Redundancy In Components; Power And Cooling Redundancy; Memory Redundancy Mechanisms; Section 6.4, "Redundancy In Components - IBM p5 590 System Handbook

Additional testing can be selected at boot time to fully verify the system memory
and check the chip interconnect wiring. When a system reboots after a hard
failure, it performs extended mode tests to verify that everything is working
properly and that nothing was compromised by the failure. This behavior can be
overridden by the systems administrator.

6.4 Redundancy in components

The p5-590 and p5-595 system design incorporates redundancy in several
components to provide fault-tolerant services where failures are not allowed.
Power supplies, fans, blowers, boot disks, I/O links, service processors, and
power cables offer redundancy to eliminate single points of failure. Some of these
features are highlighted, as described in the following sections.

6.4.1 Power and cooling redundancy

The p5-590 and p5-595 provide full power and cooling redundancy, with dual
power cords and variable-speed fans and blowers, for both the central electronics
complex (CEC) and the I/O drawers.
Within the CEC frame, the N+1 power and cooling subsystem provides complete
redundancy in case of failures in the bulk or regulated power supplies, the power
controllers, and the cooling units, as well as the power distribution cables. As on
the zSeries server, concurrent repair is supported on all of the CEC power and
cooling components.
There is also a redundant feature called internal battery features (IBF) designed
to maintain system operation during short moments of power fluctuation
conditions. For full power loss protection, optional uninterruptible power supply
(UPS) systems in addition to, or in place of, the IBF features should be used. You
should see the IBF feature as a redundancy feature only; it will not replace the
UPS capabilities.
In case of a fan or blower failure, the remaining fans automatically increase
speed to compensate for the air flow from the failed component.

6.4.2 Memory redundancy mechanisms

There are several levels of memory protection implemented on the p5-590 and
p5-595 systems. From the internal L1 caches to the main memory, several
features are implemented to assure data integrity and data recovery in case of
memory failures.
IBM Eserver p5 590 and 595 System Handbook
146