Assume that a program will experience 200 failures in infinite time. It has now experienced 100 failures. The initial failure intensity was 20 failures/CPU hr. Then the current failure intensity will be
10 failures/CPU hr.
In Musa's basic execution time model the failure intensity falls linearly with failures experienced: \(\lambda(\mu) = \lambda_{0}\left(1 - \frac{\mu}{\nu_{0}}\right)\).
Here \(\lambda_{0} = 20\), \(\mu = 100\) and \(\nu_{0} = 200\).
\(\lambda = 20\left(1 - \frac{100}{200}\right) = 20 \times 0.5 = 10\) failures/CPU hr.
Check: half the total expected failures have occurred, so the intensity is half of the initial 20.
Hence, 10 failures/CPU hr.
Similar Questions
Maximum possible value of reliability is
Which one of the following is not a source code metric ?
In software testing, how the error, fault and failure are related to each other ?
Which design metric is used to measure the compactness of the program in terms of lines of code ?
Verification :
In which testing strategy requirements established during requirements analysis are validated against developed software ?
Which one of the following is not a definition of error ?
Software testing is
Regression testing is primarily related to
Match List - I with List - II.
List - I (Testing Type) List - II (Description) (A) Unit testing (I) Testing individual components of the software. (B) Integrating testing (II) Testing the interaction between integrated components. (C) System testing (III) Testing to verify the system meets business needs. (D) Acceptance testing (IV) Testing the complete system to ensure it meets requirements. Choose the correct answer from the options given below :
Important Questions from Software Testing
Equivalence partitioning is a ______.
What is the other name of White-box testing?
- Arrange the following types of testing in the order they are usually performed in the software Development life cycle.
A. Integration testing
B. Unit testing
C. System Testing
D. Acceptance Testing
Choose the correct answer from the options given below: - Assertion(A):Rubber necking might be called non generative pattern
Reason(R): Non generative pattern describes a content and a problem but it does not provide any clear cut solution. - Which method can be applied to problems in which the input domain is relatively small but too large to accommodate exhaustive testing?
