Identifying the Behavioral Difference using Differential Slicing

Abstract

The programmer has to understand the behavior of two similar programs and then identify the execution difference which produces difference in output. When two similar programs are executed under two different environments which shows different behavior in output. The main difference exists in the program behavior is due to two different types of input. This paper proposes differential slicing based on trace alignment algorithm which produces the execution differences and generates a casual difference graph. We implement differential slicing for C# programs and identify the execution difference. The results shows that differential slicing identifies the input difference and casual difference graph reduces the amount of time for the programmers to understand the execution difference. Our experimental results show the proposed differential slicing performs better than existing approach. KeywordsCasual Difference Graph (CDG), Program Dependence Graph (PDG) 1.Introduction:Differential program Analysis is the process of finding two similar programs to determine the behavioral difference between them.Programs are frequently maintained by the programmers who are far removed from software development process. The purpose of modifications in the program development is not always clear. The actual effect on the programs behavior is the presence of a particular part of the program may be unknown,a particular line may be crucial or it may have no effect at all. The previous research shows that difference in behavior is caused by the presence or absence of particular element would help the maintainers in understanding the program. A set of automated techniques is needed to analyze the effect of modifications. We use differential slicing to focus on differences between two similar programs [1]. We proposed a differential slicing for C# programs. It identifies the aligned and disaligned region and generates a casual difference graph. We compare the differential slicing and existing approach and the results are tabulated. Our results also show that differential slicing identifies the input difference and CDG decreases the time and effort needed for an analyst to understand the observed difference. The paper is organized as follows. Section 2 describes the existing differential slicing techniques. Section 3 explains about Dynamic slicing. Section 4 explains about differential slicing. Section 5 describes about slice alignment. Section 6 describes the performance result of differential slicing. Section 7 describes conclusion. Related WorkDifferential program analysis is the task of analyzing two related programs which identified the behavioral difference between them. Joel Winsted et al presents a technique to find an input for which the two programs will produce different outputs ,Thus illustrating the behavioral difference between the two programs [1]. A combination of static and dynamic techniques are used to find the differential inputs. Sumner and Zhang [4] finds the casual path of two executions by first patching the...

Adsorptive Removal of Lead and Cadmium Ions using Cross-Linked CMC Schiff Base: Isotherm, Kinetics and Catalytic Activity

“…But in some cases it also improves the adsorption capacity, depending on the functional groups in the crosslinking agent 7 . The chitosan and chitosan crosslinked with gluteraldehyde has reported in the previous literatures [8][9] and their application in the removal of heavy metals 10 and biological property have been discussed. The metal selected for our study is lead and cadmium.…”