Browse Prior Art Database

Error Correcting Code for Storage Subsystem

IP.com Disclosure Number: IPCOM000079206D
Original Publication Date: 1973-May-01
Included in the Prior Art Database: 2005-Feb-26
Document File: 3 page(s) / 23K

Publishing Venue

IBM

Related People

Bossen, DC: AUTHOR [+3]

Abstract

In a storage hierarchy environment, such as disc and tape systems connected as shown in the figure, it is advantageous to use the same error-correcting code (ECC) in both levels.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 55% of the total text.

Page 1 of 3

Error Correcting Code for Storage Subsystem

In a storage hierarchy environment, such as disc and tape systems connected as shown in the figure, it is advantageous to use the same error- correcting code (ECC) in both levels.

Traditionally, the ECC's of disc and tape are never compatible, because the disc is normally a more reliable device than the tape and it uses a serial read/write mode versus the parallel read/write operation of the tape. This tradition should be changed when the tape is also a serial read/write device. If it is, conversions between tape and disc can be eliminated, so that a direct communication link (Analog or Digital) can be used between the two levels. This is accomplished by:
1) Using the ECC of the tape for the disc.

providing each track on the disc into independently decodable

segments, corresponding to tracks in the tape.
3) Performing no detection and correction at either the tape

or disc level.
4) Providing an ECC unit in the storage subsystem control unit.

Detection and correction are made when data enters and

leaves the storage subsystem.

Though the use of the same ECC on both disc and tape will cause extra redundancy for the disc, the direct communication link seems to offer a better payoff. If a separate ECC is still desired for disc for other reasons, a very compatible code for both units is available which has the following characteristics:
1) Data bit length = 255 x 16 x b = 4112 x b, where b denotes the number of blocks per sec...