Virtualized capacity management

1. A method for managing virtual resource capacity, comprising:

determining, for a given workload comprising two or more virtual resources from a single source, historical demand data for the two or more virtual resources by aggregating capacity consumption metrics across the two or more virtual resources as a measure of consumed infrastructure units as a function of time, the infrastructure units comprising predefined groupings of physical computational resources representing a common measure of disparate computational resources of the physical infrastructure on which the two or more virtual resources run;

generating a projection of aggregate future workload demand for the two or more virtual resources of the given workload as a function of the infrastructure units based at least in part on the historical demand data for the two or more virtual resources; and

effecting at least one configuration change for at least one of the two or more virtual resources in accordance with the projection of the aggregate future workload demand for the given workload;

wherein the method is performed using at least one processing device comprising a processor coupled to a memory.

2. The method of claim 1 , further comprising monitoring the two or more virtual resources to obtain the historical demand data.

3. The method of claim 2 , wherein the monitoring comprises gathering at least one performance metric for at least one of the two or more virtual resources and ownership information describing the given workload to which the at least one virtual resource belongs.

4. The method of claim 2 , wherein the monitoring comprises generating normalized virtual resource demand information for the two or more virtual resources by scaling capacity consumption measurements of the two or more virtual resources into the infrastructure units based at least in part on the predefined groupings of physical computational resources.

5. The method of claim 1 , wherein the generating comprises filtering the historical demand data by sorting the historical demand data in terms of numbers of infrastructure units utilized at each of a plurality of time codes and removing outlying values.

6. The method of claim 1 , wherein the generating comprises generating a regression model describing the aggregate future workload demand as a function of time.

7. The method of claim 6 , wherein the generating further comprises:

obtaining a modeling methodology from a user or selecting the modeling methodology automatically; wherein

the regression model is generated according to the modeling methodology.

8. The method of claim 6 , wherein the generating further comprises:

generating a prediction of an upper bound of the aggregate future workload demand using the regression model; and

generating a prediction of a lower bound of the aggregate future workload demand using the regression model.

9. The method of claim 8 , wherein the generating further comprises:

obtaining a forecast confidence from a user or selecting the forecast confidence automatically; wherein

the prediction of the upper bound and the prediction of the lower bound are predicted at the forecast confidence.

10. The method of claim 8 , wherein the generating further comprises:

obtaining a projection length from a user or selecting the projection length automatically; wherein

the projection length defines a time for which the prediction of the upper bound and the prediction of the lower bound are generated.

11. The method of claim 1 , wherein the generating comprises obtaining a selection of the given workload from a user or selecting the given workload automatically.

12. The method of claim 1 , wherein effecting the least one configuration change comprises:

displaying, via a user interface, the projection; and

receiving, via the user interface, a command defining the at least one configuration change.

13. The method of claim 12 , wherein effecting the least one configuration change further comprises:

determining whether the command defines a configuration change that falls outside an allowable capacity bound;

in response to determining that the configuration change falls outside the allowable capacity bound, notifying, via the user interface, the user; and

in response to determining that the configuration change does not fall outside the allowable capacity bound, effecting the defined configuration change.

14. The method of claim 1 , wherein effecting the least one configuration change comprises:

automatically evaluating the projection; and

selecting the at least one configuration change based on the evaluating.

15. A system for managing virtual resource capacity, comprising:

at least one processing device comprising a processor coupled to a memory;

the at least one processing device being configured:

to determine, for a given workload comprising two or more virtual resources from a single source, historical demand data for the two or more virtual resources by aggregating capacity consumption metrics across the two or more virtual resources as a measure of consumed infrastructure units as a function of time, the infrastructure units comprising predefined groupings of physical computational resources representing a common measure of disparate computational resources of the physical infrastructure on which the two or more virtual resources run;

to generate a projection of aggregate future workload demand for the two or more virtual resources of the given workload as a function of the infrastructure units based at least in part on the historical demand data for the two or more virtual resources; and

to effect at least one configuration change for at least one of the two or more virtual resources in accordance with the projection of the aggregate future workload demand for the given workload.

