A cost/benefit approach to performance analysis

David Maplesden, Ewan Tempero, John Hosking, John C. Grundy

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

Abstract

Most performance engineering approaches focus on understanding the use of runtime resources. However such approaches do not quantify the value being provided in return for the consumption of these resources. Without such a measure it is not possible to compare the efficiency of these components (that is whether the runtime cost is reasonable given the benefit being provided). We have created an empirical approach that measures the value being provided by a code path in terms of the visible data it generates for the rest of the application. Combining this with traditional performance cost data, creates an efficiency measure for every code path in the application. We have evaluated our approach using the DaCapo benchmark suite, demonstrating our analysis allows us to quantify the efficiency of the code in each benchmark and find real optimisation opportunities, providing improvements of up to 36% in our case studies.

Original languageEnglish
Title of host publicationProceedings of the 2016 ACM/SPEC International Conference on Performance Engineering
EditorsSteffen Becker, Xiaoyun Zhu, Jerry Rolia, Manoj Nambiar
Place of PublicationNew York NY USA
PublisherAssociation for Computing Machinery (ACM)
Pages15-26
Number of pages12
ISBN (Electronic)9781450340809
DOIs
Publication statusPublished - 2016
Externally publishedYes
EventACM/SPEC International Conference on Performance Engineering 2016 - Delft, Netherlands
Duration: 12 Mar 201616 Mar 2016
Conference number: 7th
https://icpe2016.spec.org/

Conference

ConferenceACM/SPEC International Conference on Performance Engineering 2016
Abbreviated titleICPE 2016
CountryNetherlands
CityDelft
Period12/03/1616/03/16
Internet address

Keywords

  • Blended analysis
  • Eciency analysis
  • Profiling
  • Runtime bloat

Cite this