Method, apparatus and program storage device for maintaining data consistency and cache coherency during communications failures between nodes in a remote mirror pair

• US 7,120,824 B2
• Filed: 05/09/2003
• Issued: 10/10/2006
• Est. Priority Date: 05/09/2003
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Burton, David Alan, Stewart, Alan Lee, Otterness, Noel Simen
• Primary Examiner(s): Beausoliel, Robert
• Assistant Examiner(s): Ehne, Charles
• Application Number:
  US10/435,350
• Publication Number:
  US 20040236983A1
• Time in Patent Office: 1,250 Days
• Field of Search: 714/6, 714 11- 13, 707/10, 707/201
• US Class Current: 714/6.23
• CPC Class Codes: G06F 11/2058 using more than 2 mirrored copies
• CPC Class Codes: G06F 11/2082 Data synchronisation
• CPC Class Codes: G06F 2201/855 Details of asynchronous mirroring using a journal to transfer not-yet-mirrored changes
• CPC Class Codes: Y10S 707/99952 Coherency, e.g. same view to multiple users

Abstract

A method, apparatus and program storage device for maintaining data consistency and cache coherency during communications failures between nodes in a remote mirror pair. A link between a mirror pair of storage systems is monitored. During a link failure between a first storage system and a second storage systems, reads and writes on the first and second storage systems are independently performed and write data and associated timestamps are maintained for the write data for each write in a queue on the first and second storage system. After link reestablishment, volume sets on the first and second storage systems are resynchronized using write data and associated timestamps.

Claims

1. A computer readable storage medium having computer executable instructions stored thereon to perform a method for maintaining data consistency and cache coherency during communications failures between nodes in a remote mirror pair, the method comprising:

monitoring a link between a mirror pair of storage systems, the mirror pair comprising a first storage system and a second storage system;

during a link failure between the first storage system and the second storage systems, independently performing reads and writes on the first and second storage systems and maintaining write data and associated timestamps for the write data for each write in a queue of the first and second storage system; and

after link reestablishment, resynchronizing volume sets on the first and second storage systems using the write data and associated timestamps.

2. The computer readable storage medium of claim 1, wherein the resynchronizing further comprises:

sending write data and an associated timestamp for a block of data of a volume set from the first storage system to the second storage system;

comparing the timestamp for the write data for the block of data from the first storage system to a timestamp for a write for a mirrored block of data on the second storage system;

determining whether the write data from the first storage system is the latest write; and

sending an accept to the first storage system when the write data from the first storage system is the latest write and writing the write data for the block of data to the second storage system, else sending a rejection to the first storage system when the write data from the first storage system is not the latest write and eliminating the write data for the block of data on the first storage system.

3. The computer readable storage medium of claim 2 further comprising, after the first storage system has sent to the second storage system write data and an associated timestamps for all writes received by the first storage system:

sending write data and an associated timestamp for a block of data of a volume set from the second storage system to the first storage system;

comparing the timestamp for the write data for the block of data from the second storage system to a timestamp for a write for a mirrored block of data on the first storage system;

determining whether the write data from the second storage system is the latest write; and

sending from the first storage system an accept to the second storage system when the write data from the second storage system is the latest write and writing the write data for the block of data to the first storage system, else sending a rejection to the second storage system when the write data from the second storage system is not the latest write and eliminating the write data for the block of data on the second storage system.

4. The computer readable storage medium of claim 3 further comprising, after the second storage system has sent to the first storage system write data and associated timestamps for all writes received by the second storage system exiting the resynchronizing mode and returning to normal input/output processing.

5. The computer readable storage medium of claim 1 wherein the resynchronizing further comprises eliminating write data associated with a block of data on the first and second storage systems when a write received by the first storage system after reestablishment of the link overlaps write data on the first and second storage systems.

6. The computer readable storage medium of claim 1 wherein the maintaining write data and associated timestamps for the write data for each write on the first and second storage system further comprises when receiving multiple writes for a block of data on the first storage system maintaining only the last write for the block of data on the first storage system.

7. A data storage system, comprising:

a first storage subsystem, the first storage subsystem comprising an array of drives;

a second storage subsystem comprising an array of drives and having a mirrored volume set thereon of the volume set on the first storage subsystem;

a first and second controller coupled respectively to the first and second storage subsystems; and

a host, coupled to the first and second controllers, for issuing input/output requests to the first data storage system;

wherein the first and second controller are coupled by a link, the first and second controller further being configured to monitor the link coupling the first storage system and the second storage system, during a link failure between the first storage system and the second storage systems, to independently perform reads and writes on the first and second storage systems and maintain write data and associated timestamps for the write data for each write in a queue of the first and second storage system and after link reestablishment, and to resynchronize volume sets on the first and second storage systems using the write data and associated timestamps.

