IBM RS/6000 SP Handbook page 30

The SP/CES processor nodes incorporate a dynamic CPU de-allocation
function, which enables the system to detect failing processors and take them
offline without re-booting the system. The system service processor can record
this action and notify the system administrator or service personnel of the
condition. Computing operation continues with the processor deactivated,
allowing repair to be scheduled at a time convenient to your processing
schedule. Functions such as power-on and power-off, reading the service
processor error logs and POST error logs, reading vital product data (VPD), and
accessing NVRAM can be performed remotely.
Managing large systems is always a complex process. For the SP/CES system,
a single graphical control workstation (CWS) that displays hardware, software,
jobs, and user status makes system management easier. The system
administrator uses the CWS and the PSSP software product (it is delivered with
the SP standard nodes or it has to be ordered for the SP-attached servers and
the Clustered Enterprise Servers) to perform management tasks including user
and password management, job accounting (as well as system
startup/shutdown), monitoring, and partitioning.
In addition, the SP/CES system offers a wide range of open system management
software tools for operation and administration, availability, deployment, and
security management. Included are the Tivoli Suite software for system
management, Tivoli Storage Manager for backup and recovery, LoadLeveler for
job scheduling and workload balancing, Performance Toolbox (PTX) for
performance monitoring, Distributed Computing Environment (DCE) and
Kerberos for security.
As a general-purpose parallel computer, the SP/CES systems are used
productively in a wide range of application areas and environments in the
high-end UNIX technical and commercial computing market. This broad-based
success is attributable to the highly flexible and general-purpose nature of the
system. RS/6000 SP system can be used for doing a variety of tasks including
parallel, serial, batch, and interactive jobs.
The SP system is a member of the RS/6000 product family, and it is positioned in
the top end of entire pSeries and RS/6000 product line. The following are the
features and benefits of the SP system:
Single point of administrative control
Makes system management easier with less expertise and time required for
most tasks.
Comprehensive system management
Provides the tools required to install, operate, and control an SP system and
helps maintains consistency with other enterprise AIX systems management
tools.
6 RS/6000 SP and Clustered IBM ^ pSeries Systems Handbook