Skip to Main Content
 

Global Search Box

 
 
 
 

ETD Abstract Container

Abstract Header

A Cost Effective Methodology for Quantitative Evaluation of Software Reliability using Static Analysis

Schilling, Walter William, Jr.

Abstract Details

2007, Doctor of Philosophy in Engineering, University of Toledo, Electrical Engineering.
Software reliability represents an increasing risk to overall system reliability. As systems have become larger and more complex, mission critical and safety critical systems have increasingly had functionality controlled exclusively through software. This change has resulted in a shift of the root cause of systems failure from hardware to software. Market forces have encouraged projects to reuse existing software as well as purchase COTS solutions. This has made the usage of existing reliability models difficult. Traditional software reliability models require significant testing data to be collected during development. If this data is not collected in a disciplined manner or is not made available to software engineers, these modeling techniques can not be applied. It is imperative that practical reliability modeling techniques be developed to address these issues. This dissertation puts forth a practical method for estimating software reliability. The proposed software reliability model combines static analysis of existing source code modules, functional testing with execution path capture, and a series of Bayesian Belief Networks. Static analysis is used to detect faults within the source code which may lead to failure. Code coverage is used to determine which paths within the source code are executed as well as the execution rate. The Bayesian Belief Networks combine these parameters and estimate the reliability for each method. A second series of Bayesian Belief Networks then combines the data for each method to determine the overall reliability for the system. In order to use this model, the SOSART tool is developed. This tool serves as a reliability modeling tool and a bug finding meta tool suitable for comparing the results of different static analysis tools. Verification of the model is presented through multiple experimental instances. Validation is first demonstrated through the application to a series of Open Source software packages. A second validation is provided using the Tempest Web Server, developed by NASA Glenn Research Center.
Mansoor Alam (Advisor)

Recommended Citations

Citations

  • Schilling, Jr., W. W. (2007). A Cost Effective Methodology for Quantitative Evaluation of Software Reliability using Static Analysis [Doctoral dissertation, University of Toledo]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=toledo1189820658

    APA Style (7th edition)

  • Schilling, Jr., Walter. A Cost Effective Methodology for Quantitative Evaluation of Software Reliability using Static Analysis. 2007. University of Toledo, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=toledo1189820658.

    MLA Style (8th edition)

  • Schilling, Jr., Walter. "A Cost Effective Methodology for Quantitative Evaluation of Software Reliability using Static Analysis." Doctoral dissertation, University of Toledo, 2007. http://rave.ohiolink.edu/etdc/view?acc_num=toledo1189820658

    Chicago Manual of Style (17th edition)