Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Cold Stand-by Method on Multiprocessor Workstation

IP.com Disclosure Number: IPCOM000108744D
Original Publication Date: 1992-Jun-01
Included in the Prior Art Database: 2005-Mar-22
Document File: 1 page(s) / 32K

Publishing Venue

IBM

Related People

Tokuda, S: AUTHOR

Abstract

Disclosed is a method to reduce running cost and heat radiation for fault-tolerant workstations.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 100% of the total text.

Cold Stand-by Method on Multiprocessor Workstation

       Disclosed is a method to reduce running cost and heat
radiation for fault-tolerant workstations.

      Conventional fault-tolerant systems have a backup processor
which is continuously powered on to exchange the status of the main
and sub processor.  This costs idling power, whereas it enables quick
recovery in case main processor failure is detected.

      The disclosed method realizes no power consumption of the sub
processor by i) standing-by without power on, and ii) power on after
main processor failure detection.

      The figure demonstrates the required components and the
recovery procedure.
      1) Main processor failure detected.
      2) Detection circuit informs sub power unit of the failure.
      3) Sub power unit turns sub processor on.
      4) Initialize sub processor and work environment.
      5) Sub processor takes over main processor.

      The failure detector and sub power unit procedures 2), 3), and
4) are required to establish the new method.

      This method is also applicable to multiple workstation
environments connected through LAN.  The backup server may not be
powered on until main server failure.