section7 - 7.0 PREDICTION Reliability prediction is useful...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
7-1 7.0 PREDICTION Reliability prediction is useful in a number of ways. A prediction methodology provides a uniform, reproducible basis for evaluating potential reliability during the early stages of a project. Predictions assist in evaluating the feasibility of proposed reliability requirements and provide a rational basis for design and allocation decisions. Predictions that fall short of requirements at any level signal the need for both management and technical attention. In some cases a shortfall in reliability may be offset by the use of fault tolerance techniques. For hardware, adding redundancy will often result in increased reliability. Software reliability may be improved by a focused inspection, defect removal and test effort. Software fault tolerance techniques, such as N-version programming and recovery blocks, are used as a last resort because of the high cost and controversial impact on reliability. Hardware reliability prediction provides a constant failure rate value for the "inherent reliability" of the product, the estimated reliability attainable when all design and production problems have been worked out. A hardware reliability growth model is used to monitor product reliability in the period during which the observed reliability advances toward the inherent reliability. Software reliability prediction provides a projection of the software failure rate at the start of or any point throughout system test. A software reliability growth model covers the period after the prediction, where reliability improves as the result of testing and fault correction. Hardware and software reliability predictions, when adjusted by their respective growth models to coincide with the same point in time, can be combined to obtain a prediction of the overall system reliability. Table 7-1 (page 7-3) lists the software reliability prediction procedures to use during each software development life cycle phase. When system test begins, actual failure data can be used to statistically estimate the growth model parameters (see Section 8). 7.1 Hardware Reliability Prediction . Hardware reliability prediction is a process of quantitatively assessing an equipment design. Techniques have been established so that hardware reliability predictions may be applied and interpreted uniformly. The final outcome of a prediction is a constant failure rate that can be combined with other failure rates in a system model. 7.2 Software Reliability Prediction . Metrics are used to predict a variety of measures including the initial failure rate λ 0 , final failure rate, fault density per executable lines of code, fault profile, as well as the parameters of a software reliability growth model. The final outcomes of a software reliability prediction include: Relative measures for practical use and management.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 06/03/2011 for the course TCS 402 taught by Professor Nitin during the Spring '11 term at Century College.

Page1 / 24

section7 - 7.0 PREDICTION Reliability prediction is useful...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online