Reliability, while effective for hardware, is not
enough when complex software is also involved. Software is inherently less reliable,
as every possible configuration of use and data cannot be tested prior to release;
therefore, software glitches are often revealed only when the software is actually put
into use. As systems become more and more software-driven, fault prevention has
become an unrealizable standard, now replaced by fault management. The notion of
reliability has had to be reconsidered in system design.
??? Redundancy??”Replacement. parts. (component. orientation):. Building with redundancy
addresses potential failures by establishing a backup system for critical
components. Full 2N redundancy is the duplication of components such that each
active component has its own backup. It is the simplest redundancy configuration to
manage, since processes can be mirrored on the redundant standby system for total
system failover, if needed. But full redundancy is often prohibitively expensive, unnecessary,
and impractical for complex systems (High Availability and More, 2001).
Utilizing redundancy in complex systems requires dynamic system reconfiguration in
order for the system to remain available.
Pages:
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54