APPARATUS, SYSTEM, AND METHOD FOR SELECTING A CLUSTER

• US 20090006734A1
• Filed: 06/28/2007
• Published: 01/01/2009
• Est. Priority Date: 06/28/2007
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Reid, Mark A., Niranjan, Thirumale N., Swingler, Joseph M., Peake, Jonathan W., Kishi, Gregory T.
• Granted Patent: US 7,774,094 B2
• Time in Patent Office: Days
• US Class Current: 711/111
• CPC Class Codes: G06F 12/0873 Mapping of cache memory to specific storage devices or parts thereof
• CPC Class Codes: G06F 2212/213 Tape storage
• CPC Class Codes: G06F 2212/224 Disk storage
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/0686 Libraries, e.g. tape libraries, jukebox

Abstract

An apparatus, system, and method are disclosed for selecting a source cluster in a distributed storage configuration. A measurement module measures system factors for a plurality of clusters over a plurality of instances. The clusters are in communication over a network and each cluster comprises at least one tape volume cache. A smoothing module applies a smoothing function to the system factors, wherein recent instances have higher weights. A lifespan module calculates a mount-to-dismount lifespan for each cluster from the smoothed system factors. A selection module selects a source cluster for accessing an instance of a specified volume in response to the mount-to-dismount lifespans and a user policy.

Claims

1. A computer program product comprising a computer useable medium having a computer readable program, wherein the computer readable program when executed on a computer causes the computer to:

measure system factors for a plurality of clusters over a plurality of instances, wherein the clusters are in communication over a network and each cluster comprises at least one tape volume cache;

apply a smoothing function to the system factors, wherein recent instances have higher weights;

calculate a mount-to-dismount lifespan for each cluster from the smoothed system factors; and

select a source cluster for accessing an instance of a specified volume in response to the mount-to-dismount lifespans and a user policy.

2. The computer program product of claim 1, wherein the computer readable program further causes the computer to select a consistent source volume for duplication operations.

3. The computer program product of claim 1, wherein the system factors comprise factors for the at least one tape volume cache, factors for at least one tape drive, and factors for the network.

4. The computer program product of claim 3, wherein the system factors are selected from a recall time, a free space of the tape volume cache, pre-migrated space available, throttling values, a write throughput, a read throughput, a write response time, and a read response time.

5. The computer program product of claim 4, wherein the mount-to-dismount lifespan for each cluster is calculated as l=a₀t_r+a₁e_s+a₂p_s+a₃v_t+a₄w_t+a₅r_t+a₆w_r+a₇r_r where each a is a specified constant, t_r is the recall time, e_s is the free space of the tape volume cache, p_s is the pre-migrated space available, v_t is a throttling value, w_t is the write throughput, r_t is the read throughput, w_r is the write response, and r_r is the read response.

6. The computer program product of claim 1, wherein the smoothing function is configured as an exponential smoothing function sf=k₀s₀ⁿ+k₁s₁^n-1+k₂s₂₁^n-2 . . . +k_ns_n where each k is a constant for a specified instance, s is a system factor for a specified instance, and n is a number of instances.

7. The computer program product of claim 1, wherein the user policy supersedes the mount-to-dismount lifespans in selecting the cluster.

8. The computer program product of claim 7, wherein the user policy requires that the specified volume be accessed from a local cluster.

9. The computer program product of claim 8, wherein the computer readable program further causes the computer to synchronize an instance of the specified volume on the local cluster if the local cluster specified volume instance is not consistent.

10. The computer program product of claim 7, wherein the user policy favors that the specified volume be accessed from a local cluster.

11. The computer program product of claim 7, wherein the user policy requires selection of a specified set of immediate copy clusters over deferred consistency clusters.

12. The computer program product of claim 7, wherein the user policy favors selection of a specified set of immediate copy clusters over deferred consistency clusters.

13. The computer program product of claim 7, wherein the user policy requires that a specified cluster be chosen if the specified volume is a scratch volume.

14. The computer program product of claim 7, wherein the user policy favors that a specified cluster be chosen if the specified volume is a scratch volume.

15. The computer program product of claim 7, wherein the user policy requires that a specified cluster be chosen if the specified volume is not a scratch volume.

16. The computer program product of claim 7, wherein the user policy favors that a specified cluster be chosen if the specified volume is not a scratch volume.

17. The computer program product of claim 1, wherein the computer readable program further causes the computer to:

filter out unavailable clusters;

filter out clusters that are anticipated to be unavailable;

filter out clusters with unavailable library managers; and

filter out clusters with an inhibit copy mode.

18. The computer program product of claim 1, wherein the user policy does not favor clusters with degraded library managers.

19. The computer program product of claim 1, wherein the computer readable program further causes the computer to filter out clusters with inconsistent instances of the specified volume when the specified volume is not scratch.

20. The computer program product of claim 1, wherein the computer readable program further causes the computer to access the selected volume by creating a scratch volume.

21. The computer program product of claim 1, wherein the computer readable program further causes the computer to copy the specified volume to a first cluster if a second cluster with an only consistent instance of the specified volume is anticipated to become unavailable.

22. The computer program product of claim 1, wherein the user policy favors a first cluster with the specified volume in the tape volume cache over a second cluster where the specified volume is migrated.

