Browse Prior Art Database

An Autonomic Method for Starting Partitions in a Logically Partitioned Environment Disclosure Number: IPCOM000027932D
Original Publication Date: 2004-Apr-12
Included in the Prior Art Database: 2004-Apr-12
Document File: 2 page(s) / 53K

Publishing Venue



Disclosed is an autonomic method for starting partitions in a logically partitioned environment, across diverse types of system power on or restart scenarios.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

An Autonomic Method for Starting Partitions in a Logically Partitioned Environment

A logically partitioned system consists of having one or more logical partitions that are created by the user. Each logical partition has processors, memory, and I/O assigned to it. The user is able to logically power on and off partitions, as well as power on and off the entire system. The Hypervisor is the software entity that is responsible for partitioning the physical system into logical partitions, as directed by the system operator, as well as loading, starting, and stopping logical partitions. There are a number of ways that the system can be started. Here are some examples:

User-Initiated -- A user powers on the system by pushing a power button on the front of the system. Another example is Wake-On LAN, where a "magic packet" is sent to a special Ethernet adapter, and the adapter powers on the platform.

Timed Power On -- A user sets a hardware timer to 7:00am on Monday morning. The user then powers off the system at 6:00pm on Friday. The expectation is that the system will stay powered off during the weekend, and then automatically power on Monday morning.

Auto Power Restart -- The system shuts itself down during a utility power failure event. After utility power is restored, the system automatically powers itself back on.

Platform Re-IPL Due to Firmware Changes -- After the user loads new system firmware (e.g. ,Hypervisor code), the system automatically restarts itself in order to start running with the newest firmware level.

Abnormal Terminations -- An unplanned system outage can result in the Hypervisor abnormally terminating. After a dump is captured, and any failing hardware is removed, the system automatically re-IPLs.

With a multitude of logical partitions on the system, the problem to be solved is how to determine what partitions will automatically start running when the platform is powered on or restarted. A single policy is not sufficient to cover all the examples given above. And, it is not desirable to force the user to have to...