System and method for coalescing a plurality of snapshots

• US 7,467,167 B2
• Filed: 03/19/2002
• Issued: 12/16/2008
• Est. Priority Date: 03/19/2002
• Status: Expired due to Term

• Current Assignee: Network Appliance Incorporated (NetApp, Inc.)
• Original Assignee: Network Appliance Incorporated (NetApp, Inc.)
• Inventors: Patterson, Hugo
• Primary Examiner(s): An; Meng-Ai
• Assistant Examiner(s): Truong; Camquy
• Application Number:
  US10/101,889
• Publication Number:
  US 20030182326A1
• Time in Patent Office: 2,464 Days
• Field of Search: 707/102, 707/204, 707/162, 707/1, 711161-162, 711/154, 709/219
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1461 Backup scheduling policy
• CPC Class Codes: G06F 11/1464 for networked environments

Abstract

A system and method for coalescing a plurality of snapshots into a single destination volume file system. Each snapshot and its replicated file system is replicated in a directory associated with the destination file system in which the coalesced snapshots are stored. The destination file system can be associated with a backup server and information from each of the plurality of snapshots can be transferred to the backup server by an respective backup clients over a network. Information can be passed to the backup server from the backup client(s) using a dump and restore process.

Claims

1. A method for coalescing a plurality of snapshots from a plurality of individual file systems to a destination file system, comprising:

creating a plurality of directories within the destination file system, each of the plurality of directories being associated with one of the plurality of snapshots of differing file systems, where each differing file system is associated with a separate client computer, the plurality of directories are stored in a single volume, each snapshot being a copy of a root inode of each client file system at a selected time, the root inode being a file describing an inode file associated with the each client file system;

replicating the plurality of snapshots of differing file system in the plurality of directories, each of the snapshots being replicated in a directory associated with the snapshot;

generating a snapshot of the entire destination file system, wherein each of the snapshots of differing file systems is included in the snapshot of the destination file system;

storing the snapshot of the entire destination file system on one or more storage devices; and

managing the replicating of the plurality of snapshots of differing file systems according to a user-defined methodology, wherein the user-defined methodology includes updating the replicating of the plurality of snapshots of differing file system at a user-defined interval.

2. A backup server to store a plurality of backups from each of a plurality of individual computer systems, comprising:

a plurality of separate client computers each having a computer file system of the plurality of individual file systems;

a volume on the backup server;

a plurality of directories in the volume, each directory of the plurality of directories corresponding to a file system of the plurality of individual file systems;

means for storing in the volume a snapshot of each file system of the plurality of individual file systems, the snapshot being a copy of a root inode of the volume at a selected time, the root inode being a file describing an inode file associated with the each client file system;

means for creating a snapshot of the entire volume, the snapshot being a copy of a root inode of the volume at a selected time, the root inode being a file describing an inode file associated with the entire volume;

means for storing the snapshot of the entire volume to provide backup for the plurality of individual file systems; and

managing the file system associated with each of the plurality of individual file systems according to a user-defined methodology, wherein the user-defined methodology includes updating file system associated with each of the plurality of individual file systems at a user-defined interval.

3. A computer method for backing up a plurality of file systems comprising:

sending a restore stream for each file system to a backup server, each restore stream including a copy of a corresponding file system of the plurality of file systems, where each file system of the plurality of file systems is associated with a separate backup client computer, the copy of a corresponding file system of the plurality of file systems including a snapshot, and each snapshot being a copy of a root inode of each client file system at a selected time, the root inode being a file describing an inode file associated with the each client file system;

in response to receiving each restore stream, storing the copy of the corresponding file system in a separate directory on the backup server;

creating a snapshot of the separate directories on the backup server to include the plurality of file systems in a single snapshot, the separate directories on the backup server being stored in a single volume; and

managing the copy of the corresponding file system accord to a user-defined methodology, wherein the user-defined methodology includes updating the copy of the corresponding file system at a user-defined interval.

4. The method of claim 3, further comprising:

sending the restore stream for each file system by the backup client computer associated with the file system.

5. The method of claim 3, further comprising:

requesting an update of the copy of the corresponding file system, and in response to the backup client computer accepting an update request, transmitting the update to the backup server.

