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

Retrying Commands During Handling of Error Conditions in a Data Storage Control Apparatus

IP.com Disclosure Number: IPCOM000078411D
Original Publication Date: 1972-Dec-01
Included in the Prior Art Database: 2005-Feb-25
Document File: 1 page(s) / 12K

Publishing Venue

IBM

Related People

Hancock, RJ: AUTHOR [+2]

Abstract

In a microprogrammed data storage control apparatus, such as a control unit for a disk drive employing the count, key and data-field track format of U. S. Patent 3,299,410, the handling of error conditions can be effected by retrying the command with which the error was associated. When the error condition occurs, analysis of the error yields a retry status indicator which is presented to the using system to indicate that the command is to be retried, or reissued, by the central processing unit (CPU) or data channel.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 68% of the total text.

Page 1 of 1

Retrying Commands During Handling of Error Conditions in a Data Storage Control Apparatus

In a microprogrammed data storage control apparatus, such as a control unit for a disk drive employing the count, key and data-field track format of U. S. Patent 3,299,410, the handling of error conditions can be effected by retrying the command with which the error was associated. When the error condition occurs, analysis of the error yields a retry status indicator which is presented to the using system to indicate that the command is to be retried, or reissued, by the central processing unit (CPU) or data channel.

Retry execution begins with preliminary control operations. If the error was a seek error in which the access mechanism was moved to an incorrect track, then the seek function is retried. If the error was a data read error which was uncorrectable by the control unit error-correction code (ECC), then the access mechanism can be incrementally shifted a fraction of a track width to either side of the nominal data track, in an effort to correctly read misaligned data.

In general, as a data record is read from the disk to the control unit enroute to the CPU, the Home Address and the count and the key fields are buffered in the control unit. For ECC correctable errors in these fields, correction takes place directly in the buffer. Since the length of a data field may be too long for buffering, these fields are not buffered in the control unit and ECC corrections of da...