Method for creating an application-consistent remote copy of data using remote mirroring

• US 7,464,126 B2
• Filed: 07/21/2005
• Issued: 12/09/2008
• Est. Priority Date: 07/21/2005
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Chen, Ying
• Primary Examiner(s): Vital; Pierre M.
• Assistant Examiner(s): LIANG, VEI CHUNG
• Application Number:
  US11/186,189
• Publication Number:
  US 20070022144A1
• Time in Patent Office: 1,237 Days
• Field of Search: 707/204, 714/6
• US Class Current: 707/655
• CPC Class Codes: G06F 11/2064 while ensuring consistency
• CPC Class Codes: G06F 11/2071 using a plurality of controllers
• CPC Class Codes: G06F 2201/82 Solving problems relating to consistency
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99955 Archiving or backup

Abstract

An application consistent data protection method provides application-assist and replication-technology neutral mirroring that ensures that a remote data copy is application-consistent. The method comprises a coordination protocol to coordinate application hosts across heterogeneous hosts and heterogeneous storage devices. The method utilizes a disk layout and data record format that enables use of an underlying replication ability of a storage device, minimizing development cost and utilizing customer investment. The method comprises on-demand consistency point initiation to minimize performance impact and maximize system resource usage. The method can be applied to both synchronous and asynchronous mirroring and can be incorporated into any virtualization device.

Claims

1. A method of creating an application-consistent remote copy of data using remote mirroring, comprising:

registering at least one of a plurality of application hosts associated with the application as an application-consistent host group;

intercepting a write request from at least one of a plurality of the application hosts;

appending the intercepted write request as a write record in a write record format to a local log in a local replication volume, said record format comprising content-dependent hash head and tail signatures and at least one uniquely verifiable data field;

instructing the application hosts to prepare a consistency point, said preparation including at least quiescing application updates;

generating a consistency point record to identify a set of write records comprising a consistency point data set, said consistency point being appended to said write record;

replicating the write records and the consistency point records to a remote log in a remote replication volume, said records being written in a consecutive region of said remote log;

scanning the remote log for the consistency point record until all the data associated with the consistency point has been replicated;

verifying the validity of a content of each of the write records in the remote log, said validity being determined based on criteria selected from the group consisting of;

valid value boundaries, said hashed head and tail signatures and a sequence number contained in one of said at least one uniquely verifiable data fields;

generating a validated consistency point update transaction after all of said write records have been replicated and validated, said validated consistency point record including at least a last sequence number generated; and

writing the validated write packets in a sequential fashion and the validated consistency point update transaction to a remote storage device to generate an application-consistent remote copy of the consistency point data set, said validated write packets being written in a fashion independent of a manner in which said write packets were replicated.

2. The method of claim 1, wherein the write record format further comprises any one or more of:

a request ID;

an intercept agent ID;

a timestamp;

a hash bock starting offset, S;

a disk block address; and

data from the write request.

3. The method of claim 2, wherein the timestamp is used by a remote agent to identify the write records and the consistency point records.

4. The method of claim 1, wherein the sequence number is used by a remote agent to identify the write records and the consistency point records.

5. The method of claim 1, wherein the sequence number increases sequentially.

6. The method of claim 2, wherein the hash block starting offset is dynamically and randomly generated.

7. The method of claim 2, wherein the head hash signature and the tail hash signature are calculated as hash values of the timestamp, the sequence number, the intercept agent ID, and a predetermined number, N, bytes of the data starting at the hash block starting offset, S.

8. The method of claim 2, further comprising combining an append-only circular log disk layout and the data record format with the content-dependent head hash signature and the tail hash signature hash signature, and uniquely verifiable fields, to allow a remote data processing agent to extract data and to generate a remote copy in a consistency-preserving fashion.

9. The method of claim 8, wherein combining the append-only circular log disk layout comprises appending records at the end of circular log, immediately following a previously appended record.

10. The method of claim 8, wherein the append-only circular log comprises one or more consistency point data sets.

11. The method of claim 10, further comprising deleting each consistency point data set after being written from the remote log in the remote replication volume to the remote storage.

12. The method of claim 1, wherein instructing the application hosts to prepare the consistency point comprises the application hosts completing transient transactions.

13. The method of claim 1, wherein instructing the application hosts to prepare the consistency point comprises the application hosts flushing data in write buffers to the local storage device.

14. The method of claim 1, wherein the consistency point record comprises any one or more of:

a consistency point head hash signature;

a consistency point tail hash signature;

an intercept agent ID;

a consistency point timestamp;

a consistency point sequence number;

a hash block starting offset, S;

a last timestamp of the consistency point data set; and

a last sequence number of the consistency point data set.

15. The method of claim 1, wherein scanning the remote log comprises looking for areas of unreplicated data.

16. The method of claim 15, wherein scanning the remote log comprises pausing until the areas of unreplicated data are filled.

17. The method of claim 1, wherein after writing the validated consistency point update transaction deleting the write records in the consistency point data set and the consistency point update transaction.