Techniques to improve the performance of mutation analysis.

Title: Techniques to improve the performance of mutation analysis.
Identifier: AAI9510660
identifier: 9510660
Creator: Fleyshgakker, Vladimir Nathan.
Contributor: Adviser: Stewart N. Weiss
Date: 1994
Language: English
Publisher: City University of New York.
Subject: Computer Science
Abstract: Although mutation analysis is a potentially effective method of software test data evaluation, it is computationally intensive and time-consuming. Existing algorithms are not very efficient, failing to eliminate redundant computations. We present a universal mutation analysis data structure and new serial algorithms for both strong and weak mutation analysis that on average perform much faster than existing ones, and can never do worse. We describe these algorithms and analyze their run time complexities as well as the complexities of existing algorithms.;We also present a novel, special purpose, dynamically reconfigurable multiple-SIMD architecture designed for mutation analysis. We describe the architecture and its use for mutation analysis. Our performance analysis indicates that the utilization factor of the processing elements and the potential speed-up are impressive. The results of our software simulation corroborate the theoretical analysis. On our architecture mutation analysis is a feasible method of evaluating software test data.
Type: dissertation
Source: PQT Legacy CUNY.xlsx
degree: Ph.D.