Browse Prior Art Database

Enhancement to the Dynamic Alternate Assignment of Disk Sectors

IP.com Disclosure Number: IPCOM000042339D
Original Publication Date: 1984-May-01
Included in the Prior Art Database: 2005-Feb-03
Document File: 1 page(s) / 12K

Publishing Venue

IBM

Related People

Pickering, DC: AUTHOR [+3]

Abstract

A method is provided for dynamically assigning alternate sectors on a disk for sectors which repeatedly incur read errors that are recoverable or correctable. A disk sector may show signs of failure prior to the actual failure. This change will cause an alternate assignment for a sector which shows such signs and copy the data while it is still readable. Thus, data loss is reduced. By assigning alternate sectors while the data is still readable by retrying the read operation or by applying ECC correction the method is able to recover the data and keep the user unaffected. In addition, an error log entry is created each time an alternate is assigned. This enables the customer engineer (CE) to observe the extent of damage any disk drive has experienced.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 64% of the total text.

Page 1 of 1

Enhancement to the Dynamic Alternate Assignment of Disk Sectors

A method is provided for dynamically assigning alternate sectors on a disk for sectors which repeatedly incur read errors that are recoverable or correctable. A disk sector may show signs of failure prior to the actual failure. This change will cause an alternate assignment for a sector which shows such signs and copy the data while it is still readable. Thus, data loss is reduced. By assigning alternate sectors while the data is still readable by retrying the read operation or by applying ECC correction the method is able to recover the data and keep the user unaffected. In addition, an error log entry is created each time an alternate is assigned. This enables the customer engineer (CE) to observe the extent of damage any disk drive has experienced. This ability to observe the history and extent of disk damage will allow the CE to take corrective action prior to a solid fault condition occurring. Error handling for a disk drive is shown in the flow diagram of Fig. 1 where block 100 represents the present method which includes the step of maintaining an "Occurred Before" (OB) list containing disk addresses of the last 20 occurrences of the read error (Fig. 2). Each time a recoverable or correctable read error is encountered, the OB list is interrogated to determine whether the sector which currently incurred the error has had any other recent occurrences, i.e., an entry with the failing disk address...