Identify And Analysis System's Rudimental Bug Based On Data

With the deepening of informationization, more and more software systems have been applied to a variety of environments. However, the system may also be delivered with some residual defects undetected. For this shortcoming, traditional methods are generally set software measurement information. Using experience and known defects compared to identify defects. However, in many cases may be invalid. During the post-graduate studies, the author takes part in a real running system maintenance and secondary development work at telecom operators. By studying the actual running system, discovery system run-time data have a clear distinction with the test data, these data from the running system often contains a lot of information. If use these information initiatively to identify and analyze system's rudimental defects, effects could be more obvious.For the above purpose, the following work have been done in this paper:Firstly, residual defects in the identification process. Research describing the business, data and rules model of software system. Then put forward the identification process based on these three models. Through the steps of access to information to build the three models, data acquisition, statistics and authentication, analysis of the results to identify the system's rudimental defects.Secondly, analysis of rudimental defects. Analyse the reason of rudimental defects which was found by the method proposed in this paper. Through the relationship, data items, data meaning to analyse and then look for the source of defects. To gather statistics of the impact of the defect, and then use fractal estimation wider impact of defects.Thirdly, Study the composition of monitoring model, program and module design. Design the rudimental defects monitoring program with the process of first grant price.