Browse Prior Art Database

Self Reassign Clean-Up

IP.com Disclosure Number: IPCOM000123443D
Original Publication Date: 1998-Nov-01
Included in the Prior Art Database: 2005-Apr-04
Document File: 4 page(s) / 154K

Publishing Venue

IBM

Related People

Okutsu, K: AUTHOR [+4]

Abstract

Disclosed is a method to clean-up RDM (Reassign Defect Map) for better HDD usability.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Self Reassign Clean-Up

   Disclosed is a method to clean-up RDM (Reassign Defect Map)
for better HDD usability.

   HDD (Hard Disk Drive) has a method to reassign defected
block to other location.  This method is called Reassign.  And RDM
(Reassign Defect Map) is the table which contains the information of
reassign.  RDM area is limited.  When RDM is full, HDD cannot be
usable or HDD causes error easily.  Therefore the method which
prevents RDM full condition is very useful for better usability.
Disclosed here describes this method which is made by following 3
steps.
  1.  Periodic RDM check and clean up RDM.
  2.  This is done at lower interruption level than Host command
      support.  (This prevents performance degradation.)
  3.  When multiple RDM can be cleaned up, temporary RDM image
      in the buffer is written on Disk.  This method prevents
      Host performance degradation.

   Current technique explanation Reassign is done at Read or
at Write operation as follows.
  - Reassign at Read operation
    Sometimes Disk media has the defect at some LBA
    (Logical Block Address) at customer.  Such a LBA
    data can be read by ERP (Error Recovery Procedure)
    and reassigned to other LBA.
  - Reassign at Write operation
    If servo area has defect, servo information cannot be read
    correctly, so write operation cannot be executed to that
    area.  HDD reassigns such a location.  Usually, one servo
    sector has multiple data sector.  Also, data cannot be
    written on following servo sector.  Therefore, if one servo
    sector has the defect, many data sector cannot be written
    and reassigned.

Servo data cannot be read correctly at following case.
  1.  Some kind of shock (Handling damage etc.) caused defect
      on disk media.  This defect case caused error repeatedly.
  2.  Poor mechanical fixture system.  HDD cannot be supported
      well in that system.  This causes bad HDD stability.
  3.  Customer uses HDD in the moving vehicle (car, plane etc.)
  4.  Customer uses HDD at temperature stress condition.  Head
      fly height chang...