SYSTEM AND METHOD FOR REMOTE ASYNCHRONOUS DATA REPLICATION

• US 20090030986A1
• Filed: 07/23/2008
• Published: 01/29/2009
• Est. Priority Date: 07/27/2007
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: TwinStrata, Inc. (Dell Technologies Inc.)
• Inventors: BATES, JOHN W.
• Granted Patent: US 8,073,922 B2
• Time in Patent Office: Days
• US Class Current: 709/205
• CPC Class Codes: G06F 11/2058 using more than 2 mirrored copies
• CPC Class Codes: G06F 11/2069 Management of state, configuration or failover
• CPC Class Codes: G06F 11/2074 Asynchronous techniques
• CPC Class Codes: G06F 2201/855 Details of asynchronous mirroring using a journal to transfer not-yet-mirrored changes
• CPC Class Codes: G06F 3/0607 by facilitating the process of upgrading existing storage systems, e.g. for improving compatibility between host and storage device
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes

Abstract

An asynchronous peer-to-peer data replication method implemented within a replication cluster comprising at least one master node and at least a first client node includes entering an update in a data volume of the master node and storing the update in a master node storage. Next, updating a first active session in a master log and then sending a first message from the master node to the first client node. The first message comprises a first message content and first “piggybacked” data indicating that the first active session in the master log was updated. Next, receiving the first message by the first client node, registering that the first active session in the master log was updated and signaling internally that the first active session in the master log was updated. Next, sending an update request from the first client node to the master node, processing the update request by the master node and sending the update to the first client node. Finally, receiving the update by the first client node and updating the first active session in a first client log.

Claims

1. An asynchronous data replication method comprising:

providing a replication cluster comprising at least one master node and at least a first client node, wherein said master node comprises a master log and said first client node comprises a first client log;

entering an update in a data volume of said master node and storing said update in a master node storage;

updating a first active session in said master log;

sending a first message from said master node to said first client node wherein said first message comprises a first message content and first “

piggybacked”

data indicating that said first active session in said master log was updated;

receiving said first message by said first client node, registering that said first active session in said master log was updated and signaling internally that said first active session in said master log was updated;

sending an update request from said first client node to said master node;

processing said update request by said master node and sending the update to said first client node; and

receiving the update by said first client node and updating said first active session in said first client log.

2. The method of claim 1 wherein said first active session comprises one or more segments and said master log comprises a master active session matrix storing all session segments for all cluster nodes and said first client log comprises a first client active session matrix storing all session segments for all cluster nodes.

3. The method of claim 2 wherein each of said session segments comprises an index and a timestamp and wherein said session matrices display all said session segments'"'"' indices and timestamps for all cluster nodes.

4. The method of claim 3 wherein said updating of said first active session comprises writing data in an extent structure and entering said extent structure into said first active session.

5. The method of claim 4 wherein said extent structure comprises a length larger than a session segment'"'"'s length and wherein said extent structure is entered into more than one session segments.

6. The method of claim 5 wherein said processing of said update request by said master node comprises locating in said master active session matrix all session segments that have not yet been transmitted to said first client node log and sending said identified session segments to said first client node.

7. The method of claim 6 wherein said session segments that have not yet been transmitted to said first client node log are identified as session segments registered in a row of said master active session matrix corresponding to said first client node having a timestamp earlier than the timestamp of a most recently updated session segment in a row of said master active session matrix corresponding to said master node.

8. The method of claim 7 wherein said replication cluster comprises a second client node and wherein said second client node comprises a second client log and wherein said second client log comprises a second client active session matrix storing all session segments for all cluster nodes.

9. The method of claim 8 wherein said method further comprises sending a second message from said first client node to said second client node and wherein said second message comprises a second message content and second “

piggybacked”

data indicating that said first active session was updated.

10. The method of claim 9 further comprising:

receiving said second message by said second client node, registering that said first active session in said master log and said first client log was updated and signaling internally that said first active session was updated;

sending an update request from said second client node to said first client node;

processing said update request by said first client node and sending the update to said second client node; and

receiving the update by said second client node and updating said first active session in said second client log.

11. The method of claim 10 wherein any of said messages comprises one of a “

health check”

message sent periodically from said master node to any of said client nodes, a “

health check”

message sent periodically from any of said client nodes to said master node, a “

health check”

message sent periodically from any of said client nodes to other client nodes within the replication cluster.

12. The method of claim 1 further comprising completing said first active session in said master log.

13. The method of claim 12 further comprising completing said first active session in said first client log.

14. The method of claim 13 wherein said master log further comprises an authorization and access control protocol for each cluster node.

15. The method of claim 14 wherein said access control protocol comprise one of read-write, read-only, and replicate-only.

16. A data replication system for performing peer-to-peer data replication within a replication cluster comprising at least one master node and at least a first client node, wherein said master node comprises a computing device, a primary storage and a session storage and wherein said computing device comprises a data replication application configured to enter an update in a data volume of said master node and store said update in said master node primary storage, update a first active session in a master log stored in said session storage, send a first message from said master node to said first client node wherein said first message comprises a first message content and first “

piggybacked”

data indicating that said first active session in said master log was updated, and upon receipt of an update request from said first client node process the update request and send the requested update to the client node.

17. The data replication system of claim 16 wherein said first client node comprises a computing device, a primary storage and a session storage and wherein said computing device comprises a data replication application configured to receive said first message by said master node, register that said first active session in said master log was updated and signal internally in said client node that said first active session in said master log was updated, send an update request to said master node, receive the update from said master node and update said a first active session in a first client log stored in said session storage.

18. The data replication system of claim 17 wherein said first active session comprises one or more segments and said master log comprises a master active session matrix storing all session segments for all cluster nodes and said first client log comprises a first client active session matrix storing all session segments for all cluster nodes.

19. The data replication system of claim 18 each of said session segments comprises an index and a timestamp and wherein said session matrices display all said session segments'"'"' indices and timestamps for all cluster nodes.

20. The data replication system of claim 19 wherein said updating of said first active session comprises writing data in an extent structure and entering said extent structure into said first active session.

21. The data replication system of claim 20 wherein said extent structure comprises a length larger than a session segment'"'"'s length and wherein said extent structure is entered into more than one session segments.

22. The data replication system of claim 21 wherein said processing of said update request by said master node comprises locating in said master active session matrix all session segments that have not yet been transmitted to said first client node log and sending said identified session segments to said first client node.

23. The data replication system of claim 22 wherein said session segments that have not yet been transmitted to said first client node log are identified as session segments registered in a row of said master active session matrix corresponding to said first client node having a timestamp earlier than the timestamp of a most recently updated session segment in a row of said master active session matrix corresponding to said master node.

24. The data replication system of claim 23 wherein said replication cluster comprises a second client node and wherein said second client node comprises a computing device, a primary storage and a session storage and wherein said computing device comprises a data replication application configured to receive a second message from said first client node, wherein said second message comprises a second message content and second “

piggybacked”

data indicating that said first active session was updated, register that said first active session in said master log and said first client log was updated and signal internally that said first active session was updated, send an update request to said first client node, receive the update and update said first active session in a second client log.

25. The data replication system of claim 24 wherein any of said messages comprise one of a “

health check”

message sent periodically from said master node to any of said client nodes, a “

health check”

message sent periodically from any of said client nodes to said master node, a “

health check”

message sent periodically from any of said client nodes to other client nodes within the replication cluster.

26. The data replication system of claim 25 wherein said master log further comprises an authorization and access control protocol for each cluster node.

27. The data replication system of claim 26 wherein said access control protocol comprise one of read-write, read-only, and replicate-only.