Data redistribution in data replication systems

• US 8,837,480 B2
• Filed: 11/01/2012
• Issued: 09/16/2014
• Est. Priority Date: 03/31/2009
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Zhong, Hua, Moghe, Dheer, Reddy, Sazzala Venkata
• Primary Examiner(s): Shah, Chirag
• Assistant Examiner(s): Van, Jenkey
• Application Number:
  US13/666,871
• Publication Number:
  US 20130124476A1
• Time in Patent Office: 684 Days
• Field of Search: None
• US Class Current: 370/390
• CPC Class Codes: G06F 16/27 Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes

Abstract

Data replication comprises: redistributing one or more originator data subsets among a plurality of originator nodes; determining data redistribution information pertaining to redistribution of the one or more originator data subsets among the plurality of originator nodes; and sending data redistribution information to a replica system. The data redistribution information is used by the replica system to redistribute one or more corresponding replica data subsets among a plurality of replica nodes; and the one or more corresponding replica data subsets are redistributed among the plurality of replica nodes without requiring the one or more originator data subsets to be sent to the replica system during redistribution.

Claims

1. A system comprising:

one or more processors to;

redistribute one or more originator data subsets among a plurality of originator nodes; and

determine data redistribution information pertaining to redistribution of the one or more originator data subsets among the plurality of originator nodes; and

a communication interface to send data redistribution information to a replica system comprising a plurality of replica nodes;

wherein;

the one or more originator data subsets were initially distributed among the plurality of originator nodes according to an initial distribution arrangement;

redistributing the one or more originator data subsets changes the initial distribution arrangement of the original data subsets to a new distribution arrangement;

the one or more originator data subsets were previously replicated on the replica system as one or more corresponding replica data subsets among the plurality of replica nodes;

the data redistribution information is used by the replica system to redistribute one or more corresponding replica data subsets among the plurality of replica nodes based at least in part on the new distribution arrangement;

the data redistribution information includes identification information of the originator data subsets, identification information of one or more originator nodes on which the originator data subsets were distributed prior to the redistribution, and identification information of one or more current originator nodes to which the originator data subsets have been redistributed; and

the one or more corresponding replica data subsets are redistributed among the plurality of replica nodes without requiring the one or more originator data subsets to be sent to the replica system during redistribution.

2. The system of claim 1, wherein the one or more originator data subsets include one or more data containers.

3. The system of claim 1, wherein the redistribution information includes metadata associated with one or more data containers.

4. The system of claim 1, wherein the redistribution information includes metadata associated with one or more data containers, and each of the one or more data containers includes one or more data segments.

5. The system of claim 1, wherein the one or more originator data subsets and the one or more replica data subsets include identical backup data.

6. The system of claim 1, wherein the plurality of originator nodes are included in a file system.

7. The system of claim 1, wherein the data redistribution information does not include backup data.

8. The system of claim 1, wherein the one or more originator data subsets are redistributed from one or more existing nodes to a newly added node.

9. The system of claim 1, wherein the one or more originator data subsets are redistributed to rebalance load on the plurality of originator nodes.

10. A method for data replication, comprising:

redistributing one or more originator data subsets among a plurality of originator nodes;

determining, using one or more computer processors, data redistribution information pertaining to redistribution of the one or more originator data subsets among the plurality of originator nodes; and

sending data redistribution information to a replica system comprising a plurality of replica nodes;

wherein;

the one or more originator data subsets were initially distributed among the plurality of originator nodes according to an initial distribution arrangement;

redistributing the one or more originator data subsets changes the initial distribution arrangement of the original data subsets to a new distribution arrangement;

the one or more originator data subsets were previously replicated on the replica system as one or more corresponding replica data subsets among the plurality of replica nodes;

the data redistribution information is used by the replica system to redistribute one or more corresponding replica data subsets among the plurality of replica nodes based at least in part on the new distribution arrangement;

the data redistribution information includes identification information of the originator data subsets, identification information of one or more originator nodes on which the originator data subsets were distributed prior to the redistribution, and identification information of one or more current originator nodes to which the originator data subsets have been redistributed; and

the one or more corresponding replica data subsets are redistributed among the plurality of replica nodes without requiring the one or more originator data subsets to be sent to the replica system during redistribution.

11. The method of claim 10, wherein the one or more originator data subsets include one or more data containers.

12. The method of claim 10, wherein the redistribution information includes metadata associated with one or more data containers.

