Fast backup storage and fast recovery of data (FBSRD)

• US 7,596,713 B2
• Filed: 05/04/2005
• Issued: 09/29/2009
• Est. Priority Date: 11/20/2002
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Mani-Meitav, Irit, Sarfati, Assaf
• Primary Examiner(s): Maskulinski; Michael C
• Assistant Examiner(s): Arcos; Jeison C
• Application Number:
  US11/122,508
• Publication Number:
  US 20050216788A1
• Time in Patent Office: 1,609 Days
• Field of Search: 709/202, 709/203, 707/202, 707/200, 714/6, 714/52
• US Class Current: 714/6.32
• CPC Class Codes: G06F 11/1451 by selection of backup contents
• CPC Class Codes: G06F 11/1456 Hardware arrangements for backup
• CPC Class Codes: G06F 11/1458 Management of the backup or restore process
• CPC Class Codes: G06F 11/1464 for networked environments
• CPC Class Codes: G06F 11/1469 Backup restoration techniques
• CPC Class Codes: G06F 2201/84 Using snapshots, i.e. a logical point-in-time copy of the data
• CPC Class Codes: Y10S 707/99953 Recoverability

Abstract

A Fast Backup Storage and fast Recovery of Data (FBSRD) method for a facility, preferably with a SAN, coupled to a network with servers and workstations, operating in both a storage mode and a recovery mode. Coupled to the network are a primary storage, a repository, and a Backup Appliance computer BA. The BA runs a Backup Computer Program in association with at least one computer. When in storage mode, data is retrieved out of primary storage for back up, by taking snapshots and saving the retrieved data into repository in block format. In recovery mode, backed-up data is retrieved from repository and recovered into primary storage in either one of both block format and file format. The BA runs interactively with an Agent Computer Program residing in each server and workstation, and with a Backup User Interface management computer program operated by a user.

Claims

1. A method for data storage and recovery, comprising:

accepting files, which comprise data arranged in a file format supporting objects of variable size, for storage in a primary storage device having a file system;

extracting the data for back-up from the primary storage device by producing incremental snapshots of the data on the primary storage device at successive time intervals;

backing-up the extracted data by storing the incremental snapshots in a repository in accordance with a block format having fixed-size data blocks;

prior to receiving a request to recover at least one of the files from the repository, processing the incremental snapshots in the repository so as produce file system information applicable at the successive time intervals; and

in response to receiving the request, reading one or more of the incremental snapshots from the repository, and processing the read incremental snapshots using the previously-produced file-system information so as to reconstruct the data in the at least one of the files in accordance with the file format.

2. The method according to claim 1, wherein extracting and backing-up the data comprise communicating between the repository and the primary storage device over a Storage Area Network (SAN).

3. The method according to claim 1, wherein reading the one or more of the incremental snapshots is performed when at least some of the data in the primary storage device is unusable, and wherein the method comprises

writing the at least one of the files containing the reconstructed data into the primary storage device.

4. The method according to claim 1, wherein processing the incremental snapshots to produce the file system information comprises, for a given snapshot, constructing and storing in the repository a catalog comprising the file system information and a Look-Up Table (LUT) comprising block-level data location information, such that the catalog and the LUT information are mutually-indexed for association with the data in the given snapshot, and wherein processing the read incremental snapshots using the previously-produced file-system information comprises associating indexed information stored in the catalog and in the LUT with respect to the at least one of the files, and recovering the at least one of the files using the associated information.

5. The method according to claim 4, wherein recovering the at least one of the files comprises accessing the repository over a network.

6. The method according to claim 1, wherein processing the read incremental snapshots using the previously-produced file-system information comprises communicating between the repository and the primary storage device over a Storage Area Network (SAN).

7. The method according to claim 1, wherein processing the read incremental snapshots using the previously-produced file-system information comprises connecting a computer to the repository via a Storage Area Network (SAN), running a Read File System (RFS) on the computer, creating, using the RFS, a representation of the data blocks stored in a selected snapshot in the repository, running another file system of the computer to obtain from the repository information that mutually indexes file system information and block-level data location information regarding the at least one of the files, so as to convert file data requests to block requests, accepting block read requests from the another file system of the computer and retrieving the data blocks requested in the block requests from the selected snapshot over the SAN.

8. The method according to claim 1, wherein processing the read incremental snapshots using the previously-produced file-system information comprises selecting a recovery format from a group of formats consisting of the block format and the file format, and processing the read incremental snapshots so as to produce the reconstructed data in accordance with the selected recovery format.

9. The method according to claim 8, wherein, when the selected recovery format comprises the block format, processing the read incremental snapshots comprises selecting a source snapshot, allocating a target location in the primary storage device for storage of the data that is recovered from the source snapshot, and copying the data from the source snapshot to the target location.

10. The method according to claim 9, wherein processing the read incremental snapshots comprises accessing the repository and the primary storage device over a Storage Area Network (SAN).

11. The method according to claim 9, wherein processing the read incremental snapshots further comprises temporarily protecting the target location from access operations when copying the data from the source snapshot.

12. The method according to claim 8, wherein, when the selected recovery format comprises the block format, processing the read incremental snapshots comprises:

running a Read File System (RFS) on a server coupled to the repository via a Storage Area Network (SAN);

selecting a source snapshot in the repository and representing the source snapshot as a selected file system;

accessing file data and file meta-data related to the source snapshot using another file system running on the server;

accepting requests by the RFS to read data blocks from the another file system directed to the source snapshot using data retrieved from a Look Up Table (LUT) pertaining to the source snapshot and stored in the repository, so as to locate in the repository the data blocks requested in the requests; and

reading the data blocks requested in the requests from the repository.

13. The method according to claim 1, wherein extracting and backing-up the data comprise temporarily protecting the extracted data from write operations during extraction and backing-up of the data.

14. The method according to claim 1, wherein producing the incremental snapshots comprises including in a current snapshot only a selected subset of the data blocks that are detected as valid.

15. The method according to claim 1, wherein producing the incremental snapshots comprises including in a current snapshot only a selected subset of the data blocks that are detected as changed with respect to another snapshot preceding the current snapshot.

16. The method according to claim 1, wherein storing the incremental snapshots comprises:

running an agent on a client computer, which is connected locally to the primary storage device and by a network to the repository;

sending a command to the agent, via the network, to monitor a selected region of the primary storage device;

responsively to the command, running the agent for creating an initially-empty list of incremental data blocks residing in the monitored region, and assigning the list a unique ID;

using the agent, intercepting a write instruction sent from the client computer to the primary storage device, and adding to the list a location in the primary storage data location to which the intercepted instruction is targeted; and

selecting and associating a snapshot storage operation with the unique ID assigned to the list, and copying the list from the client computer via the network.