Group stamping style asynchronous replication utilizing a loosely-accurate global clock

• US 8,290,899 B2
• Filed: 03/28/2007
• Issued: 10/16/2012
• Est. Priority Date: 03/28/2007
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Webman, Erez, Segev, Danit, Eisinger, Yigal
• Primary Examiner(s): Ruiz, Angelica
• Application Number:
  US11/692,683
• Publication Number:
  US 20080243952A1
• Time in Patent Office: 2,029 Days
• Field of Search: None
• US Class Current: 707/610
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups
• CPC Class Codes: G06F 16/1767 Concurrency control, e.g. optimistic or pessimistic approaches
• CPC Class Codes: G06F 16/2322 using timestamps

Abstract

A group stamping style asynchronous replication service utilizes a loosely-accurate global clock in a data backup and recovery storage environment. The storage environment includes a set of source storage systems illustratively embodied as source replication nodes that cooperate to maintain a consistency group that may span multiple geographical sites. The storage environment also includes one or more target storage systems illustratively embodied as target replication nodes configured to service the consistency group. The group stamping style service utilizes the loosely-accurate global clock to provide consistent replication of a storage space, e.g., a target storage space, of the consistency group.

Claims

1. A method for providing a group stamping style asynchronous replication service in a data backup and recovery storage environment, comprising:

intercepting storage requests at a set of source replication nodes configured to maintain a consistency group that spans multiple geographical sites, wherein at least one of the source replication nodes comprises a storage system;

establishing at each source replication node an interval based on a timestamp of at least one of the source replication nodes and a maximum error associated with the timestamp;

recording the intercepted storage requests intercepted within the interval to a first group of storage requests;

agreeing, among the source replication nodes, on times at which to switch from recording the first group of storage requests to a second group of storage requests;

associating the first group of storage requests with the timestamp and the maximum error;

transmitting the first group of storage requests, the timestamp and the maximum error, from the source replication nodes to one or more target replication nodes configured to service the consistency group; and

identifying the first group of storage requests at the one or more target replication nodes using the associated timestamp.

2. The method of claim 1 wherein agreeing comprises agreeing on the maximum error of the timestamp.

3. The method of claim 2 further comprising:

starting a quiet period at each source replication node using the maximum error and at least one of the times; and

suspending processing of new storage requests during the quiet period.

4. The method of claim 3 further comprising:

processing the suspended new storage requests as part of the second group of storage requests once the quiet period completes.

5. The method of claim 3 further comprising:

detecting, during the quiet period, a global clock error greater than the maximum error at at least one source replication node;

refraining from switching to the second group of storage requests at the at least one source replication node; and

continuing to record the intercepted storage requests to the first group of storage requests at the at least one source replication node.

6. The method of claim 5 further comprising:

merging the transmitted first group of storage requests with the second group of storage requests at the one or more target replication nodes.

7. A system configured to provide a group stamping style asynchronous replication service in a data backup and recovery storage environment, comprising:

a plurality of source replication nodes, at least one source replication node being a storage system, configured to cooperate and maintain a consistency group that spans multiple geographical sites, each source replication node configured to intercept storage requests, each source replication node further configured to establish an interval based on a timestamp of at least one of the source replication nodes and a maximum error associated with the timestamp, and further configured to record the intercepted storage requests intercepted within the interval to a first group of storage requests, each source replication node further configured to agree on times at which to switch from recording the first group of storage requests to a second group of storage requests, each source replication node further configured to associate the first group of storage requests with the timestamp and the maximum error; and

each source replication node further configured to be coupled to one or more target replication nodes configured to service the consistency group and to identify the first group of storage requests using the associated timestamp, each source replication node further configured to transmit the first group of storage requests, the timestamp and the maximum error, to the one or more target replication nodes.

8. The system of claim 7 wherein at least one of the source replication nodes is further configured to buffer the intercepted storage requests, write data associated with the intercepted storage requests, and metadata of the first group as an entry of a buffer.

9. The system of claim 7 wherein at least one of the storage requests is a write operation.

10. The system of claim 7 wherein each of the source replication nodes is further configured to agree on the maximum error of the timestamp.

11. The system of claim 10 wherein at least one of the source replication nodes is further configured to start a quiet period using the maximum error and at least one of the times, and further configured to suspend processing of new storage requests during the quiet period.

12. The system of claim 11 wherein the at least one source replication node is further configured to process the suspended new storage requests as part of the second group of storage requests once the quiet period completes.

13. The system of claim 11 wherein, in response to detecting a global clock error greater than the maximum error during the quiet period, at least one source replication node is further configured to refrain from switching to the second group of storage requests and further configured to continue to record the intercepted storage requests to the first group of storage requests.

14. The system of claim 7 further comprising the one or more target replication nodes configured to merge the transmitted first group of storage requests with the second group of storage requests.

15. The system of claim 7 wherein at least one source replication node comprises a write interception component configured to intercept the storage requests and assign to the first group of storage requests the timestamp and the maximum error.

16. The system of claim 14 wherein the one or more target replication nodes comprise a log configured to store the transmitted first group as a log entry.

17. The system of claim 16 wherein the one or more target replication nodes further comprise a consistency engine configured to establish, maintain and advance a consistency timeline.

18. The system of claim 17 wherein the one or more target replication nodes further comprise an apply engine configured to apply the log entry to a target storage space up to the consistency timeline to create a crash image on the target storage space.

19. A computer readable storage medium containing executable program instructions executed by a processor, comprising:

program instructions that intercept storage requests at a set of source replication nodes configured to maintain a consistency group that spans multiple geographical sites, wherein at least one of the source replication nodes comprises a storage system;

program instructions that establish at each source replication node an interval based on a timestamp of at least one of the source replication nodes and a maximum error associated with the timestamp;

program instructions that record the intercepted storage requests intercepted within the interval to a first group of storage requests;

program instructions that agree, among the source replication nodes, on times at which to switch from recording the first group of storage requests to a second group of storage requests;

program instructions that associate the first group of storage requests with the timestamp and the maximum error;

program instructions that transmit the first group of storage requests, the timestamp and the maximum error, from the source replication nodes to one or more target replication nodes configured to service the consistency group; and

program instructions that identify the first group of storage requests at the one or more target replication nodes using the associated timestamp.