Reconstructive error recovery procedure (ERP) for multiple data sets using reserved buffer

• US 8,793,552 B2
• Filed: 11/14/2012
• Issued: 07/29/2014
• Est. Priority Date: 11/14/2012
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Katagiri, Takashi, Nylander-Hill, Pamela R.
• Primary Examiner(s): Decady, Albert
• Assistant Examiner(s): Ahmed, Enam
• Application Number:
  US13/676,493
• Publication Number:
  US 20140136919A1
• Time in Patent Office: 622 Days
• Field of Search: 714/755
• US Class Current: 714/755
• CPC Class Codes: G11B 20/1833 by adding special lists or symbols to the coded information G11B20/1806, G11B20/1866 take precedence
• CPC Class Codes: G11B 2020/1853 using a product code which has inner and outer parity symbols
• CPC Class Codes: H03M 13/29 combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes
• CPC Class Codes: H03M 13/2909 Product codes
• CPC Class Codes: H03M 13/293 with erasure setting
• CPC Class Codes: H03M 13/2957 Turbo codes and decoding

Abstract

In one embodiment, a system includes logic adapted to read a plurality of data sets from a medium one or more times; logic adapted to store portions of some of the data sets to a reserved data buffer when the portions are correctable using C1-error correction code (ECC); logic adapted to aggregate all stored portions for each of the complete data sets to form assembled data sets; logic adapted to determine whether C2-ECC is capable of correcting all errors in the assembled data sets, to correct any remaining errors in the assembled data sets, and to send the corrected data sets to a host when C2-ECC is capable of correcting all errors in the assembled data sets; and logic adapted to reread at least a first uncorrected data set from the medium using a different setting when an error in the first uncorrected data set is not correctable.

Claims

1. A system for reading data, the system comprising:

logic configured to read a plurality of data sets one or more times, each data set comprising a plurality of portions which combine together to wholly form the data set, wherein during each read of at least one of the plurality of data sets, settings are changed in an attempt to better read the at least one of the plurality of data sets;

logic configured to store one or more of the portions of at least one of the plurality of data sets to a reserved data buffer when the one or more portions are correctable using C1-error correction code (ECC) to form one or more stored portions;

logic configured to aggregate, for each of the plurality of data sets that have all portions thereof stored to the reserved data buffer, all stored portions thereof to form one or more assembled data sets;

logic configured to determine whether C2-ECC is capable of correcting all errors in the one or more assembled data sets;

logic configured to correct any remaining errors in the one or more assembled data sets using C2-ECC to form one or more corrected data sets, and to send the one or more corrected data sets to a host when C2-ECC is capable of correcting all errors in the one or more assembled data sets; and

logic configured to reread at least a first uncorrected data set using a different setting in an error recovery procedure (ERP) when an error in the first uncorrected data set is not correctable using C2-ECC or an error in any portion of the first uncorrected data set is not correctable using C1-ECC, wherein the first uncorrected data set is positioned prior to any other uncorrected data set in a read direction.

2. The system as recited in claim 1, wherein the logic configured to reread the at least first uncorrected data set further comprises logic configured to reread all data sets subsequent to the first uncorrected data set.

3. The system as recited in claim 1, further comprising logic configured to store all portions of the plurality of data sets to the reserved data buffer after a first reading of the plurality of data sets.

4. The system as recited in claim 1, wherein each portion is a C1-encoded row in an array representing a data set.

5. The system as recited in claim 1, wherein the one or more stored portions is overwritten when at least one corresponding portion from the one or more portions that are correctable using C1-ECC has fewer corrected errors therein than the one or more stored portions.

6. The system as recited in claim 5, wherein at least a portion of the plurality of data sets is reread, the one or more stored portions in the reserved data buffer is overwritten with portions having less errors corrected therein, and it is determined whether C2-ECC is capable of correcting all errors in the one or more assembled data sets until either all corrected data sets are sent to the host, or until a number of allowable retries is reached and one of the plurality of data sets is reported to the host as unreadable.

7. The system as recited in claim 1, wherein the data sets are read from a magnetic tape, and wherein different settings comprise a change to at least one of:

a tape speed during reading, a data flow correction power, channel parameters, and servo tracking.

8. A tape drive, comprising:

a data buffer comprising a reserved data buffer;

logic configured to read an initial error data set followed by a plurality of data sets subsequent to the initial error data set once or more using different settings until;

all rows of the initial error data set are assembled in a reserved data buffer and no uncorrectable errors remain, or

at least one uncorrectable data set remains in a plurality of assembled data sets, or

a maximum number of rereads has been reached;

logic configured to, after each rereading of the initial error data set and one or more of the plurality of data sets subsequent to the initial error data set in the same forward motion, store rows of each data set which either have no errors or are correctable using C1-error correction code (ECC) to their respective reserved data buffer areas, unless a matching row already exists in the reserved data buffer that has fewer corrected errors therein;

logic configured to assemble the initial error data set from the rows stored for the initial error data set in the reserved data buffer to form an initial assembled data set;

logic configured to correct any remaining errors in the initial assembled data set using C2-ECC to form an initial reconstructed data set; and

logic configured to send the initial reconstructed data set to a host and/or store the initial reconstructed data set to a memory.

