SYSTEM FOR ASSEMBLING BEHAVIOR MODELS OF TECHNOLOGY COMPONENTS

• US 20090204382A1
• Filed: 04/01/2009
• Published: 08/13/2009
• Est. Priority Date: 02/12/2008
• Status: Active Grant

• Current Assignee: Accenture Global Services Limited (Accenture PLC)
• Original Assignee: Accenture Global Services GmbH (Accenture PLC)
• Inventors: Inestroza, Fausto A., Guevarra, Nikki Ann G., Tung, Teresa S.
• Granted Patent: US 8,438,125 B2
• Time in Patent Office: Days
• US Class Current: 703/21
• CPC Class Codes: G06F 1/3203 Power management, i.e. event-based initiation of a power-saving mode

Abstract

A system is described for assembling a behavior model of a technology component. The system may include a memory, and a processor. The memory may store a behavior model of a technology component. The processor may to monitor the technology component over a period of time to determine a plurality of parameter values associated with each state of a plurality of states of the technology component. The processor may process the plurality of parameter values associated with each state to determine a baseline parameter value for each state. The processor may assemble the behavior model for the technology component. The behavior model may include each state, the baseline parameter value for each state, and an association between each state and the baseline parameter value for each state. The processor may store the behavior model of the technology component in the memory.

Claims

1. A computer-implemented method for assembling a behavior model of a technology component, the method comprising:

monitoring, by the processor, a technology component over a period of time to determine a plurality of parameter values associated with each state of a plurality of states of the technology component;

processing, by the processor, the plurality of parameter values associated with each state in the plurality of states to determine a baseline parameter value for each state in the plurality of states;

assembling, by the processor, a behavior model for the technology component, wherein the behavior model comprises each state, the baseline parameter value for each state, and an association between each state and the baseline parameter value for each state; and

storing, by the processor in a data store, the behavior model of the technology component.

2. The method of claim 1 further comprising:

monitoring, by the processor, the technology component to determine a current state of the technology component and a current parameter value of the technology component;

retrieving, by the processor, the behavior model of the technology component from the data store;

inputting, by the processor, the current operating state to the behavior model to retrieve the baseline parameter value associated with the current operating state;

determining, by the processor, whether the current parameter value of the technology component is within a threshold of the baseline parameter value; and

providing, by the processor, an alert to a device of a user if the current parameter value of the technology component is not within the threshold of the baseline parameter value.

3. The method of claim 2 wherein the current parameter value comprises at least one of a current power consumption value or a current temperature value of the technology component.

4. The method of claim 1 wherein the plurality of states comprises at least one of an online state, an offline state or a startup state.

5. The method of claim 1 wherein processing, by the processor, the plurality of parameter values associated with each state further comprises, determining, by the processor, a smoothed average of the plurality of parameter values of each state in the plurality of states to determine the baseline parameter value for each state.

6. A computer-implemented method for assembling a baseline power consumption model of a technology component, the method comprising:

monitoring, by a processor, a technology component over a period of time to determine a plurality of power consumption values associated with the technology component operating in each state of a plurality of states;

processing, by the processor, the plurality of power consumption values associated with each state in the plurality of states to determine a baseline power consumption value for each state in the plurality of states;

assembling, by the processor, a baseline power consumption model for the technology component, wherein the baseline power consumption model comprises each state, the baseline power consumption value for each state, and an association between each state and the baseline power consumption value for each state; and

storing, by the processor in a data store, the baseline power consumption model of the technology component.

7. The method of claim 6 wherein processing, by the processor, the plurality of power consumption values associated with each state further comprises, determining, by the processor, a smoothed average of the plurality of power consumption values of each state in the plurality of states to determine the baseline power consumption value for each state.

8. The method of claim 6 further comprising:

monitoring, by the processor, the technology component to determine a current power consumption value and a current operating state of the technology component;

retrieving, by the processor, the baseline power consumption model of the technology component from the data store;

inputting, by the processor, the current operating state to the baseline power consumption model to retrieve the baseline power consumption value associated with the current operating state;

determining, by the processor, whether the current power consumption value is within a threshold of the baseline power consumption value; and

providing, by the processor, an alert to a device of a user if the current power consumption value of the technology component is not within the threshold of the baseline power consumption value.

