Continuous Testing - Continuous Integration - page 130

Component A
90% Reliable

Component B
90% Reliable

Component C
90% Reliable

FIGURE 6-1 A system with three components

Each component in this sample system has its reliability measured
and the values are each determined to be 90% (disregard how the 90%
value was determined). If you weren't a systems engineer, you'd prob-
ably figure the reliability of this entire system is then 90%. That
answer, however, isn't correct: .90 * .90 * .90 is actually .73. The over-
all reliability of this system is only 73%.
Ever driven across a bridge that was 73% reliable? If you had a pen
that only worked 73% of the time, wouldn't you throw it out? We
assume that most bridges we drive over are 100% reliable and most
pens we use are 100% reliable until they run out of ink. To gain that
reliability, the builders of bridges and makers of pens ensure reliability
at the lowest possible building block, because that's the only way to
ensure the overall reliability.
This is why in the 1970s sales of Japanese cars began to eclipse
sales of U.S.-made cars. The Japanese manufacturers identified and
applied this principle, and the reliability of Japanese-made cars was sim-
ply much better than their U.S. counterparts. The Japanese manufactur-
ers realized they had to ensure reliability at the lowest possible level.
Now imagine a software system (which, by the way, is nonlinear—
this essentially means you have to also consider the reliability of the
interface or connector between each object). Probably none of us has
ever worked on a software system with three components (i.e., objects)
like in Figure 6-1. Most software systems have hundreds if not thou-
sands of objects! A linear system composed of 100 components each
having 99% reliability would yield a system that is only 37% reliable.
If you wanted to build a software application that had a Service
Level Agreement of 100% (or close), you'd absolutely have to ensure
reliability at the individual object level. If you can't ensure and mea-
sure reliability at the lowest level, you can't possibly do that at the sys-
tem level. Yet this is how we, as an industry, have largely been