Engineering topics

Vouk, Mladen A.

Publications and source records attributed to Vouk, Mladen A..

18 records

NASA NTRS1999

Practical Issues in Implementing Software Reliability Measurement

Many ways of estimating software systems' reliability, or reliability-related quantities, have been developed over the past several years. Of particular interest are methods that can be used to estimate a software system's fault content prior to test, or to discriminate between components that are fault-prone and those that are not. The results of these methods can be used to: 1) More accurately focus scarce fault identification resources on those portions of a software system most in need of it. 2) Estimate and forecast the risk of exposure to residual faults in a software system during operation, and develop risk and safety criteria to guide the release of a software system to fielded use. 3) Estimate the efficiency of test suites in detecting residual faults. 4) Estimate the stability of the software maintenance process.

Nikora, Allen P.↗

NASA NTRS1993

Software reliability through fault-avoidance and fault-tolerance

Strategies and tools for the testing, risk assessment and risk control of dependable software-based systems were developed. Part of this project consists of studies to enable the transfer of technology to industry, for example the risk management techniques for safety-concious systems. Theoretical investigations of Boolean and Relational Operator (BRO) testing strategy were conducted for condition-based testing. The Basic Graph Generation and Analysis tool (BGG) was extended to fully incorporate several variants of the BRO metric. Single- and multi-phase risk, coverage and time-based models are being developed to provide additional theoretical and empirical basis for estimation of the reliability and availability of large, highly dependable software. A model for software process and risk management was developed. The use of cause-effect graphing for software specification and validation was investigated. Lastly, advanced software fault-tolerance models were studied to provide alternatives and improvements in situations where simple software fault-tolerance strategies break down.

Vouk, Mladen A.↗

NASA NTRS1992

Software reliability through fault-avoidance and fault-tolerance

Accomplishments in the following research areas are summarized: structure based testing, reliability growth, and design testability with risk evaluation; reliability growth models and software risk management; and evaluation of consensus voting, consensus recovery block, and acceptance voting. Four papers generated during the reporting period are included as appendices.

Vouk, Mladen A.↗

NASA NTRS1991

An experimental evaluation of software redundancy as a strategy for improving reliability

The strategy of using multiple versions of independently developed software as a means to tolerate residual software design faults is suggested by the success of hardware redundancy for tolerating hardware failires. Although, as generally accepted, the independence of hardware failures resulting from physical wearout can lead to substantial increases in reliability for redundant hardware structures, a similar conclusion is not immediate for software. The degree to which design faults are manifested as independent failures determines the effectiveness of redundancy as a method for improving software reliability. Interest in multi-version software centers on whether it provides an adequate measure of increased reliability to warrant its use in critical applications. The effectiveness of multi-version software is studied by comparing estimates of the failure probabilities of these systems with the failure probabilities of single versions. The estimates are obtained under a model of dependent failures and compared with the estimates obtained when failures are assumed to be independent. The experimental results are based on twenty versions of an aerospace application developed and certified by sixty programmers from four universities. Descriptions of the application, development and certifications processes, and operational evaluation are given together with an analysis of the twenty versions.

Eckhardt, Dave E.↗

NASA NTRS1991

Software reliability through fault-avoidance and fault-tolerance

Twenty independently developed but functionally equivalent software versions were used to investigate and compare empirically some properties of N-version programming, Recovery Block, and Consensus Recovery Block, using the majority and consensus voting algorithms. This was also compared with another hybrid fault-tolerant scheme called Acceptance Voting, using dynamic versions of consensus and majority voting. Consensus voting provides adaptation of the voting strategy to varying component reliability, failure correlation, and output space characteristics. Since failure correlation among versions effectively reduces the cardinality of the space in which the voter make decisions, consensus voting is usually preferable to simple majority voting in any fault-tolerant system. When versions have considerably different reliabilities, the version with the best reliability will perform better than any of the fault-tolerant techniques.

Vouk, Mladen A.↗

NASA NTRS1990

Reliability of voting in fault-tolerant software systems for small output spaces

Under a voting strategy in a fault-tolerant software system there is a difference between correctness and agreement. An independent N-version programming reliability model is proposed for treating small output spaces which distinguishes between correctness and agreement. System reliability is investigated using analytical relationships and simulation. A consensus majority voting stratey is proposed and its performance is analyzed and compared with other voting strategies. A consensus voting strategy automatically adapts the voting to diffeerent component reliability and output space cardinality characteristics. It is shown that absolute majority voting strategy provides a lower bound on the reliability provided by the consensus majority, and the 2-of-n voting strategy an upper bound. If r is the cardinality of output space it is proved that 1/r is a lower bound on the average reliability of fault-tolerant system components below which the system reliability begins to deteriorate as more versions are added.

