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

Dynamic Erase Record Gap Length Determination using Block Read-While-Write Status

IP.com Disclosure Number: IPCOM000116801D
Original Publication Date: 1995-Nov-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 4 page(s) / 147K

Publishing Venue

IBM

Related People

Bentley, SR: AUTHOR [+4]

Abstract

An improved method for implementing a write Error Recovery Procedure (ERP) in magnetic tape recording channels is disclosed. The mechanism utilizes information from the data flow error correction circuits during a read-while-write operation in order to determine the optimum location for a write retry following a write temporary error.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 42% of the total text.

Dynamic Erase Record Gap Length Determination using Block Read-While-Write
Status

      An improved method for implementing a write Error Recovery
Procedure (ERP) in magnetic tape recording channels is disclosed.
The mechanism utilizes information from the data flow error
correction circuits during a read-while-write operation in order to
determine the optimum location for a write retry following a write
temporary error.

      The ECC decoder provides information regarding media defects
during read-while-write operations via the Threshold Exceeded (TE)
register.  After identifying the portion of a block where the last
errors occur, the write formatter can be instructed to append the
appropriate extended IBG pattern through its EIBG register.

      When writing formatted blocks on tape during a Read-While-Write
(RWW) operation, defects are occasionally encountered on the media
which result in data dropouts to the extent that the readback data is
corrupted.  This may result in a condition where the decoding logic
in the dataflow Error Correction Code (ECC) circuits must utilize
greater correction power than is desireable to correct the data.  If
the data corruption is extensive, an Uncorrectable Error (UE) may
result.  The read-back status from either case initiates a Write
Error Recovery Procedure (ERP) in microcode which attempts to rewrite
the faulty block.

      After repositioning the tape in the drive and arriving at the
append point, the processor instructs the dataflow hardware to write
an Erase Record Gap (ERG) pattern for a certain period of time before
beginning to rewrite the block in error.  This pattern could be a
unique "tape mark" that is similar to, but somewhat different from,
the Inter-Block Gap (IBG) pattern.  The purpose of the ERG is to
relocate the placement of the block on the media in an attempt to
avoid the same defect that caused the write temporary error.

      If the media defect is located near, or at the beginning of,
the block written, it is likely that one or two ERG write retrys will
be sufficient to produce acceptable read-back status.  However, if
the defect is near the end of a large block, the number of ERG retrys
required before getting past the bad section of media could be large,
which adversely affects drive performance and contributes to head and
media wear.

      Ordinarily, the length of the ERG pattern is fixed and the
append point is adjusted forward.  If the maximum number of ERG
retrys are exhausted, the processor posts a Permanent Write Error.
The maximum number of ERGs is predetermined based on the maximum
block size allowable.

      Utilizing a newly developed recording format for future tape
products, much larger data block sizes (containing fixed sub-block
units referred to herein as a Minimum Interleave Unit, or MIU), are
supported.  If the same ERG strategy is used here, the ERG pattern
length would need to be increased in order to contain the po...