Efficient data consistency verification for flash storage

1. A method, comprising:

obtaining a logical block address (LBA), the LBA being associated with a data block of a file;

reading, from a solid state drive (SSD), data content that corresponds to the LBA, wherein the SSD includes one or more flash storage elements and one or more caches, and wherein the one or more flash storage elements and the one or more caches have different access latencies;

determining whether an access latency associated with the reading of the data content exceeds a threshold;

in response to a determination that the access latency does not exceed the threshold, evaluating the data content to determine whether it is consistently stored in a physical memory included in the SSD; and

in response to a determination that the data content is determined not to be consistently stored in the physical memory included in the SSD, recording an indication indicating that the LBA is not successfully verified.

2. The method of claim 1 , wherein the LBA is among a plurality of LBAs associated with a plurality of data blocks of the file.

3. The method of claim 1 , wherein:

the LBA is among a plurality of LBAs associated with a plurality of data blocks of the file; and

the method further comprises:

obtaining the plurality of LBAs from an inode associated with the file; and

selecting a subset of the plurality of LBAs to form a read list.

4. The method of claim 1 , wherein:

the LBA is among a plurality of LBAs associated with a plurality of data blocks of the file; and

the method further comprises:

obtaining the plurality of LBAs from an inode associated with the file; and

selecting a subset of the plurality of LBAs to form a read list, wherein the selected subset of the plurality of LBAs corresponds to different physical data blocks.

5. The method of claim 1 , wherein:

the evaluating of the data content to determine whether it is consistently stored in the physical memory included in the SSD includes determining whether the data content is all-zero; and

in response to a determination that the data content is determined not to be all-zero, the data content is deemed to be consistently stored in the physical memory.

6. The method of claim 1 , wherein:

the evaluating of the data content to determine whether it is consistently stored in the physical memory included in the SSD includes determining whether the data content is all-zero;

in response to a determination that the data content is determined not to be all-zero, the data content is deemed to be consistently stored in the physical memory; and

in response to a determination that the data content is determined to be all-zero, the evaluating of the data content to determine whether it is consistently stored in the physical memory included in the SSD includes:

determining one or more neighboring LBAs;

determining whether the one or more neighboring LBAs have all-zero data content and access latencies greater than the threshold; and wherein

in response to a determination that the one or more neighboring LBAs have all-zero data content and access latencies greater than the threshold, the data content is deemed to be consistently stored in the physical memory included in the SSD.

7. The method of claim 1 , wherein:

the evaluating of the data content to determine whether it is consistently stored in the physical memory included in the SSD includes determining whether the data content is all-zero;

in response to a determination that the data content is determined not to be all-zero, the data content is deemed to be consistently stored in the physical memory; and

in response to a determination that the data content is determined to be all-zero, the evaluating of the data content to determine whether it is consistently stored in the physical memory included in the SSD further includes:

determining one or more neighboring LBAs;

determining whether the one or more neighboring LBAs have all-zero data content and access latencies greater than the threshold; and

in response to a determination that the one or more neighboring LBAs do not have all-zero data content, the access latencies are not greater than the threshold, or both, performing a file-level cyclic redundancy code (CRC) check.

8. The method of claim 1 , wherein the LBA is deemed to be successfully verified in in response to a determination that the access latency exceeds the threshold.

9. The method of claim 1 , wherein in response to a determination that the data content is found not to be consistently stored in the physical memory included in the SSD, the method further comprises performing an error correction.

10. A system, comprising:

one or more processors configured to:

obtain a logical block address (LBA), the LBA being associated with a data block of a file;

read, from a solid state drive (SSD), data content that corresponds to the LBA, wherein the SSD includes one or more flash storage elements and one or more caches, and wherein the one or more flash storage elements and the one or more caches have different access latencies;

determine whether an access latency associated with the reading of the data content exceeds a threshold;

in response to a determination that the access latency does not exceed the threshold, evaluate the data content to determine whether it is consistently stored in a physical memory included in the SSD; and

in response to a determination that the data content is determined not to be consistently stored in the physical memory included in the SSD, record an indication indicating that the LBA is not successfully verified; and

one or more memories coupled to the one or more processors and configured to provide the one or more processors with instructions.

