Trimming unused blocks from a versioned image backup of a source storage that is stored in a sparse storage

1. A method for pruning unused blocks from a versioned image backup of a source storage that is stored in a sparse storage, the method comprising:

identifying a versioned image backup of a source storage that is stored in a sparse storage, the sparse storage storing first and second versions of the versioned image backup representing states of the source storage at successive first and second points in time, respectively;

during a retention process, deleting the first version of the versioned image backup from the sparse storage;

identifying a first set of blocks that were allocated in the source storage at the second point in time;

identifying a second set of blocks that are allocated in the sparse storage for the second version of the versioned image backup;

determining a third set of blocks that are included in the second set of blocks but are not included in the first set of blocks; and

after the retention process, trimming the third set of blocks from the second version of the versioned image backup that is stored in the sparse storage.

2. The method of claim 1 , wherein:

the sparse storage is a sparse block device;

the versioned image backup is stored in the sparse block device as a sparse partition of the sparse block device; and

the identifying of the first set of blocks further includes calculating offsets for the first set of blocks that correspond to actual positions of the blocks as stored in the sparse block device.

3. The method of claim 2 , wherein the sparse block device is a RADOS sparse Block Device (RBD), a ZFS sparse block device, or a BTRFS sparse block device.

4. The method of claim 3 , wherein the trimming of the third set of blocks includes trimming the third set of blocks from the second version of the versioned image backup that is stored in the sparse storage by sending one or more trim commands to a storage system of the RBD, of the ZFS sparse block device, or of the BTRFS sparse block device.

5. The method of claim 1 , wherein the sparse storage is a sparse file configured to be presented to a virtual machine as a block device by a hypervisor.

6. The method of claim 1 , wherein the trimming of the third set of blocks includes trimming the third set of blocks from the second version of the versioned image backup that is stored in the sparse storage when a system resource utilization falls below a predefined threshold.

7. One or more non-transitory computer-readable media storing one or more programs that are configured, when executed, to cause one or more processors to perform the method as recited in claim 1 .

8. A method for pruning unused blocks from a versioned image backup of a source storage that is stored in a sparse storage, the method comprising:

identifying a versioned image backup of a source storage that is stored in a sparse storage, the sparse storage storing first, second, and third versions of the versioned image backup representing states of the source storage at successive first, second, and third points in time, respectively;

during a retention process, deleting the first version of the versioned image backup from the sparse storage;

identifying a second set of blocks that were allocated in the source storage at the second point in time;

identifying a third set of blocks that were allocated in the source storage at the third point in time;

determining a first set of blocks to include blocks that are indicated as allocated in either of the second set of blocks or the third set of blocks;

identifying a fourth set of blocks that are allocated in the sparse storage for the third version of the versioned image backup;

determining a fifth set of blocks that are included in the fourth set of blocks but are not included in the first set of blocks; and

after the retention process, trimming the fifth set of blocks from the third version of the versioned image backup that is stored in the sparse storage.

9. The method of claim 8 , wherein:

the sparse storage is a sparse block device;

the versioned image backup is stored in the sparse block device as a sparse partition of the sparse block device; and

the determining of the first set of blocks further includes calculating offsets for the first set of blocks that correspond to actual positions of the blocks as stored in the sparse block device.

10. The method of claim 9 , wherein the sparse block device is a RADOS sparse Block Device (RBD), a ZFS sparse block device, or a BTRFS sparse block device.

11. The method of claim 10 , wherein the trimming of the fifth set of blocks includes trimming the fifth set of blocks from the third version of the versioned image backup that is stored in the sparse storage by sending one or more trim commands to a storage system of the RBD, of the ZFS sparse block device, or of the BTRFS sparse block device.

12. The method of claim 8 , wherein the sparse storage is a sparse file configured to be presented to a virtual machine as a block device by a hypervisor.

13. The method of claim 8 , wherein the trimming of the fifth set of blocks includes trimming the fifth set of blocks from the third version of the versioned image backup that is stored in the sparse storage when a system resource utilization falls below a predefined threshold.

14. One or more non-transitory computer-readable media storing one or more programs that are configured, when executed, to cause one or more processors to perform the method as recited in claim 8 .

15. A method for pruning unused blocks from a versioned image backup of a source partition that is stored in a sparse block device, the method comprising:

identifying a versioned image backup of a source partition that is stored as a sparse partition in a sparse block device, the sparse block device storing first and second versions of the versioned image backup representing states of the source partition at successive first and second points in time, respectively, a size of the source partition being reduced between the first and second points in time;

during a retention process, deleting the first version of the versioned image backup from the sparse block device;

identifying a first set of blocks to include blocks that are included in the source partition at the second point in time;

identifying a second set of blocks to include blocks that are included in the sparse partition at the second point in time;

determining a third set of blocks that are included in the second set of blocks but are not included in the first set of blocks; and

after the retention process, trimming the third set of blocks from the second version of the versioned image backup that is stored in the sparse block device.

16. The method of claim 15 , wherein:

the determining of the first set of blocks further includes calculating offsets for the first set of blocks that correspond to actual positions of the blocks as stored in the sparse block device.

17. The method of claim 16 , wherein:

the sparse block device is a RADOS sparse Block Device (RBD), a ZFS sparse block device, or a BTRFS sparse block device; and

the trimming of the third set of blocks includes trimming the third set of blocks from the second version of the versioned image backup that is stored in the sparse block device by sending one or more trim commands to a storage system of the RBD, of the ZFS sparse block device, or of the BTRFS sparse block device.

18. The method of claim 15 , wherein the trimming of the third set of blocks includes trimming the third set of blocks from the second version of the versioned image backup that is stored in the sparse block device when a system resource utilization falls below a predefined threshold.

19. One or more non-transitory computer-readable media storing one or more programs that are configured, when executed, to cause one or more processors to perform the method as recited in claim 15 .