Browse Prior Art Database

Method For Interlocked Deferred Partition Reset In A Logical Partitioning Environment

IP.com Disclosure Number: IPCOM000015984D
Original Publication Date: 2002-Jun-03
Included in the Prior Art Database: 2003-Jun-21
Document File: 1 page(s) / 38K

Publishing Venue

IBM

Abstract

Method For Interlocked Deferred Partition Reset In A Logical Partitioning Environment

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 64% of the total text.

Page 1 of 1

Method For Interlocked Deferred Partition Reset In A Logical Partitioning

Environment

Disclosed is a method to interlock asynchronuos resets/reboots to a logical partition signaled from the partition's operating system or the hardware management console (HMC).

In a logical partitioning computer architecture, IO/CPU/memory resources housed within one large computer system can be partitioned to form sub-computers which exhibit the same behavior and characteristics of a typical machine. These logical partitions are required to exhibit the same operational characteristics of a computer, including the ability to turn them off or reboot them without affecting any of the other logical partitions supported by the larger system framework. Specifically, the reset/reboot signals can originate from the logical partition's operating system or from a HMC external to the host machine.

In either case, these signals are propagated asynchronously to the host machine's service processor which will issue a hardware interrupt to the logical partition's processors, forcing them to execute the reset handler code. If either the host operating system or the external HMC issue a reset request before a prior reset has completed, the reset handler itself will be interrupted, causing it to lose critical state information required to terminate the logical partition gracefully and putting the reliability of future operations in jeopardy.

To avoid such a potentially catastrophic interlocked log...