Measures of Reliability and Availability
Previous work in software reliability attempted to extrapolate the mathematics of hardware reliability theory example for to the prediction of software reliability. Many hardware associated reliability models are predicated on failure due to wear rather than failure because of design defects. In hardware failures due to physical wear for example the effects of temperature corrosion shock are more likely than a design related failure. Sadly the opposite is true for software. In the fact all software failures can be traced to implementation or design problems wear does not enter into the picture.
7 There is still debate over the relationship among key concepts in hardware reliability and their applicability to software example for [LIT89], [ROO901), although an irrefutable link has still to be established it is worthwhile to consider a few simple concepts which apply to both system parts.
Now consider a computer-based system a simple measure of reliability is mean time among failures (MTBF), where
MTBF = MTTF + MTTR
The acronyms MTTR and MTTE are mean time to failure and mean time to respectively, repair.
Various researchers argue which MTBF is a far more useful measure them defects or KLOC. Begin simply an end user is concerned with failures not with the total defect count because each defect contained within a program does not have the similar failure rate the total defect count gives little indication of the reliability of a system example for let's consider a program which has been in operation for fourteen months. Many defects in this program may remain undetected for decades before they are discovered. The MTBF like obscure defect might be the year of 50 or even 100. Other defects in as still undiscovered might have a failure rate if 18 or 24 months. Even if every 1 of the 1st type of defects those with long MTBF is erased the impact on software reliability is negligible.
In the addition to a reliability measure we must established a measure of availability. Software availability is the possibility in which a program is operating according to needs at a given point in time and is defined as:
Availability = MTTE/ (MTTE + MTTR) ∗ 100%
The MTBF reliability measure is similarly sensitive to MTTR and MTTF The availability measure is somewhat must sensitive to MTTR an indirect measure of the maintainability of software.