6. A system for backing up a plurality of file systems comprising:

a plurality of separate backup client computers, each backup client computer associated with one of the plurality of file systems, each backup client computer configured to send a restore stream, the restore stream including a copy of the file system associated with the backup client, the copy of a corresponding file system of the plurality of file systems including a snapshot, and each snapshot being a copy of a root inode of each client file system at a selected time, the root inode being a file describing an inode file associated with the each client file system;

a backup server configured to receive the restore streams, the backup server storing the copy of the file system associated with each restore stream in a separate directory on the backup server;

a snapshot of the entire backup server, the snapshot created from all of the separate directories on the backup server, the snapshot including the plurality of file systems associated with the plurality of separate backup client computers;

wherein the separate directories on the backup server are stored in a single volume; and

wherein the copy of the file system associated with each backup client computer is managed according to a user-defined methodology, wherein the user-defined methodology includes updating the copy of the file system associated with the backup client computer at a user-defined interval.

7. The system of claim 6, further comprising:

an update request configured to initiate an update of the copy of the file system associated with the backup client computer, the update request sent by the backup server to the backup client computer.

8. A computer system for backing up a plurality of file systems comprising:

a plurality of client computers with each client computer having a file system and each client computer configured to send a restore stream for each file system to a backup server, each restore stream including a copy of a corresponding file system of the plurality of file systems, wherein the copy of a corresponding file system of the plurality of file systems includes a snapshot, and each snapshot being a copy of a root inode of each client file system at a selected time, the root inode being a file describing an inode file associated with the each client file system;

in response to receiving each restore stream, means for storing the copy of the corresponding file system in a separate directory on the backup server;

means for creating a snapshot of the separate directories on the backup server to include the plurality of file systems in a single snapshot;

wherein the separate directories on the backup server are stored in a single volume; and

managing the copy of the corresponding file system according to a user-defined methodology, wherein the user-defined methodology includes updating the copy of the corresponding file system at a user-defined interval.

9. A computer readable medium containing executable program instructions for execution on a processor for backing up a plurality of file systems, the executable program instructions comprising program instructions for:

sending a restore stream for each file system to a backup server, each restore stream including a copy of a corresponding file system of the plurality of file systems, where each file system of the plurality of file systems is associated with a separate client computer, the copy of a corresponding file system of the plurality of file systems including a snapshot, and each snapshot being a copy of a root inode of each client file system at a selected time, the root node being a file describing an inode file associated with the each client file system;

in response to receiving each restore stream, storing the copy of the corresponding file system in a separate directory on the backup server;

creating a snapshot of the separate directories on the backup server to include the plurality of file systems in a single snapshot, the separate directories on the backup server being stored in a single volume; and

managing the copy of the corresponding file system according to a user-defined methodology, wherein the user-defined methodology includes updating the copy of the corresponding file system at a user-defined interval.

10. A method for backing up a plurality of file systems comprising:

sending an update request to a plurality of backup client computers, wherein the update request requests changes between a previous snapshot of the file system associated with the backup client computer and a new snapshot of the file system associated with the backup client, and where each backup client computer is associated with a different file system of the plurality of file systems;

receiving at the backup server one or more updates, each update from one of the plurality of backup clients, each update including changes between the previous snapshot of the file system associated with the backup client computer and the new snapshot of the file system associated with the backup client computer, the previous snapshot and the new snapshot being a copy of a root inode of each client file system at a selected time, the root inode being a file describing an inode file associated with the each client file system at the selected time;

applying the changes in each update to a copy of a corresponding file system stored on the backup server, the backup server storing copies of each of the plurality of file systems;

creating a snapshot including the copies of the plurality of file systems stored on the backup server to coalesce the plurality of file systems into a single snapshot; and

managing the copy of the corresponding file system according to a user-defined methodology, wherein the user-defined methodology includes updating the copy of the corresponding file system at a user-defined interval.

11. The method of claim 10 further comprising:

receiving, at the backup server, one or more denials from backup client computers, each denial indicating that a new snapshot of the file system associated with the backup client computer does not exist.

12. The method of claim 10 wherein each copy of a file system stored on the backup server is stored in a separate directory.

