Technique for recovery of trapped storage space in an extent store

• US 9,830,103 B2
• Filed: 01/05/2016
• Issued: 11/28/2017
• Est. Priority Date: 01/05/2016
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Pundir, Anshul, Zheng, Ling, Bolen, David Brittain
• Primary Examiner(s): Sadler, Nathan
• Application Number:
  US14/988,435
• Publication Number:
  US 20170192710A1
• Time in Patent Office: 693 Days
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/0644 Management of space entities, e.g. partitions, extents, pools
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

A technique enables recovery of storage space trapped in an extent store from overlapping write requests associated with metadata describing volume logical storage addresses for data in the extent store. The metadata is organized as metadata entries in a multi-level dense tree metadata structure. When a level of the dense tree is full, the metadata entries of the level are merged with a next lower level of the dense tree in accordance with a dense tree merge operation. The technique may be invoked during the merge operation to process the metadata entries associated with the overlapping write requests involved in the merge operation. Processing of the overlapping write requests during the merge operation may partially overwrite extents which, in turn, may result in logical storage space being trapped in the extent store. The technique may perform read-modify-write (RMW) operations on the partially overwritten extents to recapture that trapped space.

Claims

1. A method comprising:

receiving first and second write requests directed towards a logical unit (LUN), the first and second write requests having respective first and second data, the first and second write requests representing respective first and second offset ranges of the LUN, the second offset range overlapping a portion of the first offset range, the write requests processed at a storage system having a memory;

computing first and second keys from the respective first and second data;

storing the first and second keys in respective first and second data entries of a metadata structure, the first data entry representing the first offset range of the LUN, the second data entry representing the second offset range of the LUN;

storing, in a storage array attached to the storage system, the first and second data as first and second extents respectively;

storing a valid portion of the first extent as a third extent in the storage array; and

purging the first extent and retaining the second extent such that the second extent is accessible via a read request directed towards the LUN after purging the first extent.

2. The method of claim 1 further comprising:

merging the first and second data entries to form a third data entry of the metadata structure, the third data entry storing a third key associated with the third extent.

3. The method of claim 2 further comprising:

performing a read-modify-write operation on the first extent to generate the third extent.

4. The method of claim 3 wherein the read-modify-write operation is performed out-of-band with the merge of the first and second data entries.

5. The method of claim 2 wherein the merging of the first and second data entries further comprises:

inserting an element into a read-modify-write list, the element representing the valid portion of the first extent and including the first key.

6. The method of claim 5 wherein the element further includes a relative offset from a beginning of the first extent denoting the valid portion of the first extent and a length of the valid portion of the first extent.

7. The method of claim 5 further comprising:

inserting the third data entry into the metadata structure such that the third data entry has lower precedence than a fourth data entry inserted into the metadata structure before the third data entry is inserted into the metadata structure.

8. The method of claim 7 further comprising:

in response to determining that the third data entry collides with the fourth data entry, discarding the third data entry.

9. The method of claim 1 wherein purging the first extent further comprises:

dereferencing the first key with the first extent.

10. The method of claim 1 further comprising:

performing a deferred reference count update of the first key.

11. A system comprising:

a storage system having a memory connected to a processor via a bus;

a storage array coupled to the storage system; and

a storage I/O stack executing on the processor of the storage system, the storage I/O stack configured to;

receive first and second write requests directed towards a logical unit (LUN), the first and second write requests having respective first and second data, the first and second write requests representing respective first and second offset ranges of the LUN, the second offset range overlapping a portion of the first offset range;

compute first and second keys from the respective first and second data;

store the first and second keys in respective first and second data entries of a metadata structure, the first data entry representing the first offset range of the LUN, the second data entry representing the second offset range of the LUN;

store, in the storage array, the first and second data as first and second extents respectively;

store a valid portion of the first extent as a third extent in the storage array; and

purge the first extent and retain the second extent such that the second extent is accessible via a read request directed towards the LUN after purging the first extent.

12. The system of claim 11 wherein the storage I/O stack is further configured to:

merge the first and second data entries to form a third data entry of the metadata structure, the third data entry storing a third key associated with the third extent.

13. The system of claim 12 wherein the storage I/O stack is further configured to:

perform a read-modify-write operation on the first extent to generate the third extent.

14. The system of claim 13 wherein the read-modify-write operation is performed out-of-band with the merge of the first and second data entries.

15. The system of claim 11 wherein the storage I/O stack configured to merge the first and second data entries is further configured to:

insert an element into a read-modify-write list, the element representing the valid portion of the first extent and including the first key.

16. The system of claim 15 wherein the element further includes a relative offset from a beginning of the first extent denoting the valid portion of the first extent and a length of the valid portion of the first extent.

17. The system of claim 11 wherein the storage I/O stack is further configured to:

insert the third data entry into the metadata structure such that the third data entry has lower precedence than a fourth data entry inserted into the metadata structure before the third data entry.

18. The system of claim 11 wherein the storage I/O stack configured to purge the first extent is further configured to:

dereference the first key with the first extent.

19. The system of claim 11 wherein the storage I/O stack is further configured to:

perform a deferred reference count update of the first key.

20. A non-transitory computer readable media containing instructions for executing on a central processing unit (CPU) of a storage system having a memory for a method comprising:

receiving first and second write requests directed towards a logical unit (LUN), the first and second write requests having respective first and second data, the first and second write requests representing respective first and second offset ranges of the LUN, the second offset range overlapping a portion of the first offset range;

computing first and second keys from the respective first and second data;

storing the first and second keys in respective first and second data entries of a metadata structure, the first data entry representing the first offset range of the LUN, the second data entry representing the second offset range of the LUN;

storing in a storage array attached to the storage system, the first and second data as first and second extents respectively;

storing a valid portion of the first extent as a third extent in the storage array; and

purging the first extent and retaining the second extent such that the second extent is accessible via a read request directed towards the LUN after purging the first extent.