Browse Prior Art Database

Isolating and Retaining Bus Information on Extension Units

IP.com Disclosure Number: IPCOM000060778D
Original Publication Date: 1986-May-01
Included in the Prior Art Database: 2005-Mar-09
Document File: 2 page(s) / 27K

Publishing Venue

IBM

Related People

Christiansen, RA: AUTHOR [+2]

Abstract

This article describes a technique for retaining fault information on extension units due to power faults or power interruptions by using one repowering controller to monitor the status of the other repowering controller. When a bus is extended outside the power domain, if the outside domain has a power failure, detection and isolation abilities to that other power domain are lost. The technique disclosed herein provides for logic in two power boundaries, so that information and interrupts generated will not be lost when power is lost in the slave power boundary. Logic in the local power boundary will have the fault information. The concept is illustrated in the block diagram of the drawing.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 78% of the total text.

Page 1 of 2

Isolating and Retaining Bus Information on Extension Units

This article describes a technique for retaining fault information on extension units due to power faults or power interruptions by using one repowering controller to monitor the status of the other repowering controller. When a bus is extended outside the power domain, if the outside domain has a power failure, detection and isolation abilities to that other power domain are lost. The technique disclosed herein provides for logic in two power boundaries, so that information and interrupts generated will not be lost when power is lost in the slave power boundary. Logic in the local power boundary will have the fault information. The concept is illustrated in the block diagram of the drawing. With the separation of power domains by using the power from both of the domains to set status, detection and isolation are kept in the local power domain. By having two cards, one in each power domain with each having its own local power, when a power failure occurs, the status of the fault will be recorded in the local power domain for isolation of the bus and error information. A serial string of these power domains can be as long as required for the information and can be passed in either direction. The power failure in one of the processor power domains does not bring down the system if you have multiple processing units in multiple power domains. Also, concurrent maintenance to power domains can be accomplished...