LightSys: Lightweight and Efficient CI System for Improving Integration Speed of SoftwareSEIP
Tue 25 May 2021 22:30 - 22:50 at Blended Sessions Room 1 - 1.1.1. Code Review: Automation
The complexity and size increase of software has extended the delay for developers as they wait for code analysis and code merge. With the larger and more complex software, more developers nowadays are developing software with large source code repositories. The tendency for software platforms to immediately update software packages with feature updates and bug-fixes is a significant obstacle. Continuous integration systems may help prevent software flaws during the active development of software packages, even when they are deployed and updated frequently. Herein, we present a portable and modular code review automation system that inspects incoming code changes such as code format and style, performance regressions, static analysis, build and deployment tests, and dynamic analysis before merging and changing code. The proposed mechanisms are sufficiently lightweight to be hosted on a regular desktop computer even for numerous developers. The resulting reduced costs allow developers to apply the proposed mechanism to many source code repositories. Experimental results demonstrate that the proposed mechanism drastically reduces overheads and improves usability compared with conventional mechanisms: execution time (6x faster), CPU usage (40% lower), memory consumption (1/180), and no out-of-memory occurrences.
Tue 25 MayDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
22:30 - 23:30 | 1.1.1. Code Review: AutomationSEIP - Software Engineering in Practice / Technical Track at Blended Sessions Room 1 | ||
22:30 20mPaper | LightSys: Lightweight and Efficient CI System for Improving Integration Speed of SoftwareSEIP SEIP - Software Engineering in Practice Geunsik Lim Samsung Research, Samsung Electronics, MyungJoo Ham Samsung Electronics, Jijoong Moon Samsung Electronics, Wook Song Samsung Electronics Link to publication DOI Pre-print Media Attached | ||
22:50 20mPaper | Using Machine Intelligence to Prioritise Code Review RequestsSEIP SEIP - Software Engineering in Practice Pre-print Media Attached | ||
23:10 20mPaper | Towards Automating Code Review ActivitiesTechnical Track Technical Track Rosalia Tufano Università della Svizzera Italiana, Luca Pascarella Delft University of Technology, Michele Tufano Microsoft, Denys Poshyvanyk College of William & Mary, Gabriele Bavota Software Institute, USI Università della Svizzera italiana Pre-print Media Attached |