System and method for backup storage garbage collection
A storage for storing backups of virtual machines includes a persistent storage and a garbage collector. The persistent storage stores the backups. The garbage collector identifies an end of a backup generation session; in response to identifying the end of the backup generation session, the garbage collector performs a date analysis to identify a set of potentially stale backups of the backups; performs a continuity chain analysis of each backup of the set of potentially stale backups to identify a set of stale backups; and deletes each backup of the set of stale backups.
1. A storage for storing backups of virtual machines, comprising:
a persistent storage that stores backups; and
a garbage collector programmed to:
identify an end of a backup generation session;
in response to identifying the end of the backup generation session:
perform a date analysis to identify a set of potentially stale backups of the backups that each has a respective age greater than a threshold age;
perform a continuity chain analysis of each backup of the set of potentially stale backups to identify a set of stale backups that each are not chained to other backups, wherein performing the continuity chain analysis comprises:
selecting a first backup of the set of potentially stale backups
reading a first file name of the first backup;
extracting a first portion of the first file name, wherein the first portion of the first file name references a deleted backup;
performing a first matching of the extracted first portion of the first file name to the respective file names of the backups;
making a first determination that the first backup is not chained to any of the backups based on the first matching, wherein the deleted backup is not in the backups; and
including, based on the first determination, the first backup in the set of stale backups; and
delete each backup of the set of stale backups.
2. The storage of claim 1 , wherein the set of potentially stale backups include a plurality of backups of the backups.
3. The storage of claim 2 , wherein the threshold age is based on an average age of the backups.
4. The storage of claim 1 , wherein performing the continuity chain analysis of each backup of the set of potentially stale backups to identify the set of stale backups further comprises:
selecting a second backup of the set of potentially stale backups;
reading a second file name of the second backup;
extracting a second portion of the second file name;
performing a second matching of the extracted second portion of the second file name to respective file names of the backups;
making a second determination that the second backup is chained to a third backup of the backups based on the second matching; and
excluding the second backup from the set of stale backups.
5. The storage of claim 1 , wherein the first backup of the backups is a member of the set of potentially stale backups.
6. The storage of claim 1 , wherein performing the continuity chain analysis of each backup of the set of potentially stale backups to identify the set of stale backups further comprises:
selecting a second backup of the set of potentially stale backups; and
reading a dependency identifier of the second backup,
wherein the dependency identifier is an identifier of a third backup of the backups.
7. The storage of claim 6 , wherein the first backup is based on an image of a virtual machine of the virtual machines.
8. The storage of claim 7 , wherein the second backup is based on a difference disk of the virtual machine of the virtual machines.
9. The storage of claim 7 , wherein the first backup has a first age, the second backup has a second age, and the second age is smaller than the first age.
10. The storage of claim 7 , wherein the virtual machine of the virtual machines is restorable to a first point in time associated with the first backup using only the first backup.
11. The storage of claim 10 , wherein the virtual machine of the virtual machines is not restorable to a second point in time associated with the second backup using only the second backup.
12. The storage of claim 11 , wherein the virtual machine of the virtual machines is restorable to the second point in time associated with the second backup using both of the first backup and the second backup.
13. A method for storing backups of virtual machines, comprising:
identifying an end of a backup generation session;
in response to the identification of the end:
performing a date analysis to identify a set of potentially stale backups of the backups that each has a respective age greater than a threshold age;
performing a continuity chain analysis of each backup of the set of potentially stale backups to identify a set of stale backups that each are not chained to other backups, wherein performing the continuity chain analysis comprises:
selecting a first backup of the set of potentially stale backups;
reading a first file name of the first backup;
extracting a first portion of the first file name, wherein the first portion of the first file name references a deleted backup;
performing a first matching of the extracted first portion of the first file name to the respective file names of the backups;
making a first determination that the first backup is not chained to any of the backups based on the first matching, wherein the deleted backup is not in the backups; and
including, based on the first determination, the first backup in the set of stale backups; and
deleting each backup of the set of stale backups.
14. The method of claim 13 , wherein the set of potentially stale backups include a plurality of backups of the backups.
15. The method of claim 14 , wherein the threshold age is based on an average age of the backups.
16. A non-transitory computer readable medium comprising computer readable program code, which when executed by a computer processor enables the computer processor to perform a method for storing backups of virtual machines, the method comprising:
identifying an end of a backup generation session;
in response to the identification of the end:
performing a date analysis to identify a set of potentially stale backups of the backups that each has a respective age greater than a threshold age;
performing a continuity chain analysis of each backup of the set of potentially stale backups to identify a set of stale backups that each are not chained to other backups, wherein performing the continuity chain analysis comprises:
selecting a first backup of the set of potentially stale backups
reading a first file name of the first backup;
extracting a first portion of the first file name, wherein the first portion of the first file name references a deleted backup;
performing a first matching of the extracted first portion of the first file name to the respective file names of the backups;
making a first determination that the first backup is not chained to any of the backups based on the first matching, wherein the deleted backup is not in the backups; and
including, based on the first determination, the first backup in the set of stale backups; and
deleting each backup of the set of stale backups.
17. The non-transitory computer readable medium of claim 16 , wherein the set of potentially stale backups include a plurality of backups of the backups.
18. The non-transitory computer readable medium of claim 17 , wherein the threshold age is based on an average age of the backups.