Method and system for grouping clients of a storage area network according to priorities for bandwidth allocation

1. A method for managing bandwidth allocation between groups of servers in a storage area network (SAN), the method comprising:

preparing a policy that defines access parameters for two or more groups of servers that are grouped according to priority in accessing network resources; and

allocating bandwidth resources to the groups of servers so a first group of servers with a first priority preferentially receives bandwidth for processing Input/Output (I/O) requests in the SAN as compared to a second group of servers with a second priority that is lower than the first priority.

2. The method of claim 1 , further comprising operating a resource manager in accordance with said policy, wherein said resource manager controls access to said network resources.

3. The method of claim 1 , further comprising:

receiving a plurality of Input/Output (I/O) requests from a plurality of servers;

determining a priority group to which each of said servers belong; and

dynamically allocating bandwidth resources to said plurality of servers based on the priority group to which said plurality of servers belongs.

4. The method of claim 1 , wherein preparing a policy comprises:

setting an upper limit on an amount of bandwidth allocated to a group of said servers; and

setting a lower threshold for an amount of bandwidth to be used by said group of servers.

5. The method of claim 4 , further comprising applying said limit and threshold at ports of a storage device of said storage area network.

6. The method of claim 4 , further comprising applying said limit and threshold at a logical unit of a storage device of said storage area network.

7. The method of claim 4 , wherein, if I/O requests for a group of servers drop below the corresponding lower threshold, the upper limit for another group of servers is released.

8. The method of claim 1 , wherein said preparing a policy comprises receiving user input to establish said groups and access parameters.

9. The method of claim 1 , further comprising assigning servers to said groups based on I/O performance requirements for said servers.

10. The method of claim 9 , wherein each server in one of said groups receives a same bandwidth allocation setting as other servers in that group.

11. The method of claim 1 , wherein said preparing said policy further comprises varying said access parameters with time.

12. Computer-readable instructions stored on a medium for storing computer-readable instructions, said instructions, when executed by a resource manager in a storage area network (SAN), causing said resource manager to allocate network bandwidth in the SAN according to a policy that defines access parameters for two or more groups of host computers that are grouped according to priority in accessing network resources so one group of host computers with a higher priority preferentially receives bandwidth in the SAN as compared to another group of host computers with a lower priority.

13. The instructions of claim 12 , wherein said instructions further cause said resource manager to:

receive a plurality of Input/Output (I/O) requests from a plurality of host computers;

determine a priority group to which each of said host computers belong; and

dynamically allocate bandwidth resources to said host computer based on the priority group to which that host computer belongs.

14. The instructions of claim 12 , wherein said instructions further cause said resource manager to, in accordance with said policy:

set an upper limit on an amount of bandwidth allocated to a group of said host computers; and

set a lower threshold for an amount of bandwidth to be used by said group of host computers.

15. The instructions of claim 14 , wherein said instructions further cause said resource manager to apply said limit and threshold at ports of a storage device of said storage area network.

16. The instructions of claim 14 , wherein said instructions further cause said resource manager to apply said limit and threshold at a logical unit of a storage device of said storage area network.

17. The instructions of claim 14 , wherein said instructions further cause said resource manager to release a corresponding upper limit for one group of host computers if I/O requests from another group of host computers drop below the corresponding lower threshold.

18. The instructions of claim 11 , wherein said instructions further cause said resource manager to vary allocations of network bandwidth to said groups of host computers as a function of time in accordance with said policy.

19. A system for managing bandwidth allocation in a storage area network (SAN), the system comprising:

means for preparing a policy that defines access parameters for two or more groups of servers that are grouped according to priority in accessing network resources; and

means for implementing said policy so a first group of servers with a first priority preferentially receives bandwidth for processing Input/Output (I/O) requests in the SAN as compared to a second group of servers with a second priority that is lower than the first priority.

20. The system of claim 19 , wherein said means for implementing said policy comprise a resource manager for controlling access to said network resources in response to said policy.

21. The system of claim 19 , wherein said means for implementing said policy comprise:

means for receiving a plurality of Input/Output (I/O) requests from a plurality of servers;

means for determining a priority group to which each of said servers belong; and

means for dynamically allocating bandwidth resources to said server based on the priority group to which that server belongs.

22. The system of claim 19 , wherein said means for implementing said policy comprise:

means for setting an upper limit on an amount of bandwidth allocated to a group of said servers in accordance with said policy; and

means for setting a lower threshold for an amount of bandwidth to be used by said group of servers in accordance with said policy.

23. The system of claim 22 , further comprising means for applying said limit and threshold at ports of a storage device of said storage area network.

24. The system of claim 22 , further comprising means for applying said limit and threshold at a logical unit of a storage device of said storage area network.

25. The system of claim 22 , further comprising means for releasing the corresponding upper limit for a group of servers if I/O requests from another group of servers drop below the corresponding lower threshold.

26. The system of claim 19 , further comprising user input means for receiving user input to establish said policy.

27. The system of claim 19 , further comprising means for assigning servers to said groups based on I/O performance requirements for said servers.

28. The system of claim 19 , wherein said means for implementing said policy further comprise means for varying said access parameters as a function of time in accordance with said policy.

29. A storage area network (SAN) comprising:

a plurality of servers divided into priority groups;

at least one storage device networked to said plurality of servers; and

a resource manager for allocating access to at least one storage device by said servers in accordance with a policy that defines access parameters for each of said priority groups so a first group of servers with a first priority preferentially receives bandwidth for processing Input/Output (I/O) requests in the SAN as compared to a second group of servers with a second priority that is lower than the first priority.

30. The network of claim 29 , wherein said policy further comprises:

an upper limit on an amount of bandwidth allocated to a group of said servers; and

a lower threshold for an amount of bandwidth to be used by said group of servers.

31. The network of claim 30 , wherein said resource manager applies said limit and threshold at ports of a storage device of said storage area network.

32. The network of claim 30 , wherein said resource manager applies said limit and threshold at a logical unit of a storage device of said storage area network.

33. The network of claim 30 , wherein said resource manager releases a corresponding upper limit for a group of servers if Input/Output requests from another group of servers drop below the corresponding lower threshold.

34. The network of claim 29 , further comprising a user interface through which a user can establish or update said policy.

35. The network of claim 29 , wherein said resource manager varies allocations of access to at least one storage device for said groups of servers as a function of time in accordance with said policy.

36. A resource manager for a storage device in a storage area network (SAN), said resource manager comprising:

a policy that defines access parameters for two or more priority groups of host computers in the storage area network; and

computer readable instructions, stored in a computer readable medium accessible to said resource manager, for execution by said resource manager, that allocates access to a storage device of the storage area network by the host computers in accordance with said access parameters of said policy so one group of host computers with a higher priority preferentially receives bandwidth in the SAN as compared to another group of host computers with a lower priority.