Mechanism for converting one type of mirror to another type of mirror on a storage system without transferring data

• US 7,865,475 B1
• Filed: 09/12/2007
• Issued: 01/04/2011
• Est. Priority Date: 09/12/2007
• Status: Expired due to Fees

• Current Assignee: Network Appliance Incorporated (NetApp, Inc.)
• Original Assignee: NetApp, Inc.
• Inventors: Arur, Raghu R., Chitre, Amol, Yadav, Vikas
• Primary Examiner(s): Alam; Hosain T
• Assistant Examiner(s): Ahluwalia; Navneet K
• Application Number:
  US11/854,309
• Time in Patent Office: 1,210 Days
• Field of Search: 707 1- 10, 707100-1041, 707/655, 707/659
• US Class Current: 707/655
• CPC Class Codes: G06F 11/2069 Management of state, configuration or failover
• CPC Class Codes: G06F 11/2082 Data synchronisation
• CPC Class Codes: G06F 16/128 Details of file system snapshots on the file-level, e.g. snapshot creation, administration, deletion use of snapshots for error detection or correction G06F11/14, G06F11/16
• CPC Class Codes: G06F 16/137 Hash-based content-based indexing of textual data G06F16/31
• CPC Class Codes: G06F 16/184 implemented as replicated file system
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0643 Management of files
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0683 Plurality of storage devices

Abstract

Embodiments of the present invention provide a mechanism for converting an existing mirror relationship between an original dataset at a source storage system and a copy of the original dataset at a destination system to a new mirror without discarding the existing copy of the original dataset and without transferring data to create a new dataset to replace the copy of the original dataset at the destination storage system. Rather, a directory structure (including file attributes) of the source storage system is transferred to the destination storage system and a comparison is made to identify file attributes at the destination system that match the transferred file attributes. If a match is found, a mapping between the matched file attributes is created. Subsequent updates to the destination storage system can be done using a technique other than the one that was used to create a copy of the source dataset.

Claims

1. A method comprising:

converting an existing mirror relationship between an original dataset stored at a source storage system and a mirror of the original dataset stored at a destination storage system to a new mirror relationship, a dataset including one or more data containers, the existing mirror relationship converted by;

transferring, from a source conversion module on the source storage system, a first plurality of data container attributes contained on the source storage system'"'"'s active file system in a directory structure, each attribute within the first plurality of data container attributes associated with a data container'"'"'s properties, the first plurality data container attributes transferred to the destination storage system rather than the files themselves;

receiving the first plurality of data container attributes at a destination conversion module on the destination storage system;

comparing, by the destination conversion module on the destination storage system, the first plurality of data container attributes received in the directory structure to the attributes on the destination storage system'"'"'s active file system to find common data container attributes contained on both the source storage system and the destination storage system;

for each matching attribute found in both the first plurality of data container attributes and the second plurality of data container attributes, creating a mapping of an associated inode number on the source storage system to an associated inode number on the destination storage system of each data container associated with the matched attribute; and

establishing a new mirror relationship using the mapping without discarding the mirror and without transferring data from the source storage system to create a new mirror to replace the existing mirror at the destination storage system.

2. The method of claim 1, wherein a data container is a file.

3. The method of claim 1, wherein a data container is a logical unit.

4. The method of claim 1, wherein the existing mirror relationship is a physical mirror and the new mirror relationship is a logical mirror.

5. The method of claim 1, wherein the existing mirror relationship is a logical mirror and the new mirror relationship is a logical mirror.

6. The method of claim 1, wherein creating a mapping further comprises associating an inode of a data container at the source storage system with an inode of a data container having a matched attribute at the destination storage system.