13. The method of claim 10, wherein the redistribution information includes metadata associated with one or more data containers, and each of the one or more data containers includes one or more data segments.

14. The method of claim 10, wherein the one or more originator data subsets and the one or more replica data subsets include identical backup data.

15. A computer program product for data replication, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

redistributing one or more originator data subsets among a plurality of originator nodes;

determining data redistribution information pertaining to redistribution of the one or more originator data subsets among the plurality of originator nodes; and

sending data redistribution information to a replica system comprising a plurality of replica nodes;

wherein;

the one or more originator data subsets were initially distributed among the plurality of originator nodes according to an initial distribution arrangement;

redistributing the one or more originator data subsets changes the initial distribution arrangement of the original data subsets to a new distribution arrangement;

the one or more originator data subsets were previously replicated on the replica system as one or more corresponding replica data subsets among the plurality of replica nodes;

the data redistribution information is used by the replica system to redistribute one or more corresponding replica data subsets among a plurality of replica nodes based at least in part on the new distribution arrangement;

the data redistribution information includes identification information of the originator data subsets, identification information of one or more originator nodes on which the originator data subsets were distributed prior to the redistribution, and identification information of one or more current originator nodes to which the originator data subsets have been redistributed; and

the one or more corresponding replica data subsets are redistributed among the plurality of replica nodes without requiring the one or more originator data subsets to be sent to the replica system during redistribution.

16. A system comprising:

an interface configured to receive data redistribution information from an originator system, the data redistribution information pertaining to redistribution of one or more originator data subsets among a plurality of originator nodes; and

one or more processors configured to redistribute one or more corresponding replica data subsets among a plurality of replica nodes according to the data redistribution information;

wherein;

the one or more originator data subsets were previously replicated on the system as the one or more corresponding replica data subsets among the plurality of replica nodes;

the one or more replica data subsets were initially distributed among the plurality of replica nodes according to an initial distribution arrangement;

redistributing the one or more replica data subsets among the plurality of replica nodes according to the data redistribution information changes the initial distribution arrangement of the replica data subsets to a new distribution arrangement;

the data redistribution information includes identification information of the originator data subsets, identification information of one or more originator nodes on which the originator data subsets were distributed prior to the redistribution, and identification information of one or more current originator nodes to which the originator data subsets have been redistributed; and

the one or more replica data subsets are redistributed among the plurality of replica nodes without requiring the one or more originator data subsets to be sent to the replica system during redistribution.

17. A method for data replication, comprising:

receiving, at a replica system, data redistribution information from an originator system, the data redistribution information pertaining to redistribution of one or more originator data subsets among a plurality of originator nodes; and

redistributing one or more corresponding replica data subsets among a plurality of replica nodes according to the data redistribution information;

wherein;

the one or more originator data subsets were previously replicated on the replica system as the one or more corresponding replica data subsets among the plurality of replica nodes;

the one or more replica data subsets were initially distributed among the plurality of replica nodes according to an initial distribution arrangement;

redistributing the one or more replica data subsets among the plurality of replica nodes according to the data redistribution information changes the initial distribution arrangement of the replica data subsets to a new distribution arrangement;

the data redistribution information includes identification information of the originator data subsets, identification information of one or more originator nodes on which the originator data subsets were distributed prior to the redistribution, and identification information of one or more current originator nodes to which the originator data subsets have been redistributed; and

the one or more replica data subsets are redistributed among the plurality of replica nodes without requiring the one or more originator data subsets to be sent to the replica system during redistribution.

18. A computer program product for data replication, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving data redistribution information from an originator system, the data redistribution information pertaining to redistribution of one or more originator data subsets among a plurality of originator nodes; and

redistributing one or more corresponding replica data subsets among a plurality of replica nodes according to the data redistribution information;

wherein;

the one or more originator data subsets were previously replicated on a replica system as the one or more corresponding replica data subsets among the plurality of replica nodes;

the one or more replica data subsets were initially distributed among the plurality of replica nodes according to an initial distribution arrangement;

redistributing the one or more replica data subsets among the plurality of replica nodes according to the data redistribution information changes the initial distribution arrangement of the replica data subsets to a new distribution arrangement;

the data redistribution information includes identification information of the originator data subsets, identification information of one or more originator nodes on which the originator data subsets were distributed prior to the redistribution, and identification information of one or more current originator nodes to which the originator data subsets have been redistributed; and

the one or more replica data subsets are redistributed among the plurality of replica nodes without requiring the one or more originator data subsets to be sent to the replica system during redistribution.