Estimating power consumption in a computing environment

1. A computer implemented method for determining power consumption in a data storage system, the method comprising:

processing data access requests submitted by way of a host system operating in a computing environment, wherein the data access requests are directed to at least a first storage device operating in a virtualized storage system in said computing environment;

determining disk level data access operations for the first storage device based on processing the data access requests, wherein due to virtualization of the storage system, the first storage device is unable to independently provide information that identify individual disk level data access operations performed by the first storage device, in response to receiving the data access requests submitted by way of the host system; and

calculating power consumption for the storage system by interpolating costs associated with each of the operations performed by the first storage device, wherein the cost associated with each of said operations is determined based on:

(1) various levels of activities for the first storage device and a mix of workload characteristics, and

(2) predetermined power consumption measurements obtained from one or more benchmarks for same operations performed by a second storage device in a test environment, wherein the second storage device simulates the first storage device in a non-virtualized storage environment such that power consumption measurements for a series of operations performed on the second storage device are accurately obtained by way of test equipment directly connected to the second storage device.

2. The method of claim 1 , wherein the second storage device is same as the first storage device.

3. The method of claim 1 , wherein various levels of activities are associated with I/O rate for the first storage device.

4. The method of claim 1 , wherein the workload characteristics are associated with levels with which the first storage device is utilized.

5. The method of claim 1 , wherein the power consumption measurements are stored in a data structure so that the power consumption of the storage system can be determined based on interpolation of values stored in the data structure.

6. The method of claim 5 , wherein the data structure comprises values for power consumed by each of the operations.

7. The method of claim 1 , further comprising reporting power consumed by a storage device in the storage system.

8. The method of claim 7 , wherein the storage system may comprise at least one of a disk drive, a disk RAID, a controller, just a bunch of disks (JBOD), a solid state disk (SSD) or a combination thereof.

9. The method of claim 1 , wherein various levels of activities for the first storage device and the workload characteristics are determined based on at least one of sequential read I/O rate, sequential write I/O rate, random read I/O rate, random write I/O rate, read and write bytes per transfer, read hit probability, average seek distance per action, stripe and logical track sizes, read/write response time, read/write queue size or a combination thereof.

10. A system for determining power consumption in a data storage system, the system comprising:

one or more processors for processing data access requests submitted by way of a host system operating in a computing environment, wherein the data access requests are directed to at least a first storage device operating in a virtualized storage system in said computing environment;

a logic unit for determining disk level data access operations for the first storage device based on processing the data access requests, wherein due to virtualization the first storage device is unable to independently provide information that identify individual disk level data access operations performed by the first storage device, in response to receiving the data access requests submitted by way of the host system; and

a logic unit for calculating power consumption for the storage system by interpolating costs associated with each of the operations performed by the first storage device, wherein the cost associated with each of said operations is determined based on:

(1) various levels of activities for the first storage device and a mix of workload characteristics, and

(2) predetermined power consumption measurements obtained from one or more benchmarks for same operations performed by a second storage device in a test environment, wherein the second storage device simulates the first storage device in a non-virtualized storage environment such that power consumption measurements for a series of operations performed on the second storage device are accurately obtained by way of test equipment directly connected to the second storage device.

11. The system of claim 10 , wherein the second storage device is same as the first storage device.

12. The system of claim 10 , wherein various levels of activities are associated with I/O rate for the first storage device.

13. The system of claim 10 , wherein the workload characteristics are associated with levels with which the first storage device is utilized.

14. The system of claim 10 , wherein the power consumption measurements are stored in a data structure so that the power consumption of the storage system can be determined based on interpolation of values stored in the data structure.

15. The system of claim 14 , wherein the data structure comprises values for power consumed by each of the operations.

16. The system of claim 10 , further comprising a logic unit for reporting power consumed by a storage device in the storage system.

17. The system of claim 16 , wherein the storage system may comprise at least one of a disk drive, a disk RAID, a controller, just a bunch of disks (JBOD), a solid state disk (SSD) or a combination thereof.

18. The system of claim 10 , wherein various levels of activities for the first storage device and the workload characteristics are determined based on at least one of sequential read I/O rate, sequential write I/O rate, random read I/O rate, random write I/O rate, read and write bytes per transfer, read hit probability, average seek distance per action, stripe and logical track sizes, read/write response time, read/write queue size or a combination thereof.

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

process data access requests submitted by way of a host system operating in a computing environment, wherein the data access requests are directed to at least a first storage device operating in a virtualized storage system in said computing environment;

determine disk level data access operations for the first storage device based on processing the data access requests, wherein due to virtualization the first storage device is unable to independently provide information that identify individual disk level data access operations performed by the first storage device, in response to receiving the data access requests submitted by way of the host system; and

calculate power consumption for the storage system by interpolating costs associated with each of the operations performed by the first storage device, wherein the cost associated with each of said operations is determined based on:

(1) various levels of activities for the first storage device and a mix of workload characteristics, and

(2) predetermined power consumption measurements obtained from one or more benchmarks for same operations performed by a second storage device in a test environment, wherein the second storage device simulates the first storage device in a non-virtualized storage environment such that power consumption measurements for a series of operations performed on the second storage device are accurately obtained by way of test equipment directly connected to the second storage device.

20. The computer program product of claim 19 , wherein various levels of activities for the first storage device and the workload characteristics are determined based on at least one of sequential read I/O rate, sequential write I/O rate, random read I/O rate, random write I/O rate, read and write bytes per transfer, read hit probability, average seek distance per action, stripe and logical track sizes, read/write response time, read/write queue size or a combination thereof.