System and method for re-allocating storage area network resources

• US 7,069,468 B1
• Filed: 06/28/2002
• Issued: 06/27/2006
• Est. Priority Date: 11/15/2001
• Status: Active Grant

• Current Assignee: Innovations In Memory LLC
• Original Assignee: Xiotech Corporation
• Inventors: Pittelkow, Michael Henry, Olson, Mark David
• Primary Examiner(s): Baderman, Scott
• Assistant Examiner(s): Bonura, Timothy M.
• Application Number:
  US10/183,949
• Time in Patent Office: 1,460 Days
• Field of Search: 714/4, 714/5, 714/6, 714/7, 714/9, 714/11, 714/12, 714/13
• US Class Current: 714/4.2
• CPC Class Codes: G06F 11/2092 Techniques of failing over between control units

Abstract

A fiber channel storage area network (SAN) provides virtualized storage space for a number of servers to a number of virtual disks implemented on various virtual redundant array of inexpensive disks (RAID) devices striped across a plurality of physical disk drives. The SAN includes plural controllers and communication paths to allow for fail-safe and fail-over operation. The plural controllers can be loosely-coupled to provide n-way redundancy and more than one independent channel for communicating with one another. In the event of a failure involving a controller or controller interface, the virtual disks that are accessed via the affected interfaces are re-mapped to another interface in order to continue to provide high data availability. In particular a resource re-allocation routine operating on one of the controllers employs predetermined criteria against a family of potential back-up controllers in order to select appropriate back-up controllers for a particular device that fails. The criteria, for example, can include an explicit primary/secondary table to control selecting a secondary controller for any primary controller or be more complex so as to consider such statistics as data transfer loads among the interfaces, data transfer loads among the controllers, or the number of targets assigned to each interfaces.

Claims

1. A storage area network comprising:

a plurality of front-end interfaces, each respective front-end interface providing access, to one or more associated targets, to one of a plurality of connected servers;

a resource management application configured to determine to which of the plurality of front-end interfaces to move any target associated with a first front-end interface;

wherein the plurality of front-end interfaces are distributed across a plurality of storage controllers which are loosely-coupled to provide an arrangement including a master storage controller and one or more slave storage controllers

the resource management application configured to execute on the master storage controller; and

wherein the resource management application, in determining to which of the plurality of front-end interfaces to move any target associated with a first front-end interface, is configured to base the determination on at least one of;

a controller identity of a particular storage controller with which the first front-end interface is associated, and a physical location of the particular storage controller.

2. The storage area network according to claim 1, further comprising:

a table of relationships, each relationship in the table indicating a correlation between one of the plurality of front-end interfaces and a fail-over front-end interface from among the other front-end interfaces, and

wherein the resource management application is configured so as to determine to move any target associated with the first front-end interface to the correlated fail-over front-end interface in accordance with the table of relationships.

3. The storage area network according to claim 1, further comprising:

a compilation of statistics regarding data transfers through each of the plurality of front-end interfaces, and

wherein the resource management application determines to which of the plurality of front-end interfaces to move any target associated with the first front-end interface, based on the compilation of statistics.

4. The storage area according to claim 3, wherein the resource management application is further configured to determine to which of the plurality of front-end interfaces to move any target so as to more evenly balance a data transfer load among the plurality of front-end interfaces.

5. The storage area network according to claim 1, further comprising:

storage configured to maintain an indication of a respective number of targets associated with each of the plurality of front-end interfaces; and

wherein the resource management application determines to which of the plurality of front-end interfaces to move any target associated with the front-end interface based on the indication of the number of targets associated with each of the plurality of front-end interfaces.

6. The storage area network according to claim 1, wherein all targets associated with the first front-end interface are determined to be moved to a single other one of the plurality of front-end interfaces.

7. The storage area network according to claim 1, wherein the resource management application is configured to be responsive to a failure detected with respect to the first front-end interface.

8. The storage area network according to claim 1, wherein the physical location of the particular storage controller relates to one of:

a computer rack in which the particular storage controller is located, a room in which the particular storage controller is located, a building in which the particular storage controller is located, and a geographical area in which the particular storage controller is located.

9. The storage area network according to claim 1, wherein:

each of the plurality of storage controllers maintains a respective indication of an ownership relationship between each of the plurality of front-end interfaces and each of the targets, and

the resource management application is further configured to move any target associated with the first front-end interface, wherein moving a particular target associated with the first front-end interface so as to be associated with a second front-end interface includes changing the indicated relationship, for that particular target, from the first front-end interface to the second front-end interface.

10. A method, in a storage area network comprising plural loosely-coupled redundant controllers comprising a plurality of front-end interfaces, for determining to which fail-over front-end interface to map a target associated with a first front-end interface, comprising the steps of:

identifying an operational failure that affects the first front-end interface on a first controller;

identifying one or more targets owned by the first front-end interface, wherein each target is a virtual disk presented to a server via the storage area network; and

determining, for each identified target, a respective fail-over front-end interface;

wherein the step of determining further includes the steps of;

determining information relating to the physical location of each of the plurality of front-end interfaces; and

identifying each respective fail-over interface based on at least a portion of the determined information relating to physical location.

11. The method according to claim 10, further comprising the steps of:

modifying ownership information of the identified one or more targets such that each of the one or more targets are no longer associated with the first front-end interface but are associated with their respective fail-over front-end interface; and

updating respective configuration information used by each of the redundant controllers to reflect the modified ownership information of the identified one or more targets.

12. The method according to claim 10, wherein the step of determining further includes the steps of:

analyzing operational statistics regarding the plurality of front-end interfaces; and

identifying each respective fail-over front-end interface in order to more evenly distribute a data transfer load among the plurality of front-end interfaces.

13. The method according to claim 10, wherein the step of determining further includes the step of:

storing a table which indicates for each of the plurality of front-end interfaces a respective predetermined fail-over front-end interface, and

identifying each respective fail-over front-end interface based on the table.