Browse Prior Art Database

Method of Spindle Control for Raid Subsystem

IP.com Disclosure Number: IPCOM000118006D
Original Publication Date: 1996-Aug-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 2 page(s) / 38K

Publishing Venue

IBM

Related People

Iwasa, H: AUTHOR [+3]

Abstract

Disclosed is a method of spindle control for the RAID Subsystem to improve reliability and sustain performance of the RAID Subsystem by extending the life of HDDs and preventing deterioration of the HDD's error rate. This is achieved by controlling the rotation of the spindle motor of the HDDs which are used in the RAID subsystem. The following procedure shows an example of this method. This assumes several HDDs, named No. 1, No. 2-No. N, which are working in the RAID subsystem: 1. Select one of the HDDs, for example, select No. 1 HDD. 2. Change RAID mode from normal condition mode to 1 HDD failure mode. Failure HDD is assumed as No. 1 HDD and RAID subsystem works by (N-1) HDDs. 3. Power off No. 1 HDD. 4. Wait a moment. 5. Power on No. 1 HDD again. 6.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 90% of the total text.

Method of Spindle Control for Raid Subsystem

      Disclosed is a method of spindle control for the RAID Subsystem
to improve reliability and sustain performance of the RAID Subsystem
by extending the life of HDDs and preventing deterioration of the
HDD's error rate.  This is achieved by controlling the rotation of
the spindle motor of the HDDs which are used in the RAID subsystem.
The following procedure shows an example of this method.  This
assumes several HDDs, named No. 1, No. 2-No. N, which are working in
the RAID subsystem:
  1.  Select one of the HDDs, for example, select No. 1 HDD.
  2.  Change RAID mode from normal condition mode to 1 HDD
       failure mode.  Failure HDD is assumed as No. 1 HDD and
       RAID subsystem works by (N-1) HDDs.
  3.  Power off No. 1 HDD.
  4.  Wait a moment.
  5.  Power on No. 1 HDD again.
  6.  Check if the host accessed RAID subsystem during RAID
       subsystem is 1 HDD failure mode.
  7.  If accessed, then generate the data which should have be
       written into No. 1 HDD and write that data into No. 1 HDD.
  8.  Check whether all HDD's power have been once off at item 3.
  9.  If all HDD's power have been once off, then complete this
       method.
 10.  Else return to item 1, but select HDD which have not been
       yet powered off.

      This method doesn't limit the controlling spindle motor to
power off/on HDDs, but the following method has the equivalen...