Browse Prior Art Database

Intermittent Array Failure Identification

IP.com Disclosure Number: IPCOM000060439D
Original Publication Date: 1986-Apr-01
Included in the Prior Art Database: 2005-Mar-08
Document File: 1 page(s) / 11K

Publishing Venue

IBM

Related People

Datres, JH: AUTHOR [+3]

Abstract

"Scrubbing" of data can be used to find problem areas of storage which fail only intermittently. The scrubbing consists of continuous slow speed fetching of all memory locations and correction of all located single bit errors, thus preventing errors caused by either alpha particles or other intermittent phenomena from accumulating and ultimately causing an uncorrectable error. In addition, these intermittent errors are specifically identified by causing a refetch of each corrected error and examining to see if the correction really did occur. In the case where the hardware had corrected the error successfully, that error was an intermittent one and the error data (failing bit and its location) is mapped. The data accumulated in the error map is used to locate any array cards which have an intermittent data problem.

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

Page 1 of 1

Intermittent Array Failure Identification

"Scrubbing" of data can be used to find problem areas of storage which fail only intermittently. The scrubbing consists of continuous slow speed fetching of all memory locations and correction of all located single bit errors, thus preventing errors caused by either alpha particles or other intermittent phenomena from accumulating and ultimately causing an uncorrectable error. In addition, these intermittent errors are specifically identified by causing a refetch of each corrected error and examining to see if the correction really did occur. In the case where the hardware had corrected the error successfully, that error was an intermittent one and the error data (failing bit and its location) is mapped. The data accumulated in the error map is used to locate any array cards which have an intermittent data problem. When the scrub action finds a multiple-bit error, invocation of a store complement/fetch/store complement error-correcting algorithm corrects that multiple bit error using the previously accumulated error data.

1