEFP 2019
Country/TerritoryHungary
CityBudapest
Period6/17/196/21/19

Keywords

  • Cache performance
  • Energy efficiency
  • Matrix transposition

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'An Empirical Study on the Energy Efficiency of Matrix Transposition Algorithms'. Together they form a unique fingerprint.

Cite this