Data reallocation upon detection of errors

• US 9,678,864 B2
• Filed: 12/03/2014
• Issued: 06/13/2017
• Est. Priority Date: 12/03/2014
• Status: Active Grant

• Current Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Original Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Inventors: Cai, Jun, Sutiawan, AndiSumaryo, Kella, Jeetandra, Ong, ChuanPeng, Gaertner, Mark Allen, Edgar, Brian T.
• Primary Examiner(s): Chappell, Daniel C
• Application Number:
  US14/559,327
• Publication Number:
  US 20160162208A1
• Time in Patent Office: 923 Days
• Field of Search: 711103
• CPC Class Codes: G06F 11/1016 Error in accessing a memory location, i.e. addressing error
• CPC Class Codes: G06F 11/1072 in multilevel memories
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory
• CPC Class Codes: G06F 2206/1014 One time programmable [OTP] memory, e.g. PROM, WORM
• CPC Class Codes: G06F 2212/7201 Logical to physical mapping or translation of blocks or pages
• CPC Class Codes: G11B 20/10527 Audio or video recording; Data buffering arrangements G11B20/12 - G11B20/18 take precedence
• CPC Class Codes: G11B 20/1217 on discs
• CPC Class Codes: G11B 20/1889 with discs

Abstract

A device includes one or more data storage media having a main storage area, and includes a non-volatile cache memory and a controller. The controller stores a plurality of data packets into a plurality of physical locations in the main storage area. Each of the data packets is associated with a different logical block address (LBA), and each of the physical locations is associated with a different physical location address (PLA). The controller generates mapping information that links the different LBAs to the different PLAs. Upon detecting a soft error when reading a data packet stored in a physical location, the controller relocates the data packet associated with the soft error to a physical location of the non-volatile cache memory. The controller also marks the physical location as a suspect location. The controller updates the mapping information to reflect the relocation of the data packet associated with the soft error.

Claims

1. A method comprising:

storing a plurality of data packets into a plurality of physical locations in a main storage area of one or more data storage media, wherein each of the plurality of data packets is associated with a different logical block address (LBA), and wherein each of the plurality of physical locations is associated with a different physical location address;

generating mapping information that links the different LBAs associated with the different data packets to the different physical location addresses associated with the different physical locations;

upon detecting a soft error when reading at least one data packet of the plurality of data packets stored in at least one physical location of the plurality of physical locations, relocating the at least one data packet associated with the soft error to at least one physical location of a non-volatile memory;

when the soft error is detected, marking the at least one physical location of the plurality of physical locations as a suspect location, wherein the suspect location marking is indicative of an intent to carry out later-in-time damage testing of the at least one physical location, and wherein any determination as to whether the relocation of the at least one data packet to the at least one physical location in the non-volatile memory is temporary or permanent is deferred until after the later-in-time damage testing of the at least one physical location with the suspect location marking is carried out; and

updating the mapping information to reflect the relocation of the at least one data packet associated with the soft error to the at least one physical location in the non-volatile memory.

2. The method of claim 1 and further comprising, upon detecting the soft error, incrementing a soft reallocation count of the at least one physical location of the plurality of physical locations, wherein the soft reallocation count is indicative of a number of data reallocations from the at least one physical location of the plurality of physical locations resulting from soft errors being detected at that physical location.

3. The method of claim 2 and further comprising flagging the at least one physical location of the plurality of physical locations as damaged if the soft reallocation count of the at least one physical location of the plurality of physical locations is greater than a predetermined soft reallocation count threshold.

4. The method of claim 1 and wherein the non-volatile memory is a media cache included in the one or more data storage media or a non-volatile memory that is of a different type than the one or more data storage media.

5. The method of claim 1 and wherein storing the plurality of data packets into the plurality of physical locations in the main storage area is carried out using a shingled magnetic recording technique.

6. The method of claim 1 and wherein the plurality of data packets is a part of a first outer codeword, and wherein the first outer codeword comprises the plurality of data packets and a first outer codeword redundancy.

7. The method of claim 6 and wherein the first outer codeword redundancy is stored in a physical location, in the main storage area, which is different from the plurality of physical locations at which the plurality of data packets is stored.

