Compensating for unavailability in a storage virtualization system

• US 7,617,252 B2
• Filed: 05/29/2006
• Issued: 11/10/2009
• Est. Priority Date: 09/07/2001
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Shah, Ashish, Ghate, Ranjit, Nguyen, Tuan, Ng, Chan, Chiang, Gen-Hwa, Ibrahim, Rahim
• Primary Examiner(s): Wong; Don
• Assistant Examiner(s): Nguyen; Kim T
• Application Number:
  US11/420,781
• Publication Number:
  US 20060206494A1
• Time in Patent Office: 1,261 Days
• Field of Search: 707/10, 707/200
• US Class Current: 1/1
• CPC Class Codes: G06F 11/2092 Techniques of failing over between control units
• CPC Class Codes: G06F 2201/815 Virtual middleware or OS functionality using virtual machines to implement generic software techniques for error detection or fault masking G06F11/1484
• CPC Class Codes: G06F 3/0617 in relation to availability
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99955 Archiving or backup

Abstract

A storage virtualization system for accessing at least one storage device for a virtual logical unit (VLUN) from at least one host. The system includes at least one upstream processing element (UPE) which is adapted for coupling to a corresponding host server. The system is further adapted for coupling to at least one storage device. The controller also includes a primary central processing element (CPE) that can be configurably coupled to the UPEs. The primary CPE permits data transfer to occur between a host and a storage device via the corresponding UPE. The controller further includes a backup CPE that can be configurably coupled to the UPEs. The backup CPE is configurable to permit the data transfer when the primary CPE becomes unavailable.

Claims

1. A method for operating a storage virtualization controller arrangement, comprising:

providing at least two central processing elements both designed with the same functionality of arbitrating access between at least one host and at least one storage device disposed within the storage virtualization controller arrangement to facilitate an in-band storage virtualization configuration;

configuring the storage virtualization controller arrangement to connect at least one host and at least one storage device through the storage virtualization controller arrangement and the at least two central processing elements wherein the storage virtualization controller arrangement is distinct from both the at least one host and the at least one storage device and uses features of the in-band storage virtualization configuration to perform both data transfer and communication over the access arbitrated between the at least one host and at least one storage device;

assigning one of the at least two central processing elements within the storage virtualization controller arrangement, distinct from both the at least one host and the at least one storage device, as a primary central processing element to arbitrate access from the at least one host to the at least one storage device; and

upon detection of an unavailability of the primary central processing element within the storage virtualization controller arrangement yet distinct from both the at least one host and the at least one storage device, automatically reassigning at least one of the at least two different one of the central processing elements as a backup central processing element within the storage virtualization controller to arbitrate the access from the at least one host to the at least one storage device using features of the in-band storage virtualization configuration.

2. The method of claim 1, wherein the controller arrangement is a single controller.

3. The method of claim 1, wherein the configuring includes:

connecting individual ones of the at least one host and the at least one storage device to the storage virtualization system via a connection method taken from a group consisting of a direct point-to-point connection and a switching network connection.

4. The method of claim 1, wherein the unavailability results from a failure of the primary central processing element.

5. The method of claim 1, wherein the storage virtualization controller arrangement includes an upstream processing element adapted to connect to the host, the assigning further including:

establishing a communication link between the primary central processing element and the upstream processing element.

6. The method of claim 5, wherein the controller arrangement includes at least two controllers, wherein the upstream processing element is disposed in a first controller, and wherein at least one of the primary central processing element and the backup central processing element is disposed in a second controller.

7. The method of claim 6, wherein the unavailability results from an event selected from the group consisting of a failure condition of the controller in which the primary central processing element is disposed, and an elective shutdown of the controller in which the primary central processing element is disposed.

8. The method of claim 6, further comprising:

communicatively coupling tire first controller arid the second controller to at least one same one of the at least one host; and

communicatively coupling the first controller and the second controller to at least one same one of the at least one storage device, so as to provide two separate paths from the at least one same one of the at least one host to the at least one same one of the at least one storage device.

