Granted Patent B2

US 8,566,635 · App. 13/011,195 · Granted Oct 22, 2013

Inventors: Yanling Qi (Austin, TX); Scott W. Kirvan (Austin, TX); Guy Martin (Wichita, KS); Robert R. Stankey (Wichita, KS)

Assignee: LSI Corporation

Abstract

Claims (82)

1. A system comprising:

a primary storage system, the primary storage system comprising:

a primary physical storage volume having an associated primary physical volume device identifier; and

a primary storage controller coupled with the primary physical storage volume;

a secondary storage system, the secondary storage system comprising:

a secondary physical storage volume having an associated secondary physical volume device identifier different than the primary physical volume device identifier; and

a secondary storage controller coupled with the secondary physical storage volume;

a replication communication link coupling the primary storage controller with the secondary storage controller,

wherein a logical volume is defined within both the primary storage system and the secondary storage system, the logical volume having an associated logical volume device identifier common to both the primary storage controller and the secondary storage controller,

wherein the secondary storage system is adapted to process a write I/O request directed to the logical volume device identifier from an attached host system by shipping, via the replication link, the write I/O request to the primary storage controller for processing by the primary storage system,

wherein the primary storage system is adapted to process a write I/O request directed to the logical volume device identifier received from an attached host system or received from the secondary storage system by storing write data on the primary physical storage volume and transferring the data to the secondary storage controller, via the replication link, to be stored on the secondary physical storage volume.

2. The system of claim 1

wherein the secondary storage system is further adapted to sense a failure of the primary storage system, and

wherein the secondary storage system is further adapted to reconfigure as a backup primary storage system in response to sensing the failure of the primary storage, wherein the backup primary storage system is adapted to process a received write I/O request directed to the logical volume device identifier received from an attached host system by storing write data on the secondary physical storage volume.

3. The system of claim 2

wherein the primary storage system is further adapted to transmit periodic watchdog signals to the secondary storage system via the replication link, and

wherein the secondary storage system is further adapted to sense the failure of the primary storage system by sensing loss of the periodic watchdog signal from the primary storage system.

4. The system of claim 1

the primary storage system and the secondary storage system each comprise a memory storing a pre-defined split-brain policy wherein the split-brain policy indicates which of the primary storage system and the secondary storage system should continue processing I/O requests in response to sensing a failure of the replication link,

wherein both the primary storage system and the secondary storage system are further adapted to sense a failure of the replication link, and

wherein either the primary storage system or the secondary storage system rejects further I/O requests responsive to sensing the failure of the replication link based on the pre-defined split-brain policy.

5. The system of claim 1 further comprising:

a storage management system coupled with the primary storage system and coupled with the secondary storage system,

wherein each of the storage management system, the primary storage system, and the secondary storage system are adapted to sense a failure of the replication link,

wherein both the primary storage system and the secondary storage system are further adapted to reject further I/O requests responsive to sensing the failure of the replication link,

wherein the storage management system is further adapted, responsive to sensing the failure of the replication link, to communicate with either the primary storage system or the secondary storage system to resume processing of I/O requests as a backup primary storage system.

6. A method for managing storage replication between a primary storage system and a secondary storage system coupled by a replication link, the method comprising:

defining a primary physical storage volume having an associated primary physical volume device identifier within the primary storage system;

defining a secondary physical storage volume having an associated secondary physical volume device identifier within the secondary storage system;

defining a logical volume within both the primary storage system and the secondary storage system, the logical volume having an associated single logical volume device identifier common to both the primary storage system and the secondary storage system;

processing a write I/O request received in the primary storage system from an attached host system requesting access to the logical volume wherein the step of processing further comprises:

storing data associated with the request on the primary physical storage volume; and

transmitting the data from the primary storage system to the secondary storage system over a replication link coupling the primary storage system with the secondary storage system; and

storing the data on the secondary physical storage volume;

processing another write I/O request received in the secondary storage system from an attached host system requesting access to the logical volume wherein the step of processing the another write I/O request further comprises:

shipping, over the replication link, the request to the primary storage system for processing as a request received by the primary storage system.

7. The method of claim 6 further comprising:

sensing, within the secondary storage system, a failure of the primary storage system;

reconfiguring the secondary storage system as a backup primary storage system in response to sensing failure of the primary storage system; and

processing, in the backup primary storage system, a received write I/O request directed to the logical volume device identifier received from an attached host system by storing write data on the secondary physical storage volume.

