Fast incremental backup method and system

• US 10,380,141 B1
• Filed: 09/30/2015
• Issued: 08/13/2019
• Est. Priority Date: 09/30/2014
• Status: Active Grant

• Current Assignee: MidCap Financial Trust
• Original Assignee: Acronis International GmbH (Acronis AG)
• Inventors: Chepel, Dmitry, Per, Yuri, Beloussov, Serguei M., Protasov, Stanislav, Lyadvinsky, Maxim V.
• Primary Examiner(s): Alam, Shahid A
• Application Number:
  US14/870,452
• Time in Patent Office: 1,413 Days
• Field of Search: 707618
• CPC Class Codes: G06F 11/1451 by selection of backup contents
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 16/273 Asynchronous replication or reconciliation
• CPC Class Codes: G06F 16/275 Synchronous replication
• CPC Class Codes: G06F 2201/835 Timestamp

Abstract

Various embodiments are directed to systems and methods for creating an archive of a set of logical storage units stored on at least one storage medium. A first plurality of LSU descriptors may be generated. Each LSU descriptor may describe a current version of one of the set of LSUs at about the time of the generating. A first archive may be read to produce a first archive plurality of LSU descriptors. Each of the first archive plurality of LSU descriptors may describe a version of one of the set of LSUs when the first archive was created. The first archive plurality of LSU descriptors may be compared with the first plurality of LSU descriptors to identify a first LSU from the set of LSUs that has an associated LSU descriptor that does not match a corresponding LSU descriptor from the first archive plurality of LSU descriptors.

Claims

1. A method for creating a logical storage unit-level archive of a set of logical storage units stored on at least one storage medium, the set of logical storage units (LSUs) comprising a plurality of LSUs, the method comprising:

generating, by a processor in communication with the at least one storage medium, a first plurality of LSU descriptors, wherein each LSU descriptor from the first plurality of LSU descriptors describes a current version of one of the set of LSUs at about the time of the generating, wherein the first plurality of LSU descriptors comprises, for each LSU from the set of LSUs, a location descriptor and a timestamp descriptor;

reading, by the processor, from a first archive of the set of LSUs, a first archive plurality of LSU descriptors, wherein each of the first archive plurality of LSU descriptors describes a version of one of the set of LSUs when the first archive was created;

comparing, by the processor, the first archive plurality of LSU descriptors with the first plurality of LSU descriptors;

identifying, by the processor, a first LSU from the set of LSUs that has an associated LSU descriptor from the first plurality of LSU descriptors that does not match a corresponding LSU descriptor from the first archive plurality of LSU descriptors;

generating, by the processor, a second archive of the set of LSUs, wherein the second archive comprises the first file and the first plurality of LSU descriptors;

determining that a timestamp descriptor for a second LSU at the first plurality of LSU descriptors does not match a corresponding timestamp descriptor for the second LSU at the first archive plurality of LSU descriptors;

determining that a location descriptor for the second LSU at the first plurality of LSU descriptors does not match a corresponding location descriptor for the second LSU at the first archive plurality of LSU descriptors;

determining that the second LSU has not changed since when the first archive was created; and

omitting the second LSU from the second archive of the set of LSU.

2. The method of claim 1, wherein the first archive plurality of LSU descriptors are hashed and the first plurality of LSU descriptors are hashed.

3. The method of claim 1, wherein the first archive plurality of LSU descriptors and the first plurality of LSU descriptors each comprise at least one timestamp descriptor and at least one location descriptor.

4. The method of claim 3, wherein the hashed timestamp descriptor comprises a hashed version of a timestamp for an LSU operation associated with the timestamp, wherein the LSU operation is selected from the group consisting of:

creating the LSU;

modifying the LSU;

accessing the LSU; and

modifying a descriptor of the LSU.

5. The method of claim 1, wherein generating the first plurality of LSU descriptors comprises receiving a first LSU descriptor from an entry for the first LSU, the entry at a log file maintained by a file system for the at least one storage medium.

6. The method of claim 1, wherein generating the second archive of the set of LSUs comprises writing to the second archive the first plurality of LSU descriptors and, for each of the first plurality of LSU descriptors, an associated indicator of an LSU from the set of LSUs that is described by the LSU descriptor.

7. The method of claim 1, further comprising examining, by the processor, the first LSU to verify whether the first LSU has changed since when the first archive was created.

8. The method of claim 1, wherein the LSUs from the set of LSUs are selected from the group consisting of a folder and a file.

9. A method for creating a logical storage unit-level archive of a set of logical storage units stored on at least one storage medium, the set of logical storage units (LSUs) comprising a plurality of LSUs, the method comprising:

generating, by a processor in communication with the at least one storage medium, a first plurality of LSU descriptors, wherein each LSU descriptor from the first plurality of LSU descriptors describes a current version of one of the set of LSUs at about the time of the generating;

reading, by the processor, from a first archive of the set of LSUs, a first archive plurality of LSU descriptors, wherein each of the first archive plurality of LSU descriptors describes a version of one of the set of LSUs when the first archive was created;

comparing, by the processor, the first archive plurality of LSU descriptors with the first plurality of LSU descriptors;

identifying, by the processor, a first LSU from the set of LSUs that has an associated LSU descriptor from the first plurality of LSU descriptors that does not match a corresponding LSU descriptor from the first archive plurality of LSU descriptors; and

generating, by the processor, a second archive of the set of LSUs, wherein the second archive comprises the first file and the first plurality of LSU descriptors; and

after generating the second archive of the set of LSUs, generating, by the processor, a third plurality of LSU descriptors, wherein each LSU descriptor of the third plurality of LSU descriptors describes a version of one of the set of LSUs at about the time of the generating;

reading, by the processor, the first plurality of LSU descriptors from the second archive;

comparing, by the processor, the third plurality of LSU descriptors to the first plurality of LSU descriptors;

identifying, by the processor, a second LSU from the set of LSUs that has an associated LSU descriptor from the first plurality of LSU descriptors that does not match a corresponding LSU descriptor from the third plurality of LSU descriptors; and

generating, by the processor, a third archive of the set of LSUs, wherein the third archive comprises the second file and the third plurality of LSU descriptors.