Mcallister, David F.↗

NASA NTRS1990

Software reliability through fault-avoidance and fault-tolerance

The use of back-to-back, or comparison, testing for regression test or porting is examined. The efficiency and the cost of the strategy is compared with manual and table-driven single version testing. Some of the key parameters that influence the efficiency and the cost of the approach are the failure identification effort during single version program testing, the extent of implemented changes, the nature of the regression test data (e.g., random), and the nature of the inter-version failure correlation and fault-masking. The advantages and disadvantages of the technique are discussed, together with some suggestions concerning its practical use.

Vouk, Mladen A.↗

NASA NTRS1990

An experimental evaluation of software redundancy as a strategy for improving reliability

The strategy of using multiple versions of independently developed software as a means to tolerate residual software design faults is suggested by the success of hardware redundancy for tolerating hardware failures. Although, as generally accepted, the independence of hardware failures resulting from physical wearout can lead to substantial increases in reliability for redundant hardware structures, a similar conclusion is not immediate for software. The degree to which design faults are manifested as independent failures determines the effectiveness of redundancy as a method for improving software reliability. Interest in multi-version software centers on whether it provides an adequate measure of increased reliability to warrant its use in critical applications. The effectiveness of multi-version software is studied by comparing estimates of the failure probabilities of these systems with the failure probabilities of single versions. The estimates are obtained under a model of dependent failures and compared with estimates obtained when failures are assumed to be independent. The experimental results are based on twenty versions of an aerospace application developed and certified by sixty programmers from four universities. Descriptions of the application, development and certification processes, and operational evaluation are given together with an analysis of the twenty versions.

Eckhardt, Dave E., Jr.↗

NASA NTRS1990

Multiversion software reliability through fault-avoidance and fault-tolerance

In this project we have proposed to investigate a number of experimental and theoretical issues associated with the practical use of multi-version software in providing dependable software through fault-avoidance and fault-elimination, as well as run-time tolerance of software faults. In the period reported here we have working on the following: We have continued collection of data on the relationships between software faults and reliability, and the coverage provided by the testing process as measured by different metrics (including data flow metrics). We continued work on software reliability estimation methods based on non-random sampling, and the relationship between software reliability and code coverage provided through testing. We have continued studying back-to-back testing as an efficient mechanism for removal of uncorrelated faults, and common-cause faults of variable span. We have also been studying back-to-back testing as a tool for improvement of the software change process, including regression testing. We continued investigating existing, and worked on formulation of new fault-tolerance models. In particular, we have partly finished evaluation of Consensus Voting in the presence of correlated failures, and are in the process of finishing evaluation of Consensus Recovery Block (CRB) under failure correlation. We find both approaches far superior to commonly employed fixed agreement number voting (usually majority voting). We have also finished a cost analysis of the CRB approach.

Vouk, Mladen A.↗

NASA NTRS1990

Analysis of faults detected in a large-scale multi-version software development experiment

In a multiversion software experiment, twenty programs were built to the same specification of an inertial navigation problem. The programs were then subjected to a three-phase testing and debugging process: an acceptance test, a certification test, and an operational test. Less than 20 percent of the faults discovered during the certification and operational testing were nonunique, i.e., the same or very similar faults would be found in more than one program. However, some of these common faults spanned as many as half of the versions. Faults discovered during the certification testing were due to specification errors and ambiguities, inadequate programmer background knowledge, insufficient programming experience, incomplete analysis, and insufficient acceptance testing. Faults discovered during the operational testing were of a more subtle nature, and were mostly due to various programmer knowledge defects and incomplete analysis errors. Techniques that might have prevented the observed faults are discussed.

Vouk, Mladen A.↗

NASA NTRS1989

Multi-version software reliability through fault-avoidance and fault-tolerance

A number of experimental and theoretical issues associated with the practical use of multi-version software to provide run-time tolerance to software faults were investigated. A specialized tool was developed and evaluated for measuring testing coverage for a variety of metrics. The tool was used to collect information on the relationships between software faults and coverage provided by the testing process as measured by different metrics (including data flow metrics). Considerable correlation was found between coverage provided by some higher metrics and the elimination of faults in the code. Back-to-back testing was continued as an efficient mechanism for removal of un-correlated faults, and common-cause faults of variable span. Software reliability estimation methods was also continued based on non-random sampling, and the relationship between software reliability and code coverage provided through testing. New fault tolerance models were formulated. Simulation studies of the Acceptance Voting and Multi-stage Voting algorithms were finished and it was found that these two schemes for software fault tolerance are superior in many respects to some commonly used schemes. Particularly encouraging are the safety properties of the Acceptance testing scheme.

