Parallelization of Software Systems Test Case Selection Algorithm Based on Singular Value Decomposition

TL;DR

This paper presents a parallelized approach using singular value decomposition to efficiently select test cases in software regression testing, aiming to improve speed and scalability for large systems.

Contribution

It introduces a novel parallel algorithm leveraging SVD for clustering change impacts, enhancing test case selection efficiency in software testing.

Findings

Achieved significant speedup with parallel processing

Effective clustering of system changes based on SVD

Scalable approach for large software systems

Abstract

When developing a software system, a change in one part of the system may lead to unwanted changes in other parts of the system. These affected parts may interfere with system performance, so regression testing is used to deal with these disorders. This test seeks to re-measure these sections to prevent these abnormalities, but it is difficult to identify these sections for re-examination. We try to cluster the changes of our software system based on the system functions by singular value decomposition, to be able to use to identify these parts during a new change, to perform the test again. In order to increase speedup, our calculations were performed in parallel on shared memory systems so that by increasing the scale of software systems, an optimal answer could be obtained.