23. An apparatus to select a cluster, the apparatus comprising:

a measurement module comprising a computer readable program executing on a processor and configured to measure system factors for a plurality of clusters over a plurality of instances, wherein the clusters are in communication over a network and each cluster comprises at least one tape volume cache;

a smoothing module comprising a computer readable program executing on the processor and configured to apply a smoothing function to the system factors, wherein recent instances have higher weights;

a lifespan module comprising a computer readable program executing on the processor and configured to calculate a mount-to-dismount lifespan for each cluster from the smoothed system factors; and

a selection module comprising a computer readable program executing on the processor and configured to select a source cluster for accessing an instance of a specified volume in response to the mount-to-dismount lifespans and a user policy.

24. The apparatus of claims 23, wherein the system factors comprise factors for the at least one tape volume cache, factors for at least one tape drive, and factors for the network.

25. The apparatus of claim 24, wherein the system factors are selected from a recall time, a free space of the tape volume cache, pre-migrated space available, throttling values, a write throughput, a read throughput, a write response time, and a read response time.

26. The apparatus of claim 25, wherein the mount-to-dismount lifespan for each cluster is calculated as l=a₀t_r+a₁e_s+a₂p_s+a₃v_t+a₄w_t+a₅r_t+a₆w_r+a₇r_r where each a is a specified constant, t_r is the recall time, e_s is the free space of the tape volume cache, p_s is the pre-migrated space available, v_t is a throttling value, w_t is the write throughput, r_t is the read throughput, w_r is the write response, and r_r is the read response.

27. A system to select a cluster, the system comprising:

a network;

a plurality of sites in communication over the network, each site comprising at least one host and a storage system comprising a plurality of clusters, each cluster comprising at least one tape drive configured to access volumes stored on magnetic tape, at least one tape volume cache, and a cluster manager configured to execute computer readable programs using a processor and a memory, wherein the software readable programs comprise

a measurement module configured to measure system factors for the plurality of clusters over a plurality of instances;

a smoothing module configured to apply a smoothing function to the system factors, wherein recent instances have higher weights;

a communication module configured to communicate system factors between the measurement modules of the clusters;

a lifespan module configured to calculate a mount-to-dismount lifespan for each cluster from the smoothed system factors; and

a selection module configured to select a source cluster for accessing an instance of a specified volume in response to the mount-to-dismount lifespans and a user policy.

28. The system of claim 27, wherein each cluster includes at least one library manager that embodies the at least one tape drive.

29. A method for deploying computer infrastructure, comprising integrating computer readable program into a computing system, wherein the program in combination with the computing system is capable of performing the following:

measuring system factors for a plurality of clusters over a plurality of instances, wherein the clusters are in communication over a network and each cluster comprises at least one tape drive configured to access volumes stored on magnetic tape and at least one tape volume cache;

applying a smoothing function to the system factors, wherein recent instances have higher weights;

calculating a mount-to-dismount lifespan for each cluster from the smoothed system factors; and

selecting a source cluster for accessing an instance of a specified volume in response to the mount-to-dismount lifespans and a user policy.

30. The method of claim 29, further comprising selecting a consistent source volume for duplication operations.

31. The method of claim 29, wherein the system factors comprise factors for the at least one tape volume cache, factors for the at least one tape drive, and factors for the network.

32. The method of claim 31, wherein the system factors are selected from a recall time, a free space of the tape volume cache, pre-migrated space available, throttling values, a write throughput, a read throughput, a write response time, and a read response time.

33. The method of claim 32, wherein the mount-to-dismount lifespan for each cluster is calculated as l=a₀t_r+a₁e_s+a₂p_s+a₃v_t+a₄w_t+a₅r_t+a₆w_r+a₇r_r where each a is a specified constant, t_r is the recall time, e_s is the free space of the tape volume cache, p_s is the pre-migrated space available, v_t is a throttling value, w_t is the write throughput, r_t is the read throughput, w_r is the write response, and r_r is the read response.

34. The method of claim 33, wherein the smoothing function is configured as an exponential smoothing function sf=k₀s₀ⁿ+k₁s₁^n-1+k₂s₂₁^n-2 . . . +k_ns_n where each k is a constant for a specified instance, s is a system factor for a specified instance, and n is a number of instances.

35. An apparatus to select a cluster, the apparatus comprising:

means for measuring system factors for a plurality of clusters over a plurality of instances, wherein the clusters are in communication over a network and each cluster comprises at least one tape drive configured to access volumes stored on magnetic tape and at least one tape volume cache and wherein the system factors are selected from a recall time, a free space of the tape volume cache, pre-migrated space available, throttling values, a write throughput, a read throughput, a write response time, and a read response time;

means for applying an exponential smoothing equation sf=k₀s₀ⁿ+k₁s₁^n-1+k₂ s₂₁^n-2 . . . +k_ns_n to the system factors, where each k is a constant for a specified instance, s is a system factor for a specified instance, and n is a number of instances;

means for calculating a mount-to-dismount lifespan for each cluster from the smoothed system factors; and

means for selecting a source cluster for accessing an instance of a specified volume in response to the mount-to-dismount lifespans and a user policy.