9. The tape drive as recited in claim 8, wherein the logic configured to read the plurality of data sets subsequent to the initial error data set more times further comprises logic configured to read all data sets subsequent to and including a first uncorrected data set, wherein the first uncorrected data set has an error that is not correctable using C2-ECC or an error in any row that is not correctable using C1-ECC.

10. The tape drive as recited in claim 8, further comprising logic configured to copy all rows of the plurality of data sets to their respective reserved data buffer areas after a first reading of the plurality of data sets.

11. The tape drive as recited in claim 8, wherein the initial error data set and the data sets are read from a magnetic tape.

12. The tape drive as recited in claim 8, wherein a row comprises a C1-encoded row in an array representing a data set.

13. The tape drive as recited in claim 8, wherein the different settings comprise a change to at least one of:

speed during reading, a data flow correction power, channel parameters, and servo tracking.

14. A method for reading data, the method comprising:

reading a plurality of data sets one or more times, each data set comprising a plurality of portions which combine together to wholly form the data set, wherein during each read of at least one of the plurality of data sets, settings are changed in an attempt to better read the at least one of the plurality of data sets;

storing one or more of the portions of at least one of the plurality of data sets to a reserved data buffer when the one or more portions are correctable using C1-error correction code (ECC) to form one or more stored portions;

aggregating, for each of the plurality of data sets that have all portions thereof stored to the reserved data buffer, all stored portions thereof to form one or more assembled data sets;

determining whether C2-ECC is capable of correcting all errors in the one or more assembled data sets;

correcting any remaining errors in the one or more assembled data sets using C2-ECC to form one or more corrected data sets, and sending the one or more corrected data sets to a host when C2-ECC is capable of correcting all errors in the one or more assembled data sets; and

rereading at least a first uncorrected data set using a different setting in an error recovery procedure (ERP) when an error in the first uncorrected data set is not correctable using C2-ECC or an error in any portion of the first uncorrected data set is not correctable using C1-ECC, wherein the first uncorrected data set is positioned prior to any other uncorrected data set in a read direction.

15. The method as recited in claim 14, wherein the rereading at least the first uncorrected data set further comprises rereading all data sets subsequent to the first uncorrected data set.

16. The method as recited in claim 14, further comprising storing all portions of the plurality of data sets to the reserved data buffer after a first reading of the plurality of data sets.

17. The method as recited in claim 14, wherein each portion is a C1-encoded row in an array representing a data set.

18. The method as recited in claim 14, wherein the one or more stored portions is overwritten when at least one corresponding portion from the one or more portions that are correctable using C1-ECC has less corrected errors therein than the one or more stored portions.

19. The method as recited in claim 18, wherein at least a portion of the plurality of data sets is reread, the one or more stored portions in the reserved data buffer is overwritten with portions having less errors corrected therein, and it is determined whether C2-ECC is capable of correcting all errors in the one or more assembled data sets until either all corrected data sets are sent to the host, or a number of allowable retries is reached and one of the plurality of data sets is reported to the host as unreadable.

20. The method as recited in claim 14, wherein the data sets are read from a magnetic tape.

21. The method as recited in claim 14, wherein the different settings comprise a change to at least one of:

speed during reading, a data flow correction power, channel parameters, and servo tracking.

22. A method for reading data, the method comprising:

reading an initial error data set followed by a plurality of data sets subsequent to the initial error data set once or more using different settings until;

all rows of the initial error data set are assembled in a reserved data buffer and no uncorrectable errors remain, or

at least one uncorrectable data set remains in a plurality of assembled data sets, or

a maximum number of rereads has been reached;

after each rereading of the initial error data set and one or more of the plurality of data sets subsequent to the initial error data set in the same forward motion, storing rows of each data set which either have no errors or are correctable using C1-error correction code (ECC) to their respective reserved data buffer areas, unless a matching row already exists in the reserved data buffer that has fewer corrected errors therein;

assembling the initial error data set from the rows stored for the initial error data set in the reserved data buffer to form an initial assembled data set;

correcting any remaining errors in the initial assembled data set using C2-ECC to form an initial reconstructed data set; and

sending the initial reconstructed data set to a host and/or storing the initial reconstructed data set to a memory.

23. The method as recited in claim 22, wherein the reading the plurality of data sets subsequent to the initial error data set more times further comprises reading all data sets subsequent to and including a first uncorrected data set, wherein the first uncorrected data set has an error that is not correctable using C2-ECC or an error in any row that is not correctable using C1-ECC.

24. The method as recited in claim 22, further comprising copying all rows of the plurality of data sets to their respective reserved data buffer areas after a first reading of the plurality of data sets.

25. The method as recited in claim 22, wherein the initial error data set and the data sets are read from a magnetic tape, and wherein a row comprises a C1-encoded row in an array representing a data set.