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

Fault Prediction Method for Hard Disk Drive

IP.com Disclosure Number: IPCOM000122512D
Original Publication Date: 1991-Dec-01
Included in the Prior Art Database: 2005-Apr-04
Document File: 2 page(s) / 46K

Publishing Venue

IBM

Related People

Arai, K: AUTHOR [+3]

Abstract

Disclosed is a simple method for predicting faults such as a firm seek fault and possible drive fail in the near future.

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

Fault Prediction Method for Hard Disk Drive

      Disclosed is a simple method for predicting faults such
as a firm seek fault and possible drive fail in the near future.

      When a conventional hard disk drive (HDD) reports 'fail', it is
too late.  Sometimes, the fail is the loss of data.  Even if we
proceed to higher quality, defect probability of 0 is unrealizable
unless HDD only consists of parts of infinite MTBF. The fault
prediction is not done currently for low-end storage devices.

      This method gives the following advantages to users:
      (1)  Sends alarm for possible fatal damage.
      (2)  Gives user time for taking back-up of data.
      (3)  Relieves a user of emotional disturbance caused by HDD
fail.

      This method checks the servo pattern to see if seek fault or
read errors occur exceeding its retry limits. The reason for using
the servo pattern is that it has a history of damage after being
shipped from the factory. And the HDD fatal fail is predicted by (1)
increased soft errors (hard errors) or by (2) seek errors.  The
figure shows its algorithm with current realization. The servo
pattern checking method is to get position signals for serveral
revolutions and the extremely bad deviation of one sector defined to
be 'broken', and the HDD sends alarm to the user system.