8. The method of claim 7 further comprising:

transmitting periodic watchdog signals from the primary storage system to the secondary storage system via the replication link,

wherein the step of sensing further comprises sensing, in the secondary storage system, loss of the periodic watchdog signal from the primary storage system.

9. The method of claim 6 , further comprising:

providing a split-brain policy in both the primary storage system and the secondary storage system wherein the split-brain policy indicates which of the primary storage system and the secondary storage system should continue processing I/O requests in response to sensing a failure of the replication link;

sensing, within the primary storage system and within the secondary storage system, a failure of the replication link, and

rejecting further I/O requests, within either the primary storage system or the secondary storage system, responsive to sensing the failure of the replication link based on the split-brain policy.

10. The method of claim 6 further comprising:

a storage management system coupled with the primary storage system and coupled with the secondary storage system,

sensing, within the primary storage system and the secondary storage system, a failure of the replication link;

rejecting further I/O requests, within both the primary storage system and the secondary storage system, responsive to sensing the failure of the replication link;

receiving a signal, within either the primary storage system or within the secondary storage system, from a storage management system responsive to sensing the failure of the replication link;

resuming processing of I/O requests in the primary storage system responsive to receipt of the signal within the primary storage system; and

resuming processing of I/O requests within the secondary storage system as a backup primary storage system responsive to receipt of the signal within the secondary storage system.

11. A non-transitory computer readable medium embodying programmed instructions which, when executed by a computer system of one or more storage systems, perform a method for managing storage replication between a primary storage system and a secondary storage system coupled by a replication link, the method comprising:

defining a primary physical storage volume having an associated primary physical volume device identifier within the primary storage system;

defining a secondary physical storage volume having an associated secondary physical volume device identifier within the secondary storage system;

defining a logical volume within both the primary storage system and the secondary storage system, the logical volume having an associated single logical volume device identifier common to both the primary storage system and the secondary storage system;

processing a write I/O request received in the primary storage system from an attached host system requesting access to the logical volume wherein the step of processing further comprises:

storing data associated with the request on the primary physical storage volume; and

transmitting the data from the primary storage system to the secondary storage system over a replication link coupling the primary storage system with the secondary storage system; and

storing the data on the secondary physical storage volume;

processing another write I/O request received in the secondary storage system from an attached host system requesting access to the logical volume wherein the step of processing the another write I/O request further comprises:

shipping, over the replication link, the request to the primary storage system for processing as a request received by the primary storage system.

12. The medium of claim 11 , wherein the method further comprises:

sensing, within the secondary storage system, a failure of the primary storage system;

reconfiguring the secondary storage system as a backup primary storage system in response to sensing failure of the primary storage system; and

processing, in the backup primary storage system, a received write I/O request directed to the logical volume device identifier received from an attached host system by storing write data on the secondary physical storage volume.

13. The medium of claim 12 wherein the method further comprises:

transmitting periodic watchdog signals from the primary storage system to the secondary storage system via the replication link,

wherein the step of sensing further comprises sensing, in the secondary storage system, loss of the periodic watchdog signal from the primary storage system.

14. The medium of claim 11 wherein the method further comprises:

providing a split-brain policy in both the primary storage system and the secondary storage system wherein the split-brain policy indicates which of the primary storage system and the secondary storage system should continue processing I/O requests in response to sensing a failure of the replication link;

sensing, within the primary storage system and within the secondary storage system, a failure of the replication link, and

rejecting further I/O requests, within either the primary storage system or the secondary storage system, responsive to sensing the failure of the replication link based on the split-brain policy.

15. The medium of claim 11 wherein the method further comprises:

a storage management system coupled with the primary storage system and coupled with the secondary storage system,

sensing, within the primary storage system and the secondary storage system, a failure of the replication link;

rejecting further I/O requests, within both the primary storage system and the secondary storage system, responsive to sensing the failure of the replication link;

receiving a signal, within either the primary storage system or within the secondary storage system, from a storage management system responsive to sensing the failure of the replication link;

resuming processing of I/O requests in the primary storage system responsive to receipt of the signal within the primary storage system; and

resuming processing of I/O requests within the secondary storage system as a backup primary storage system responsive to receipt of the signal within the secondary storage system.

Assignments (9)

Continuity (1)