13. The method of claim 12 wherein all the separate directories are stored in a single volume.

14. A system for backing up a plurality of file systems comprising:

a plurality of separate backup client computers, each backup client computer associated with a different file system of the plurality of file systems, each backup client computer responsive to an update request and configured to send an update that includes changes between a previous snapshot of the file system associated with the backup client computer and a new snapshot of the file system associated with the backup client computer, the previous snapshot and the new snapshot being a copy of a root inode of each client file system at a selected time, the root inode being a file describing an inode file associated with the each client file system;

a backup server configured to receive updates, the backup server configured to apply the changes in each update to a copy of a corresponding file system stored on the backup server, the backup server storing copies of each of the plurality of file systems;

a snapshot of the backup server, the snapshot including the copies of the plurality of file systems stored on the backup server;

wherein all the separate directories are stored in a single volume; and

managing the copy of the corresponding file system according to a user-defined methodology, wherein the user-defined methodology includes updating the copy of the corresponding file system at a user-defined interval.

15. The system of claim 14 further comprising:

each backup client computer is further configured to send a denial indicating that a new snapshot of the file system associated with the backup client computer does not exist.

16. The system of claim 14 further comprising:

each copy of a file system stored on the backup is stored in a separate directory.

17. A computer system having processor for backing up a plurality of file systems comprising:

means for sending an update request to a plurality of backup client computers, wherein the update request requests changes between a previous snapshot of the file system associated with the backup client computer and a new snapshot of the file system associated with the backup client, and where each backup client computer associated with a different file system of the plurality of file systems, the previous snapshot and the new snapshot being a copy of a root inode of each client file system at a selected time, the root inode being a file describing an node file associated with the each client file system;

means for receiving at the backup server one or more updates, each update from one of the plurality of backup client computers, each update including changes between the previous snapshot of the file system associated with the backup client computer and the new snapshot of the file system associated with the backup client computer;

means for applying the changes in each update to a copy of a corresponding file system stored on the backup server, the backup server storing copies of each of the plurality of file systems;

means for creating a snapshot including the copies of the plurality of file systems stored on the backup server to coalesce the plurality of file systems into a single snapshot; and

means for managing the copy of the corresponding file system according to a user-defined methodology, wherein the user-defined methodology includes updating the copy of the corresponding file system at a user-defined interval.

18. A computer readable medium containing executable program instructions for backing up a plurality of file systems, the executable program instructions comprising instructions adapted for:

sending an update request to a plurality of backup client computers, wherein the update request requests changes between a previous snapshot of the file system associated with the backup client computer and a new snapshot of the file system associated with the backup client, and where each backup client computer associated with a different file system of the plurality of file systems, the previous snapshot and the new snapshot being a copy of a root inode of each client file system at a selected time, the root inode being a file describing an inode file associated with the each client file system;

receiving at the backup server one or more updates, each update from one of the plurality of backup client computers, each update including changes between the previous snapshot of the file system associated with the backup client computer and the new snapshot of the file system associated with the backup client computer;

applying the changes in each update to a copy of a corresponding file system stored on the backup server, the backup server storing copies of each of the plurality of file systems;

creating a snapshot including the copies of the plurality of file systems stored on the backup server to coalesce the plurality of file systems into a single snapshot; and

managing the copy of the corresponding file system according to a user-defined methodology, wherein the user-defined methodology includes updating the copy of the corresponding file system at a user-defined interval.

19. A method, comprising:

organizing a backup server with a plurality of directories;

storing in a first directory of the plurality of directories a first snapshot, where the first snapshot is of a first client computer'"'"'s file system;

storing in a second directory of the plurality of directories a second snapshot, where the second snapshot is of a second client computer'"'"'s file system, the first and second snapshot of a client computer'"'"'s file system being a copy of a root inode of each client file system at a selected time, the root inode being a file describing an inode file associated with the each client file system;

generating a single snapshot of the entire backup server, where the single snapshot is of the first snapshot of the first file system and is of the second snapshot of the second file system , wherein the first directory and the second directory are stored in a single volume; and

managing the copy of the first snapshot of the first file system and the copy of the second snapshot of the second file system according to a user-defined methodology, wherein the user-defined methodology includes updating the first snapshot of the first file system and the second snapshot of the second file system at a user-defined interval.

20. The method of claim 19, further comprising:

storing in the first directory a plurality of first snapshots with respect to the first file system to store updates to the first file system;

storing in the second directory a plurality of second snapshots with respect to the second file system to store updates to the second file system;

generating a new single snapshot of the entire backup server, where the new single snapshot comprises a copy of the plurality of first snapshots and a copy of the plurality of second snapshots.