N-Version Design vs. One Good Version

Evidence indicates that n-version development techniques are more reliable than producing one "good" version-and cost effective in the long run. The author concludes that diverse, independent channels used in parallel are significantly superior to even the current state of the art, especially in situations where cost of failure is high.

[1]  Shari Lawrence Pfleeger,et al.  Investigating the Influence of Formal Methods , 1997, Computer.

[2]  John D. Musa,et al.  Software reliability - measurement, prediction, application , 1987, McGraw-Hill series in software engineering and technology.

[3]  John D. Musa,et al.  Software reliability measurement , 1984, J. Syst. Softw..

[4]  Nancy G. Leveson,et al.  Analysis of Faults in an N-Version Software Experiment , 1990, IEEE Trans. Software Eng..

[5]  Peter T. Popov,et al.  The effect of testing on the reliability of single version and 1-out-of-2 software systems , 1995, Proceedings of Sixth International Symposium on Software Reliability Engineering. ISSRE'95.

[6]  Andy Roberts,et al.  How Accurate Is Scientific Software? , 1994, IEEE Trans. Software Eng..

[7]  Les Hatton,et al.  Reexamining the Fault Density-Component Size Connection , 1997, IEEE Softw..

[8]  Nancy G. Leveson,et al.  An experimental evaluation of the assumption of independence in multiversion programming , 1986, IEEE Transactions on Software Engineering.

[9]  Bev Littlewood,et al.  The Need for Evidence from Disparate Sources to Evaluate Software Safety , 1993 .

[10]  David F. McAllister,et al.  An Experimental Evaluation of Software Redundancy as a Strategy For Improving Reliability , 1991, IEEE Trans. Software Eng..

[11]  Peter G. Neumann,et al.  Computer-related risks , 1994 .

[12]  Leslie Hatton,et al.  Software failures-follies and fallacies , 1997 .

[13]  Peter G. Bishop,et al.  PODS — A project on diverse software , 1986, IEEE Transactions on Software Engineering.