Recover corrupted data through speculative bitflip and cross-validation

Patent metadata

Field	Value
Publication number	US-12393485-B2
Application number	US-202218054246-A
Country	US
Kind code	B2
Filing date	Nov 10, 2022
Priority date	Jan 28, 2022
Publication date	Aug 19, 2025
Grant date	Aug 19, 2025

Abstract

Recovering corrupted data through speculative bitflip and cross-validation, including altering a first data portion stored in a storage device, wherein the first data portion contains an error; calculating a hash value using the altered first data portion; comparing the hash value calculated using the altered first data portion to a stored hash value for the first data portion; and based on the comparing, determining whether the altering corrects the error in the first data portion.

First claim

What is claimed is:

1. A method comprising:
generating, after detection of an error caused by corrupted data in a first data portion in a storage device, an altered first data portion, wherein the first data portion is included within a plurality of data portions of a RAID stripe;
calculating a hash value using the altered first data portion;
comparing the hash value calculated using the altered first data portion to a stored hash value for the first data portion;
based on the comparing, determining that the altering corrects the error in the first data portion by reconstructing a second data portion of the RAID stripe using the altered first data portion to cross-validate the altered first data portion; and
overwriting the first data portion using the altered first data portion.

2. The method of claim 1 , wherein generating the altered first data portion further comprises flipping at least one bit in the first data portion.

3. The method of claim 2 , wherein flipping the at least one bit further comprises flipping the at least one bit within a header portion of the first data portion.

4. The method of claim 1 , wherein calculating the hash value using the altered first data portion further comprises calculating a cyclic redundancy check (CRC) value using the altered first data portion.

5. The method of claim 1 further comprising:
reconstructing a second data portion of the RAID stripe with parity calculations that use the altered first data portion;
calculating a second hash value using the reconstructed second data portion;
determining that the calculated second hash value matches a stored hash value within the second data portion; and
based on the determination, determining that the altering corrects the error in the first data portion.

6. The method of claim 5 , further comprising:
determining that the second data portion is unavailable; and
based on the determination, obtaining the stored hash value for the second data portion from the storage device, wherein the storage device stores the stored hash value in a primary header associated with the second data portion.

7. The method of claim 5 , further comprising:
determining that a first number of the plurality of data portions of the RAID stripe does not contain an error;
based on the determination, reconstructing a third data portion with parity calculations that are performed using the third data portion and a first subset of the plurality of data portions that includes the altered first data portion; and
performing a memory comparison between the third data portion and the reconstructed third data portion.

8. An apparatus comprising:
a memory; and
a processing device operatively coupled to the memory, the processing device configured to:
generate, after detection of an error caused by corrupted data in a first data portion in a storage device, an altered first data portion, wherein the first data portion is included within a plurality of data portions of a RAID stripe;
calculate a hash value using the altered first data portion;
compare the hash value calculated using the altered first data portion to a stored hash value for the first data portion;
based on the comparing, determine that the altering corrects the error in the first data portion by reconstructing a second data portion of the RAID stripe using the altered first data portion to cross-validate the altered first data portion; and
overwrite the first data portion using the altered first data portion.

9. The apparatus of claim 8 , wherein the processing device is further configured to flip at least one bit in the first data portion.

10. The apparatus of claim 9 , wherein the processing device is further configured to flip the at least one bit within a header portion of the first data portion.

11. The apparatus of claim 8 , wherein the processing device is further configured to calculate a cyclic redundancy check (CRC) value using the altered first data portion.

12. The apparatus of claim 8 , wherein the processing device is further configured to:
reconstruct a second data portion of the RAID stripe with parity calculations that use the altered first data portion;
calculate a second hash value using the reconstructed second data portion;
determine that the calculated second hash value matches a stored hash value within the second data portion; and
based on the determination, determine that the altering corrects the error in the first data portion.

13. The apparatus of claim 12 , wherein the processing device is further configured to:
determine that the second data portion is unavailable; and
based on the determination, obtain the stored hash value for the second data portion from the storage device, wherein the storage device stores the stored hash value in a primary header associated with the second data portion.

14. The apparatus of claim 12 , wherein the processing device is further configured to:
determine that a first number of the plurality of data portions of the RAID stripe does not contain an error;
based on the determination, reconstruct a third data portion with parity calculations that are performed using the third data portion and a first subset of the plurality of data portions that includes the altered first data portion; and
perform a memory comparison between the third data portion and the reconstructed third data portion.

15. A non-transitory computer readable storage medium storing instructions which, when executed, cause a processing device to:
generate, after detection of an error caused by corrupted data in a first data portion in a storage device, an altered first data portion, wherein the first data portion is included within a plurality of data portions of a RAID stripe;
calculate a hash value using the altered first data portion;
compare the hash value calculated using the altered first data portion to a stored hash value for the first data portion;
based on the comparing, determine that the altering corrects the error in the first data portion by reconstructing a second data portion of the RAID stripe using the altered first data portion to cross-validate the altered first data portion; and
overwrite the first data portion using the altered first data portion.

16. The non-transitory computer readable storage medium of claim 15 , wherein the processing device is further configured to flip at least one bit in the first data portion.

17. The non-transitory computer readable storage medium of claim 15 , wherein the processing device is further configured to calculate a cyclic redundancy check (CRC) value using the altered first data portion.

18. The non-transitory computer readable storage medium of claim 15 , wherein the processing device is further configured to:
reconstruct a second data portion of the RAID stripe with parity calculations that use the altered first data portion;
calculate a second hash value using the reconstructed second data portion;
determine that the calculated second hash value matches a stored hash value within the second data portion; and
based on the determination, determine that the altering corrects the error in the first data portion.

19. The non-transitory computer readable storage medium of claim 18 , wherein the processing device is further configured to:
determine that the second data portion is unavailable; and
based on the determination, obtain the stored hash value for the second data portion from the storage device, wherein the storage device stores the stored hash value in a primary header associa

Assignees

• Pure Storage Inc

Inventors

• Fay Matthew
• Wang Runmin
• Zhao Randy
• Bernat Andrew

Classifications

• G06F11/1088

  Reconstruction on already foreseen single or plurality of spare disks · CPC title

• G06F11/108

  Parity data distribution in semiconductor storages, e.g. in SSD · CPC title

• G06F2201/84

  Using snapshots, i.e. a logical point-in-time copy of the data · CPC title

• G06F11/1435

  using file system or storage system metadata · CPC title

• G06F11/2097

  maintaining the standby controller/processing unit updated (initialisation or re-synchronisation thereof G06F11/1658 and subgroups) · CPC title