11. The system of claim 10 , wherein the LBA is among a plurality of LBAs associated with a plurality of data blocks of the file.

12. The system of claim 10 , wherein:

the LBA is among a plurality of LBAs associated with a plurality of data blocks of the file; and

the one or more processors are further configured to:

obtain the plurality of LBAs from an inode associated with the file; and

select a subset of the plurality of LBAs to form a read list.

13. The system of claim 10 , wherein:

the LBA is among a plurality of LBAs associated with a plurality of data blocks of the file; and

the one or more processors are further configured to:

obtain the plurality of LBAs from an inode associated with the file; and

select a subset of the plurality of LBAs to form a read list, wherein the selected subset of the plurality of LBAs corresponds to different physical data blocks.

14. The system of claim 10 , wherein:

to evaluate the data content to determine whether it is consistently stored in the physical memory included in the SSD includes to determine whether the data content is all-zero; and

in response to a determination that the data content is determined not to be all-zero, the data content is deemed to be consistently stored in the physical memory.

15. The system of claim 10 , wherein:

to evaluate the data content to determine whether it is consistently stored in the physical memory included in the SSD includes to determine whether the data content is all-zero;

in response to a determination that the data content is determined not to be all-zero, the data content is deemed to be consistently stored in the physical memory; and

in response to a determination that the data content is determined to be all-zero, to evaluate the data content to determine whether it is consistently stored in the physical memory included in the SSD includes to:

determine one or more neighboring LBAs;

determine whether the one or more neighboring LBAs have all-zero data content and access latencies greater than the threshold; and wherein

in response to a determination that the one or more neighboring LBAs have all-zero data content and access latencies greater than the threshold, the data content is deemed to be consistently stored in the physical memory included in the SSD.

16. The system of claim 10 , wherein:

to evaluate the data content to determine whether it is consistently stored in the physical memory included in the SSD includes to determine whether the data content is all-zero;

in response to a determination that the data content is determined not to be all-zero, the data content is deemed to be consistently stored in the physical memory; and

in response to a determination that the data content is determined to be all-zero, to evaluate the data content to determine whether it is consistently stored in the physical memory included in the SSD includes to:

determine one or more neighboring LBAs;

determine whether the one or more neighboring LBAs have all-zero data content and access latencies greater than the threshold; and

in response to a determination that the one or more neighboring LBAs do not have all-zero data content, the access latencies are not greater than the threshold, or both, perform a file-level cyclic redundancy code (CRC) check.

17. The system of claim 10 , wherein the LBA is deemed to be successfully verified in in response to a determination that the access latency exceeds the threshold.

18. The system of claim 10 , wherein in response to a determination that the data content is found not to be consistently stored in the physical memory included in the SSD, the system further comprises performing an error correction.

19. The system of claim 10 , wherein the one or more flash storage elements include a NAND flash.

20. A non-transitory computer program product for data verification, the non-transitory computer program product being embodied in a tangible computer readable storage medium and comprising computer instructions for:

obtaining a logical block address (LBA), the LBA being associated with a data block of a file;

reading, from a solid state drive (SSD), data content that corresponds to the LBA, wherein the SSD includes one or more flash storage elements and one or more caches, and wherein the one or more flash storage elements and the one or more caches have different latencies;

determining whether an access latency associated with the reading of the data content exceeds a threshold;

in response to a determination that the access latency does not exceed the threshold, evaluating the data content to determine whether it is consistently stored in a physical memory included in the SSD; and

in response to a determination that the data content is determined not to be consistently stored in the physical memory included in the SSD, recording an indication indicating that the LBA is not successfully verified.

21. A method, comprising:

obtaining a logical block address (LBA), the LBA being associated with a data block of a file;

reading, from a solid state drive (SSD), data content that corresponds to the LBA, wherein the SSD includes one or more flash storage elements and one or more caches, and wherein the one or more flash storage elements and the one or more caches have different access latencies;

determining whether an access latency associated with the reading of the data content exceeds a threshold;

in response to the access latency exceeding the threshold, recording an indication indicating that the LBA is successfully verified.

22. The method of claim 1 , wherein the indication includes a location of the data content.

23. The system of claim 10 , wherein the indication includes a location of the data content.