9. The method of claim 5, wherein the reassigning further comprises:

the primary central processing element rejecting a request from the upstream processing element for permission to perform an I/O command.

10. The method of claim 1, wherein the primary central processing element includes a pending queue of I/O commands, the reassigning further comprising:

the primary central processing element executing the pending queue of I/O commands.

11. The method of claim 5, wherein the reassigning further comprises:

terminating the communication link between the primary central processing element and the upstream processing element; and

establishing a communication link between the backup central processing element and the upstream processing element.

12. The method of claim 1, wherein the reassigning further comprises:

predetermining the at least one different one of the central processing elements before the unavailability occurs.

13. The method of claim 1, wherein the reassigning further comprises:

determining the at least one different one of the central processing elements after the unavailability occurs.

14. The method of claim 13, wherein the determining includes:

balancing a workload among the available ones of the at least two central processing elements in the controller arrangement.

15. A storage virtualization controller arrangement, comprising:

means for providing at least two central processing elements both designed with the same functionality of arbitrating access between at least one host and at least one storage device disposed within the storage virtualization controller arrangement to facilitate an in-band storage virtualization configuration;

means for configuring the storage virtualization controller arrangement to connect at least one host and at least one storage device through the storage virtualization controller arrangement and the at least two central processing elements wherein the storage virtualization controller arrangement is distinct from both the at least one host and the at least one storage device and uses features of the in-band storage virtualization configuration to perform both data transfer and communication over the access arbitrated between the at least one host and at least one storage device;

means for assigning one of the at least two central processing elements within the storage virtualization controller arrangement and distinct from both the at least one host and the at least one storage device as a primary central processing element to arbitrate access from the at least one host to the at least one storage device; and

means for automatically reassigning at least one of the at least two different one of the central processing elements as a backup central processing element within the storage virtualization controller to arbitrate the access from the at least one host to the at least one storage device using features of the in-band storage virtualization configuration upon detection of an unavailability of the primary central processing element.

16. A computer program product stored on a computer readable storage medium encoded with instructions for operating a storage virtualization controller arrangement, comprising instructions in the computer program product operable to cause a programmable processor to:

provide at least two central processing elements both designed with the same functionality of arbitrating access between at least one host and at least one storage device disposed within the storage virtualization controller arrangement for an in-band storage virtualization configuration;

configure the storage virtualization controller arrangement to connect at least one host and at least one storage device through the storage virtualization controller arrangement and the at least two central processing elements wherein the storage virtualization controller arrangement is distinct from both the at least one host and the at least one storage device and uses features of the in-band storage virtualization configuration to perform both data transfer and communication over the access arbitrated between the at least one host and at least one storage device;

assign one of the at least two central processing elements within the storage virtualization controller arrangement and distinct from both the at least one host and the at least one storage device as a primary central processing element to arbitrate access from the at least one host to the at least one storage device; and

upon detection of an unavailability of the primary central processing element within the storage virtualization controller arrangement yet distinct from both the at least one host and the at least one storage device, automatically reassign at least one of the at least two different one of the central processing elements as a backup central processing element within the storage virtualization controller to arbitrate the access from the at least one host to the at least one storage device using features of the in-band storage virtualization configuration.

17. The computer program product of claim 16. wherein the controller arrangement is a single controller.

18. The computer program product of claim 16, wherein the configuring further includes instructions that:

connect individual ones of the at least one host and the at least one storage device to the storage virtualization system via a connection method taken from a group consisting of a direct point-to-point connection and a switching network connection.

19. The computer program product of claim 16, wherein the unavailability results from a failure of the primary central processing element.

20. the computer program product of claim 16, wherein the storage virtualization controller arrangement includes an upstream processing element adapted to connect to the host and the instructions that assign further includes instructions that:

establish a communication link between the primary central processing element and the upstream processing element.