8. The method of claim 7 and further comprising, upon detecting a soft error when reading at least one data packet of the plurality of data packets stored in at least one physical location of the plurality of physical locations, relocating the plurality of data packets without the first outer codeword redundancy to a plurality of physical locations of the non-volatile memory.

9. The method of claim 8 and further comprising generating a second outer codeword redundancy for the plurality of data packets relocated to the non-volatile memory, the second outer codeword redundancy and the plurality of data packets comprising a second outer codeword.

10. The method of claim 7 and further comprising, upon detecting a soft error when reading at least one data packet of the plurality of data packets stored in at least one physical location of the plurality of physical locations, relocating the plurality of data packets and the first outer codeword redundancy to a plurality of physical locations of the non-volatile memory.

11. A method comprising:

detecting a soft error when reading a data packet stored in a non-volatile location of a data storage medium;

recovering the data packet from the non-volatile location;

when the soft error is detected, marking the non-volatile location as suspect, wherein the suspect location marking is indicative of an intent to carry out later-in-time damage testing of the non-volatile location;

when the soft error is detected, relocating the data packet to a different non-volatile location, wherein any determination as to whether the relocation of the data packet to the different non-volatile location is temporary or permanent is deferred until after the later-in-time damage testing of the non-volatile location with the suspect location marking is carried out; and

updating mapping information to reflect the relocation.

12. The method of claim 11 and further comprising carrying out the later-in-time damage testing of the non-volatile location marked as suspect to determine whether the suspect non-volatile location is damaged.

13. The method of claim 12 and further comprising storing a received data packet in the suspect non-volatile location if a determination is made that the suspect non-volatile location is not damaged.

14. The method of claim 11 and further comprising incrementing a soft reallocation count of the suspect non-volatile location, wherein the soft reallocation count is indicative of a number of data reallocations from the at least one physical location of the plurality of physical locations resulting from soft errors being detected at that physical location.

15. The method of claim 11 wherein the data storage medium comprises a main storage area having a plurality of main storage physical locations including the suspect non-volatile location, and wherein the different non-volatile location is one of a plurality of media cache locations, and wherein the data packet is one of a plurality of data packets protected by an outer codeword redundancy, and wherein the plurality of data packets and the outer codeword redundancy are stored in the plurality of main storage physical locations.

16. The method of claim 15 and further comprising relocating the plurality of data packets to at least some of the plurality of media cache locations.

17. The method of claim 15 and further comprising relocating only the data packet associated with the soft error to the different non-volatile location without relocating the remaining ones of the plurality of data packets and the outer codeword redundancy to the media cache such that a participation of the relocated data packet associated with the soft error in an outer codeword that comprises the plurality of data packets and the outer codeword redundancy is retained.

18. A device comprising:

one or more data storage media having a main storage area;

a non-volatile cache memory; and

a controller configured to;

store a plurality of data packets into a plurality of physical locations in the main storage area of the one or more data storage media, wherein each of the plurality of data packets is associated with a different logical block address (LBA), and wherein each of the plurality of physical locations is associated with a different physical location address;

generate mapping information that links the different LBAs associated with the different data packets to the different physical location addresses associated with the different physical locations;

upon detecting a soft error when reading at least one data packet of the plurality of data packets stored in at least one physical location of the plurality of physical locations, relocate the at least one data packet associated with the soft error to at least one physical location of a non-volatile memory;

when the soft error is detected, mark that the at least one physical location of the plurality of physical locations as a suspect location, wherein the suspect location marking is indicative of an intent to carry out later-in-time damage testing of the at least one physical location;

update the mapping information to reflect the relocation of the at least one data packet associated with the soft error to the at least one physical location in the non-volatile memory; and

determine as to whether the relocation of the at least one data packet to the at least one physical location in the non-volatile memory is temporary or permanent after the later-in-time damage testing of the at least one physical location with the suspect location marking is carried out.

19. The device of claim 18 and wherein the controller is further configured to, upon detecting the soft error, increment a soft reallocation count of the at least one physical location of the plurality of physical locations, wherein the soft reallocation count is indicative of a number of data reallocations from the at least one physical location of the plurality of physical locations resulting from soft errors being detected at that physical location.

20. The device of claim 19 and wherein the controller is further configured to flag the at least one physical location of the plurality of physical locations as damaged if the soft reallocation count of the at least one physical location of the plurality of physical locations is greater than a predetermined soft reallocation count threshold.