Browse Prior Art Database

Location of Failing Component in Variable Configuration System

IP.com Disclosure Number: IPCOM000114237D
Original Publication Date: 1994-Nov-01
Included in the Prior Art Database: 2005-Mar-28
Document File: 2 page(s) / 57K

Publishing Venue

IBM

Related People

Apperley, N: AUTHOR [+4]

Abstract

Computer peripheral products may be obtained as low-cost (entry-level) systems or as fully fault-tolerant machines. It is advantageous for the customer to be able to upgrade an entry-level machine at any time to add the fault-tolerance feature. Various devices such as power-supplies and fans need to be monitored so that the correct action can be taken. In an entry-level machine, any component failure must cause the machine to abort data transfers and report the error (by means of a front-panel indicator and/or message to the operator console). In a fault-tolerant machine it must report the error, but continue to handle customer data. However, if a second component subsequently fails in the fault-tolerant machine then the procedure must be to abort data transfers.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 59% of the total text.

Location of Failing Component in Variable Configuration System

      Computer peripheral products may be obtained as low-cost
(entry-level) systems or as fully fault-tolerant machines.  It is
advantageous for the customer to be able to upgrade an entry-level
machine at any time to add the fault-tolerance feature.  Various
devices such as power-supplies and fans need to be monitored so that
the correct action can be taken.  In an entry-level machine, any
component failure must cause the machine to abort data transfers and
report the error (by means of a front-panel indicator and/or message
to the operator console).  In a fault-tolerant machine it must report
the error, but continue to handle customer data.  However, if a
second component subsequently fails in the fault-tolerant machine
then the procedure must be to abort data transfers.

      A typical product allowing for various configurations may
include for example three open bays into which any one of three
different components may be plugged.  The customer might order any of
eight valid combinations, and then change the combination at will (an
upgrade).

      To indicate component failure a sense wire, connected to each
component, is monitored for continuity.

      In order to identify a failing component, while minimising the
amount of additional cabling a different output impedance is
presented to the sense wire by each type of component.  Any good
component presents a zero Ohm output impeda...