Restore process using incremental inversion
In one example, a method is provided for creating an inversion patch that can be used to locally restore a file to an earlier version of that file. Initially, a datacenter receives information concerning a target file version that resides at the datacenter and that is associated with a file stored locally at a client. The information includes an identification of the file and an identification of the target file version. Next, byte ranges are identified that are affected by changes that took place with respect to the file after the target file version was created. The byte ranges are then combined to form a unified set of byte ranges. Finally, the original content of the byte ranges in the unified set is identified and used to create the inversion patch.
1. A method for creating an inversion patch, comprising:
receiving, at a datacenter, information concerning a target file version that resides at the datacenter and that is associated with a file stored locally at a client, wherein the information includes an identification of the file and an identification of the target file version;
identifying byte ranges of the locally stored file, wherein the identified byte ranges include bytes of the locally stored file that were changed after the target file version was created;
combining the identified byte ranges to form a unified set of byte ranges;
identifying original content of the changed bytes in the unified set of byte ranges; and
creating an inversion patch that includes the original content.
2. The method as recited in claim 1 , wherein identification of the byte ranges is performed by reviewing respective tables of contents of each incremental backup of the file.
3. The method as recited in claim 1 , wherein one of the byte ranges includes local changes that have been made to the file and which have not been stored at the datacenter.
4. The method as recited in claim 3 , wherein the local changes and associated byte range are identified in the information received by the datacenter concerning the target file version.
5. The method as recited in claim 1 , wherein the original content resides in one or both of an incremental backup that was stored prior to creation of the target file version, and a baseline dataset associated with the file.
6. The method as recited in claim 1 , wherein the file in connection with which the inversion patch is created is one of a database, a mailbox, or a virtual machine disk file.
7. The method as recited in claim 1 , further comprising transmitting the inversion patch to a client from which the target file version information was received.
8. The method as recited in claim 1 , wherein the unified set of byte ranges includes all the changes that occurred in the local copy of the file after the creation of the target file version.
9. The method as recited in claim 1 , wherein identification of the byte ranges is performed without examination of the bytes in the byte ranges.
10. The method as recited in claim 1 , wherein identification of the byte ranges is performed by a local operating system.
11. The method as recited in claim 1 , wherein identification of the original content is performed by searching, in reverse chronological order, each incremental backup of the file that was created prior to creation of the target file version.
12. The method as recited in claim 11 , further comprising searching a baseline dataset after the incremental backups have been searched.
13. A non-transitory storage medium having stored therein computer-executable instructions which, when executed by one or more hardware processors, perform the following operations:
receiving, at a datacenter, information concerning a target file version that resides at the datacenter and that is associated with a file stored locally at a client, wherein the information includes an identification of the file and an identification of the target file version;
identifying byte ranges of the locally stored file, wherein the identified byte ranges include bytes of the locally stored file that were changed after the target file version was created;
combining the identified byte ranges to form a unified set of byte ranges;
identifying original content of the changed bytes in the unified set of byte ranges; and
creating an inversion patch that includes the original content.
14. A physical device, wherein the physical device comprises:
one or more hardware processors; and
the non-transitory storage medium of claim 13 .