Measuring Software Reliability under the Influence of an Infected Patch

Abstract

Patching service provides software firms an option to deal with the leftover bugs and is thereby helping firms to keep track of their product. More and more software firms are making use of this concept of prolonging testing. But this framework of releasing unprepared software in the market involves a huge risk. The hastiness of vendors in releasing of software patch at times can be dangerous as there are chances of firms releasing an infected patch. The infected patch (es) might lead to a hike in bug and error count of the software and might make the software more vulnerable. The current proposal presents an understanding of such a situation through a mathematical model. Wherein, the role of differing behavior of testers (during in-house testing and eld testing) and users is described. The proposed model has been validated on two software failure data sets of Tandem Computers and Brazilian Electronic Switching System, TROPICO R-1500.