Automatic Location of Root Causes of Performance Regression in Web-based Systems, Joint Venture Concordia University, ERA Environmental Software Solutions, & NSERC
A joint study between Concordia University and ERA Environmental Software solutions in collaboration with Natural Science and Engineering Research Council of Canada (NSERC) enabled the automated locating of root causes of performance regressions in web-based systems (Liao et al., 2021). This study was published in IEEE Transactions on Software Engineering on November 30th, 2021.
MONTREAL, Dec. 17, 2021 /PRNewswire-PRWeb/ -- A joint study between Concordia University and ERA Environmental Software solutions in collaboration with Natural Science and Engineering Research Council of Canada (NSERC) enabled the automated locating of root causes of performance regressions in web-based systems (Liao et al., 2021). This study was published in IEEE Transactions on Software Engineering on November 30th, 2021.
Large-scale software providers invest large sums of money into performance evaluation for their systems. Individual testing is time-consuming, and there is a need for a testing method that could run in the background while the software is running. The study designed approaches to automatically locate the root causes of performance regressions in programs without time-consuming performance testing. The developed procedure uses black box performance models to show the relationship between how a system performs and its run time activity. Statistical analysis was used to investigate the root causes of performance regression further. The open-source subject systems used were TeaStore, OpenMRS, and Cloud store.
The approach was tested by its ability to accurately detect and locate injected performance regressions on three open-source projects and an industrial product. Injected performance regressions included additional calculations, excessive garbage collection, excessive I/O access, and excessive use of multi-threading. The method successfully detected the causes of three injected performance regressions and was applied to a large-scale industrial system, ERA environmental' s EHS software, (on daily basis over a 12-month period). The next steps in the study will be to use these approaches in updated versions of the program.
This study is beneficial in saving time spent on performance testing and mitigating financial losses associated with performance issues for large-scale software companies. It is currently open for companies that would like to employ it in their operations.
The research resulted from a joint research effort between the Department of Computer Science and Software Engineering at Concordia University and the research and development department at ERA Environmental Software solutions. ERA is a provider of sustainable software solutions in fields of environmental compliance, sustainability, health safety.
(1) Liao, L., Chen, J., Li, H., Zeng, Y., Shang, W., Sporea, C., ... & Sajedi, S. (2021). Locating Performance Regression Root Causes in the Field Operations of Web-based Systems: An Experience Report. IEEE Transactions on Software Engineering.
Media Contact
Sarah Sajedi, ERA Environmental Management Solutions, 1 438-799-6208, [email protected]
Weiyi(Ian) Shang, Concordia University, (514) 848-2424 7801, [email protected]
SOURCE ERA Environmental Management Solutions

Share this article