Table
2 summarizes the main features for the three switchover modes.
o Cold.switchover: This mode requires initialization of the standby component upon
switchover. Of the three, this mode has the slowest recovery time, and it cannot
maintain transaction state; however, it does conserve power and resources.
o Warm.switchover: Data can be saved to the standby component, but the application
needs to be configured at the time of switchover, adding restart time to the
switchover process. This mode cannot support uninterrupted connections during
switchover (if an application has to be rebooted when a server fails, transactions
in process at the time of the failure are lost).
o Hot.switchover: This mode has the fastest recovery time, but it is the most
complex to implement: Application and transaction state data must be continuously
checkpointed to the standby device, faults must be quickly detected, and
the system must be reconfigured. Hot switchover redundancy designs support
the higher levels of availability, but simply having redundancy obviously does
not produce system availability; it is the management of hardware, software,
and data redundancy; the management of faults; and the management of system
reconfiguration that yields availability.
Pages:
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56