Browse Prior Art Database

Data Integrity Improvement Method

IP.com Disclosure Number: IPCOM000113606D
Original Publication Date: 1994-Sep-01
Included in the Prior Art Database: 2005-Mar-27
Document File: 4 page(s) / 87K

Publishing Venue

IBM

Related People

Hashimoto, M: AUTHOR [+3]

Abstract

This invention is to improve the storage device data integrity. The peripherals are connected through the bus to the host. Data are protected by the mean of ECC or parity within the host system and bus. But once the peripherals receives the data, it is the peripheral's responsibility to protect the data integrity. This invention is to improve this integrity totally.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Data Integrity Improvement Method

      This invention is to improve the storage device data integrity.
The peripherals are connected through the bus to the host.  Data are
protected by the mean of ECC or parity within the host system and
bus.  But once the peripherals receives the data, it is the
peripheral's responsibility to protect the data integrity.  This
invention is to improve this integrity totally.

      Key Items of Invention - In this section, disk storage is used
to explain the concept.  But this technique can be used by other mean
of peripherals.

      A disk drive receives a command for the request to save the
data to a certain location of the disk.  The disk prepares the data
transfer from the host.

      At this time, data location parameter is set to the Host
Interface Module.  In disk storage case, it is usually defined as a
logical block address.

      Host interface module starts data transfer.  Some sort of error
detection code, such as CRC, will be added to the each block of data.
This detection code is generated based on (Block Address) + (Data).

      (Data) + (Error Detection Code) will travel through internal
hardware, and is finally written on the location specified by the
block address on the disk.

      A disk drive receives a command for the request to retrieve the
data from a certain location on the disk.  The disk internal logic
fetches the data specified by the block address.

      When the data is transferred out from the host interface
module, error detection code is generated based on.(Requested Block
Address) + (Fetched Data).  This error detection code is compared
with the error detection code fetched along with the data from the
disk.  If these detection code does not match, the exception handling
process is invoked to recover the truly requested data.

      Fig. 2.  Data Integri...