Vulnerability discovery in multiple version software systems : open source and commercial software systems
Kim, Jin Yoo
MetadataShow full item record
The vulnerability discovery process for a program describes the rate at which the vulnerabilities are discovered. A model of the discovery process can be used to estimate the number of vulnerabilities likely to be discovered in the near future. Past studies have considered vulnerability discovery only for individual software versions, without considering the impact of shared code among successive versions and the evolution of source code. These affecting factors in vulnerability discovery process need to be taken into account estimate the future software vulnerability discovery trend more accurately. This thesis examines possible approaches for taking these factors into account in the previous works. We implemented these factors on vulnerability discovery process. We examine a new approach for quantitatively vulnerability discovery process, based on shared source code measurements among multiple version software system. The applicability of the approach is examined using Apache HTTP Web server and Mysql DataBase Management System (DBMS). The result of this approach shows better goodness of fit than fitting result in the previous researches. Using this revised software vulnerability discovery process, the superposition effect which is an unexpected vulnerability discovery in the previous researches could be determined by software discovery model. The multiple software vulnerability discovery model (MVDM) shows that vulnerability discovery rate is different with single vulnerability discovery model’s (SVDM) discovery rate because of newly considered factors. From these result, we create and applied new SVDM for open source and commercial software. This single vulnerability process is examined, and the model testing result shows that SVDM can be an alternative modeling. The modified vulnerability discovery model will be presented for supporting previous researches’ weakness, and the theoretical modeling will be discuss for more accurate explanation.