16. The system of claim 15 , wherein the at least one processing device is further configured to monitor the two or more virtual resources to obtain the historical demand data.

17. The system of claim 16 , wherein the monitoring comprises gathering at least one performance metric for at least one of the two or more virtual resources and ownership information describing the given workload to which the at least one virtual resource belongs.

18. The system of claim 16 , wherein the monitoring comprises generating normalized virtual resource demand information for the two or more virtual resources by scaling capacity consumption measurements of the two or more virtual resources into the infrastructure units based at least in part on the predefined groupings of physical computational resources.

19. The system of claim 15 , wherein the generating comprises filtering the historical demand data by sorting the historical demand data in terms of numbers of infrastructure units utilized at each of a plurality of time codes and removing outlying values.

20. The system of claim 15 , wherein the generating comprises generating a regression model describing the aggregate future workload demand as a function of time.

21. The system of claim 20 , wherein the generating further comprises:

obtaining a modeling methodology from a user or selecting the modeling methodology automatically; wherein

the regression model is generated according to the modeling methodology.

22. The system of claim 20 , wherein the generating further comprises:

generating a prediction of an upper bound of the aggregate future workload demand using the regression model; and

generating a prediction of a lower bound of the aggregate future workload demand using the regression model.

23. The system of claim 22 , wherein the generating further comprises:

obtaining a forecast confidence from a user or selecting the forecast confidence automatically; wherein

the prediction of the upper bound and the prediction of the lower bound are predicted at the forecast confidence.

24. The system of claim 22 , wherein the generating further comprises:

obtaining a projection length from a user or selecting the projection length automatically; wherein

the projection length defines a time for which the prediction of the upper bound and the prediction of the lower bound are generated.

25. The system of claim 15 , wherein the generating comprises obtaining a selection of the given workload from a user or selecting the given workload automatically.

26. The system of claim 15 , wherein effecting the least one configuration change comprises:

displaying, via a user interface, the projection; and

receiving, via the user interface, a command defining the at least one configuration change.

27. The system of claim 26 , wherein effecting the least one configuration change further comprises:

determining whether the command defines a configuration change that falls outside an allowable capacity bound;

in response to determining that the configuration change falls outside the allowable capacity bound, notifying, via the user interface, the user; and

in response to determining that the configuration change does not fall outside the allowable capacity bound, effecting the defined configuration change.

28. The system of claim 15 , wherein effecting the least one configuration change comprises:

automatically evaluating the projection; and

selecting the at least one configuration change based on the evaluating.

29. The method of claim 6 , wherein the regression model comprises a seasonal integrated autoregressive moving average regression model.

30. The method of claim 6 , wherein the regression model decomposes the historical demand data into a trend component, a seasonal component and a high-frequency component, the high-frequency component being fit with a stationary integrated autoregressive moving average model that is summed with the trend component and the seasonal component to produce an additive model of the historical demand.

31. The method of claim 6 , wherein the regression model comprises a sum of a Fourier series representing a seasonal component of the historical demand and a stationary integrated autoregressive moving average model.

32. A computer program product comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device:

to determine, for a given workload comprising two or more virtual resources from a single source, historical demand data for the two or more virtual resources by aggregating capacity consumption metrics across the two or more virtual resources as a measure of consumed infrastructure units as a function of time, the infrastructure units comprising predefined groupings of physical computational resources representing a common measure of disparate computational resources of the physical infrastructure on which the two or more virtual resources run;

to generate a projection of aggregate future workload demand for the two or more virtual resources of the given workload as a function of the infrastructure units based at least in part on the historical demand data for the two or more virtual resources; and

to effect at least one configuration change for at least one of the two or more virtual resources in accordance with the projection of the aggregate future workload demand for the given workload.