Vouk, Mladen A.↗

NASA NTRS1989

Experiments in fault tolerant software reliability

Twenty functionally equivalent programs were built and tested in a multiversion software experiment. Following unit testing, all programs were subjected to an extensive system test. In the process sixty-one distinct faults were identified among the versions. Less than 12 percent of the faults exhibited varying degrees of positive correlation. The common-cause (or similar) faults spanned as many as 14 components. However, a majority of these faults were trivial, and easily detected by proper unit and/or system testing. Only two of the seven similar faults were difficult faults, and both were caused by specification ambiguities. One of these faults exhibited variable identical-and-wrong response span, i.e. response span which varied with the testing conditions and input data. Techniques that could have been used to avoid the faults are discussed. For example, it was determined that back-to-back testing of 2-tuples could have been used to eliminate about 90 percent of the faults. In addition, four of the seven similar faults could have been detected by using back-to-back testing of 5-tuples. It is believed that most, if not all, similar faults could have been avoided had the specifications been written using more formal notation, the unit testing phase was subject to more stringent standards and controls, and better tools for measuring the quality and adequacy of the test data (e.g. coverage) were used.

Mcallister, David F.↗

NASA NTRS1988

A large scale second generation experiment in multi-version software - Description and early results

The Second-Generation Experiment is a large-scale empirical study of the development and operation of multiversion software systems that has engaged researchers at five universities and three research institutes. This paper presents the history and current status of this experiment. A preliminary analysis of the multiple software versions has been performed and is reported here.

Kelly, John P. J.↗

NASA NTRS1987

Experiments in fault tolerant software reliability

The reliability of voting was evaluated in a fault-tolerant software system for small output spaces. The effectiveness of the back-to-back testing process was investigated. Version 3.0 of the RSDIMU-ATS, a semi-automated test bed for certification testing of RSDIMU software, was prepared and distributed. Software reliability estimation methods based on non-random sampling are being studied. The investigation of existing fault-tolerance models was continued and formulation of new models was initiated.

Mcallister, David F.↗

NASA NTRS1987

Reliability of voting in fault-tolerant software systems for small output spaces

Under a voting strategy in a fault-tolerant software system there is a difference between correctness and agreement. An independent N-version programming reliability model is proposed for treating small output spaces which distinguishes between correctness and agreement. System reliability is investigated using analytical relationships and simulation. A consensus majority voting strategy is proposed and its performance is analyzed and compared with other voting strategies. Consensus majority strategy automatically adapts the voting to different component reliability and output space cardinality characteristics. It is shown that absolute majority voting strategy provides a lower bound on the reliability provided by the consensus majority, and 2-of-n voting strategy an upper bound. If r is the cardinality of the output space it is proved the 1/r is a lower bound on the average reliability of fault-tolerant system components below which the system reliability begins to deteriorate as more versions are added.

Mcallister, David F.↗

NASA NTRS1987

Effectiveness of back-to-back testing

Three models of back-to-back testing processes are described. Two models treat the case where there is no intercomponent failure dependence. The third model describes the more realistic case where there is correlation among the failure probabilities of the functionally equivalent components. The theory indicates that back-to-back testing can, under the right conditions, provide a considerable gain in software reliability. The models are used to analyze the data obtained in a fault-tolerant software experiment. It is shown that the expected gain is indeed achieved, and exceeded, provided the intercomponent failure dependence is sufficiently small. However, even with the relatively high correlation the use of several functionally equivalent components coupled with back-to-back testing may provide a considerable reliability gain. Implications of this finding are that the multiversion software development is a feasible and cost effective approach to providing highly reliable software components intended for fault-tolerant software systems, on condition that special attention is directed at early detection and elimination of correlated faults.

Vouk, Mladen A.↗

NASA NTRS1986

On testing of functionally equivalent components of fault-tolerant software

Six functionally equivalent programs were tested with specification based random and extremal/special value (ESV) test cases. Statement and branch coverage were used to measure and compare the attained testing effectiveness. It was observed that both measures reached a nearly steady state value after 25 to 75 random test cases. Coverage saturation curves appear to follow an exponential growth model. However, the steady state values for branch coverage of different components, but the same input cases, differed by as much as 22 percent. The effect is the result of the differences in the detailed structure of the components. Improvement in coverage provided by the random test data, after the ESV cases were executed, was only about 1 percent. Results indicate that extensive random testing can be a process of diminishing returns, and that in the FTS context functional ('black box') testing can provide a very uneven execution coverage of the functionally equivalent software, and therefore should be supplemented by structure based testing.

Vouk, Mladen A.↗

NASA NTRS1986

An experimental evaluation of the effectiveness of random testing of fault-tolerant software

Results of a fault-tolerant software (FTS) experiment are used to show deficiencies of the simple random testing approach. Testing was performed using randomly generated test cases supplemented with extremal and special value (ESV) cases. Error detection efficiency of the random testing approach, with emphasis on correlated errors, was compared to the error detecting capabilities of the ESV data and found deficient. The use of carefully designed test cases as a supplement to random testing, as well as the use of structure based testing are recommended.

Vouk, Mladen A.↗