Browse Prior Art Database

Integrated CRC for Sector Buffer Memory and Disk Data

IP.com Disclosure Number: IPCOM000123080D
Original Publication Date: 1998-May-01
Included in the Prior Art Database: 2005-Apr-04
Document File: 2 page(s) / 65K

Publishing Venue

IBM

Related People

Katoh, K: AUTHOR [+4]

Abstract

This article describes a system for using common CRC to assure data on the sector buffer memory and disk such as a hard disk drive that have memory manage logic (MM) to assure data on the sector buffer memory by CRC (cyclic redundancy code) and error correcting code logic (ECC) to avoid miscorrection of the disk data by CRC. Conventional system uses different CRC in MM logic and ECC logic each other. The system should stores two CRCs on the disk. But this system can detect erroneous data on the sector buffer memory and disk without increasing CRC bytes (this system uses only one CRC). This is accomplished by that MM uses ECC's CRC generator polynomial and MM performs CRC generation.

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

Integrated CRC for Sector Buffer Memory and Disk Data

      This article describes a system for using common CRC to
assure data on the sector buffer memory and disk such as a hard disk
drive that have memory manage logic (MM) to assure data on the sector
buffer memory by CRC (cyclic redundancy code) and error correcting
code logic (ECC) to avoid miscorrection of the disk data by CRC.
Conventional system uses different CRC in MM logic and ECC logic each
other.  The system should stores two CRCs on the disk.  But this
system can detect erroneous data on the sector buffer memory and disk
without increasing CRC bytes (this system uses only one CRC).  This
is accomplished by that MM uses ECC's CRC generator polynomial and MM
performs CRC generation.

   The block diagram of writing (host logic (HOST) to Disk)
is shown as Fig 1.  At first HOST transfers a data (Fig 1(1)) to MM,
then MM creates CRC and stores the data and CRC on the sector buffer
memory (Fig 1(2)).  Afterward MM reads it from the sector buffer
memory, and execute CRC check.  If MM detects erroneous data, then
report CRC error else transfers the data and CRC (Fig. 1 (4)) to
drive logic (DRIVE).  DRIVE stores it on the disk and transfers to
ECC (Fig. 1 (5)).  Then ECC creates ECC bytes (Fig (6)) for error
correction from it and transfers it to DRIVE.  Then DRIVE stores the
data, CRC, ECC bytes (Fig. 1 (7)) on the disk.

   The block diagram of reading (Disk to HOST) is shown as
Fig. 2.  At first DRIV...