Context-sensitive delta inference for identifying workload-dependent performance bottlenecks

Xusheng Xiao, Shi Han, Dongmei Zhang, Tao Xie

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

Abstract

Software hangs can be caused by expensive operations in responsive actions (such as time-consuming operations in UI threads). Some of the expensive operations depend on the input workloads, referred to as workload-dependent performance bottlenecks (WDPBs). WDPBs are usually caused by workload-dependent loops (i.e., WDPB loops) that contain relatively expensive operations. Traditional performance testing and single-execution profiling may not reveal WDPBs due to incorrect assumptions of workloads. To address these issues, we propose the ΔInfer approach that predicts WDPB loops under large workloads via inferring iteration counts of WDPB loops using complexity models for the workload size. ΔInfer incorporates a novel concept named context-sensitive delta inference that consists of two parts: temporal inference for inferring the complexity models of different program locations, and spatial inference for identifying WDPB loops as WDPB candidates. We conducted evaluations on two popular open-source GUI applications, and identified impactful WDPBs that caused 10 performance bugs.

Original languageEnglish (US)
Title of host publication2013 International Symposium on Software Testing and Analysis, ISSTA 2013 - Proceedings
Pages90-100
Number of pages11
DOIs
StatePublished - 2013
Externally publishedYes
Event22nd International Symposium on Software Testing and Analysis, ISSTA 2013 - Lugano, Switzerland
Duration: Jul 15 2013Jul 20 2013

Publication series

Name2013 International Symposium on Software Testing and Analysis, ISSTA 2013 - Proceedings

Other

Other22nd International Symposium on Software Testing and Analysis, ISSTA 2013
CountrySwitzerland
CityLugano
Period7/15/137/20/13

Keywords

  • Performance analysis
  • model prediction

ASJC Scopus subject areas

  • Software

Fingerprint Dive into the research topics of 'Context-sensitive delta inference for identifying workload-dependent performance bottlenecks'. Together they form a unique fingerprint.

Cite this