9. The method of claim 6 wherein at least one state in the plurality of states comprises of an online state.

10. The method of claim 9 wherein the technology component operates in the online state at a utilization value in a plurality of utilization values.

11. The method of claim 10 wherein the plurality of utilization values represent a percentage of a resource of the technology component which is in operation.

12. The method of claim 11 wherein the resource of the technology component comprises at least one of a memory resource or a central processing unit resource.

13. The method of claim 10 further comprising:

monitoring, by the processor, the technology component over the period of time to determine a plurality of power consumption values associated with the technology component operating at each utilization value in the plurality of utilization values;

processing, by the processor, the plurality of power consumption values associated with each utilization value in the plurality of utilization values to determine a baseline power consumption value for each utilization value of the plurality of utilization values; and

assembling, by the processor, the baseline power consumption model for the technology component, wherein the baseline power consumption model further comprises each utilization value, the baseline power consumption value for each utilization value, and an association between each utilization value and the baseline power consumption value for each state.

14. The method of claim 13 further comprising:

applying, by the processor, each utilization value and each baseline power consumption value to a curve fitting algorithm to generate a continuous representation of each baseline power consumption value as a function of each utilization value; and

assembling, by the processor, the baseline power consumption model for the technology component, wherein the baseline power consumption model further comprises the continuous representation of each baseline power consumption value as a function of each utilization value.

15. The method of claim 14 wherein the curve fitting algorithm comprises at least one of a Gauss-Newton algorithm or a linear least square method.

16. The method of claim 6 further comprising:

receiving, by the processor, an initial technology configuration of a data center, wherein the initial technology configuration comprises of a plurality of initial technology components each initial technology component being associated with a baseline power consumption model;

processing, by the processor, the baseline power consumption model of each initial technology component to generate an initial power consumption estimate of the data center over a period of time; and

providing, by the processor, a graph of the initial power consumption estimate of the data center over a period of time.

17. The method of claim 16 further comprising:

receiving, by the processor, an energy saving initiative and a reduction factor from a user, wherein the energy saving initiative provides a strategy for reducing the initial power consumption estimate of the data center by the reduction factor;

processing, by the processor, the initial technology configuration, the energy saving initiative, and the reduction factor to determine a recommended technology configuration capable of reducing the initial power consumption estimate by the reduction factor, wherein the recommended technology configuration comprises of a plurality of recommended technology components, each recommended technology component being associated with a baseline power consumption model;

determining, by the processor, an recommended power consumption estimate of the recommended technology configuration over the period of time based on the baseline power consumption model of each recommended technology component; and

providing, by the processor, a second graph displaying the initial power consumption estimate and the recommended power consumption estimate.

18. A system for assembling a behavior model of a technology component, the system comprising:

a memory to store a behavior model of a technology component; and

a processor operatively connected to the memory, the processor operative to monitor the technology component over a period of time to determine a plurality of parameter values associated with each state of a plurality of states of the technology component, process the plurality of parameter values associated with each state in the plurality of states to determine a baseline parameter value for each state in the plurality of states, assemble the behavior model for the technology component, wherein the behavior model comprises each state, the baseline parameter value for each state, and an association between each state and the baseline parameter value for each state, and store the behavior model of the technology component in the memory.

19. The system of claim 18 further comprising an interface operatively connected to the memory and the processor, and operative to communicate with a device of a user, wherein:

the processor is further operative to monitor the technology component to determine a current state of the technology component and a current parameter value of the technology component, retrieve the behavior model of the technology component from the data store, input the current operating state to the behavior model to retrieve the baseline parameter value associated with the current operating state, determine, whether the current parameter value of the technology component is within a threshold of the baseline parameter value, and provide an alert to the device of the user via the interface if the current parameter value of the technology component is not within the threshold of the baseline parameter value.

20. The system of claim 19 wherein the current parameter value comprises at least one of a current power consumption value or a current temperature value of the technology component.

21. The system of claim 18 wherein the plurality of states comprises at least one of an online state, an offline state or a startup state.

22. The system of claim 18 wherein the processor is further operative to determine a smoothed average of the plurality of parameter values of each state in the plurality of states to determine the baseline parameter value for each state.