7. A method comprising:

converting an existing mirror relationship between an original dataset stored at a source storage system and a copy of the original dataset stored at a destination storage system to a new mirror relationship, each dataset including one or more data containers, the existing mirror relationship converted by;

transferring, from a source conversion module on the source storage system, a first plurality of data container attributes contained on the source storage system'"'"'s active file system in a directory structure, each attribute within the first plurality of data container attributes associated with a data container'"'"'s properties, the first plurality data container attributes transferred to the destination storage system rather than the files themselves;

receiving the first plurality of data container attributes at a destination conversion module on the destination storage system;

finding attributes of data containers at the destination storage system matching attributes of data containers at the source storage system by comparing the attributes received in the directory structure to the attributes on the destination storage systems active file system;

for each matching attribute found in both the first plurality of data container attributes and the second plurality of data container attributes, changing identifiers of the data containers at the destination storage system having attributes that match the attributes of the data containers at the source storage system; and

establishing the new mirror relationship between the original dataset and the copy of the original dataset without discarding the copy of the original dataset and without transferring data to create a new dataset to replace the copy of the original dataset at the destination storage system.

8. The method of claim 7, wherein an identifier is an inode number identifying a data container and the changing step further comprises:

copying an inode number of the data container having a matching attribute to a new location; and

updating a directory structure to indicate the change of the inode number.

9. A system for converting an existing mirror relationship between an original dataset stored at a source storage system and a copy of the original dataset stored at a destination storage system to a new mirror relationship, each dataset including one or more data containers, the system comprising:

a first conversion module executed at the source storage system configured to transfer a first plurality of data container attributes contained on the source storage system'"'"'s active file system in a directory structure, each attribute within the plurality of attributes associated with a data container'"'"'s properties, the first plurality data container attributes transferred to the destination storage system rather than the files themselves;

a second conversion module executed at the destination storage system configured to receive the first plurality of data container attributes at a destination conversion module on the destination storage system, to find attributes of data containers at the destination storage system matching the received attributes of data containers by comparing the first plurality of data container attributes received in the directory structure to attributes on the destination storage system'"'"'s active file system, and for each matching attribute found in both the first plurality of data container attributes and the second plurality of data container attributes, to create a mapping between the data containers having matching attributes; and

a first mirroring module at the destination storage system configured to provide updates to the copy of the original dataset to reflect changes in the original dataset using the mapping without discarding the copy of the original dataset and without transferring data from the source storage system to create a new dataset to replace the copy of the original dataset at the destination storage system.

10. The system of claim 9, further comprising a second mirroring module at the source storage system configured to:

create a current snapshot at the source storage system;

identify modified data blocks; and

provide, to the destination storage system, the modified data blocks.

11. The system of claim 9, wherein the second conversion module is further adapted to create a mapping of an inode of a data container at the source storage system and an inode of a data container having a matching attribute at the destination storage system.

12. The system of claim 9, wherein the first conversion module and the second conversion module each comprising program code executed on one or more processors of the storage system.

13. The system of claim 9, wherein the first mirroring module comprises program code executed on one or more processors of the storage system.

14. A computer-program product containing executable program instructions executed by a processor, comprising:

a computer-readable medium having computer program code embodied thereon for converting an existing mirror relationship between an original dataset stored at a source storage system and a copy of the original dataset stored at a destination storage system to a new mirror relationship, each dataset including one or more data containers, the computer program code configured to;

transfer, from a source conversion module on the source storage system, a first plurality of data container attributes contained on the source storage system'"'"'s active file system in a directory structure, each attribute within the first plurality of data container attributes associated with a data container'"'"'s properties, the first plurality data container attributes transferred to the destination storage system rather than the files themselves;

receive the first plurality of data container attributes at a destination is conversion module on the destination storage system;

compare, by the destination conversion module on the destination storage system, the first plurality of data container attributes received in the directory structure to attributes on the destination storage system'"'"'s active file system to find common data container attributes contained on both the source storage system and the destination storage system;

for each matching attribute found in both the first plurality of data container attributes and the second plurality of data container attributes, create a mapping of an associated inode number on the source storage system to an associated inode number on the destination storage system of each data container associated with the matched attribute; and

establish a new mirror relationship using the mapping without discarding the copy of the original dataset and without transferring data from the source storage system to create a new dataset to replace the copy of the original dataset at the destination storage system.

15. The computer program code of claim 14, wherein the existing mirror relationship is a physical mirror and the new mirror relationship is a logical mirror.

16. The computer program code of claim 14, wherein the existing mirror relationship is a logical mirror and the new mirror relationship is a logical mirror.