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

Recovery Algorithms on Disk Drives for Certain Errors on Write Operations

IP.com Disclosure Number: IPCOM000019223D
Original Publication Date: 2003-Sep-05
Included in the Prior Art Database: 2003-Sep-05
Document File: 1 page(s) / 39K

Publishing Venue

IBM

Abstract

Disclosed is the use of write verify algorithms to recover from errors written on a disk drive that resulted from either loss of signal or erasure . Those hard errors can subsequently be rewritten reliably at the same location of the original hard error via the proposed algorithm. A certain class of problems and customer applications make this a very cost effective solution -- eliminating the need to replace the drive in question with the subsequent cost and productivity impacts.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 65% of the total text.

Page 1 of 1

Recovery Algorithms on Disk Drives for Certain Errors on Write Operations

The key claim of this disclosure is the implementation of a selective use of the Write-Verify command after a write command confined to a subset of the disk's available cylinders followed by an immediate retry of the preceeding write command if an error is detected. This solves the case of a miswrite that would otherwise go undetected when the design proves weak or marginal in a restricted area of the hard disk drive (HDD). For example a recoverable write error can be created due to some mechanical interaction between the head and disk in the outside diameter (OD) or any other portion of the disk. The error may be due to a temporary local spacing loss induced by, for example, excess lube droplet formation, hydrocarbon contamination or a soft particle contamination In the normal case this would result in data loss to the customer. Other localized recoverable errors can result from head to disk impacts during head unloading off a load/unload ramp results in thermal erasures at the OD. The use of the write-verify after a write is issued to a sector inside the suspect OD region would reduce or eliminate this type of hard error. The impact on performance is small since the frequency of the extra two revolutions required to perform the algorithm will be, in practically all customer applications, very low since customer data will be spread across the disk surface and not confined to the targeted o...