8. The data storage system of claim 7, wherein the first controller resynchronizes volume sets by:

sending write data and an associated timestamp for a block of data of a volume set from the first storage system to the second storage system;

comparing the timestamp for the write data for the block of data from the first storage system to a timestamp for a write for a mirrored block of data on the second storage system;

determining whether the write data from the first storage system is the latest write; and

sending an accept to the first storage system when the write data from the first storage system is the latest write and writing the write data for the block of data to the second storage system, else sending a rejection to the first storage system when the write data from the first storage system is not the latest write and eliminating the write data for the block of data on the first storage system.

9. The data storage system of claim 8 wherein the second controller, after the first storage system has sent to the second storage system write data and an associated timestamps for all writes received by the first storage system:

sends write data and an associated timestamp for a block of data of a volume set from the second storage system to the first storage system;

compares the timestamp for the write data for the block of data from the second storage system to a timestamp for a write for a mirrored block of data on the first storage system;

determines whether the write data from the second storage system is the latest write; and

sends from the first storage system an accept to the second storage system when the write data from the second storage system is the latest write and writes the write data for the block of data to the first storage system, else sends a rejection to the second storage system when the write data from the second storage system is not the latest write and eliminates the write data for the block of data on the second storage system.

10. The data storage system of claim 9 wherein the first and second controller, after the second controller has sent to the first storage system write data and associated timestamps for all writes received by the second storage system, exits the resynchronizing mode and returns to normal input/output processing.

11. The data storage system of claim 7 wherein the first and second controller resynchronize the volume sets by eliminating write data associated with a block of data when a write received by the first storage system after reestablishment of the link overlaps write data on the first storage system.

12. The data storage system of claim 7 wherein the first controller upon receiving multiple writes for a block of data maintains only the last write for the block of data.

13. A data processing apparatus, comprising:

a memory comprising executable instructions; and

a processor for executing instructions provided from the memory;

wherein the processor is configured to perform the executable instructions from the memory to monitor a link to a second storage system, during a failure of the link, to independently perform reads and writes and maintain write data and timestamps associated with the write data for each write in the memory and, after link reestablishment, to resynchronize volume sets with the second storage system using the write data and associated timestamps.

14. The data processing apparatus of claim 13, wherein the processor resynchronizes volume sets by:

sending write data and an associated timestamp for a block of data of a volume set to the second storage system; and

receiving an accept from the second storage system when the sent write data is the latest write, else receiving a rejection from the second storage system when the sent write data is not the latest write and eliminating the write data for the block of data in the memory.

15. The data processing apparatus of claim 14 wherein the processor, after sending the second storage system write data and an associated timestamps for all writes received in the queue:

receives write data and an associated timestamp for a block of data of a volume set from the second storage system;

compares the timestamp for the write data for the block of data from the second storage system to a timestamp for a write for a mirrored block of data in the memory;

determines whether the write data from the second storage system is the latest write; and

sends an accept to the second storage system when the write data from the second storage system is the latest write and writes the write data for the block of data, else sends a rejection to the second storage system when the write data from the second storage system is not the latest write.

16. The data processing apparatus of claim 15 wherein the processor, after the second storage system has sent write data and associated timestamps for all writes received by the second storage system, exits the resynchronizing mode and returns to normal input/output processing.

17. The data processing apparatus of claim 13 wherein the processor further resynchronizes volume sets by eliminating write data associated with a block of data in the memory when a write received after reestablishment of the link overlaps write data in the memory.

18. The data processing apparatus of claim 13 wherein the processor maintains comprises upon receiving multiple writes for a block of data maintains only the last write for the block of data in the memory.

19. A method for maintaining data consistency and cache coherency during communications failures between nodes in a remote mirror pair, the method comprising:

monitoring a link between a mirror pair of storage systems, the mirror pair comprising a first storage system and a second storage system;

during a link failure between the first storage system and the second storage systems, independently performing reads and writes on the first and second storage systems and maintaining write data and associated timestamps for the write data for each write in a queue of the first and second storage system; and

after link reestablishment, resynchronizing volume sets on the first and second storage systems using the write data and associated timestamps.

20. A data storage system, comprising:

a first storage means for storing a volume set thereon;

a second storage means for providing a mirrored volume set thereon of the volume set on the first storage subsystem;

means, coupled to the first and second storage means, for controlling the first and second storage means; and

host means, coupled to the means for controlling, for issuing input/output requests to the first data storage system;

wherein the means for controlling include linking means, the means for controlling further monitoring the linking means and, during a link failure between the first storage means and the second storage means, independently performing reads and writes on the first and second storage means and maintaining write data and associated timestamps for the write data for each write in a queue of the first and second storage means and after link reestablishment, and resynchronizing volume sets on the first and second storage means using the write data and associated timestamps.

21. A data processing apparatus, comprising:

means for providing memory that includes executable instructions; and

processing means for executing instructions provided from the means for providing memory;

wherein the processing means is configured to perform the executable instructions from the means for providing memory for monitoring a link to a second storage means, during a failure of the link, independently performing reads and writes and maintaining write data and timestamps associated with the write data for each write in the means for providing memory and, after link reestablishment, resynchronizing volume sets with the second storage means using the write data and associated timestamps.