Method for predicting power usage effectiveness and data center infrastructure efficiency within a real-time monitoring system

• US 8,401,833 B2
• Filed: 08/17/2009
• Issued: 03/19/2013
• Est. Priority Date: 08/15/2008
• Status: Active Grant

• Current Assignee: Bentley Systems Incorporated
• Original Assignee: Power Analytics Corporation (PAC Global, Inc.)
• Inventors: Radibratovic, Branislav, Meagher, Kevin, Needle, Barry
• Primary Examiner(s): ALHIJA, SAIF A
• Application Number:
  US12/542,472
• Publication Number:
  US 20100049494A1
• Time in Patent Office: 1,310 Days
• Field of Search: 703/2, 703/13
• US Class Current: 703/13
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202

Abstract

A system for making real-time predictions about power usage efficiency (PUE) and/or data center infrastructure efficiency (DCiE) of an electrical system comprises a data acquisition component communicatively connected to a sensor configured to acquire real-time data output from the electrical system; an analytics server communicatively connected to the data acquisition component and comprising a virtual system modeling engine configured to generate predicted data output for the electrical system using a virtual system model of the electrical system, an analytics engine configured to monitor the real-time data output and the predicted data output of the electrical system, and a PUE/DCiE simulation engine configured to use the virtual system model updated based in the real-time data to forecast the PUE/DCiE.

Claims

1. A system for making real-time predictions about power usage efficiency (PUE) of an electrical system, the system comprising:

a data acquisition component communicatively connected to one or more sensors configured to acquire real-time data output from the electrical system of a data center, wherein the real-time data output comprises real-time data for one or more transformers in the data center and real-time data for one or more power supplies providing power to information technology (IT) equipment in the data center; and

an analytics server communicatively connected to the data acquisition component, the analytics server comprising

a virtual system modeling engine configured to generate predicted data output for the electrical system using a virtual system model of the electrical system,

an analytics engine configured to monitor the real-time data output and the predicted data output of the electrical system, and validate the virtual system model using the real-time data output, and

a PUE simulation engine configured to receive one or more modifications for the data center, modify a copy of the validated virtual system model based on the received modifications, and use the modified copy of the validated virtual system model, to forecast the PUE for the data center with the one or more modifications, wherein the PUE comprises a ratio of total power dedicated solely to the data center to a measure of load associated with the IT equipment in the data center.

2. The system for making real-time predictions about PUE of an electrical system, as recited in claim 1, further comprising a client terminal communicatively connected to the analytics server and configured to communicate the forecasted PUE.

3. The system for making real-time predictions about PUE of an electrical system, as recited in claim 1, wherein the forecasted PUE is communicated by way of graphics on a display interfaced with the client terminal.

4. The system for making real-time predictions about PUE of an electrical system, as recited in claim 1, wherein the forecasted PUE is communicated by way of text on a display interfaced with the client terminal.

5. The system for making real-time predictions about PUE of an electrical system, as recited in claim 1, wherein the forecasted PUE is communicated by way of a paper report generated by a printing device interfaced with the client terminal.

6. The system for making real-time predictions about PUE of an electrical system, as recited in claim 1, wherein the PUE simulation engine is further configured to use the validated virtual system model to forecast data center infrastructure efficiency (DCiE).

7. The system for making real-time predictions about PUE of an electrical system, as recited in claim 1, wherein the analytics server is configured to:

create a base-line PUE model of the electrical system; and

calculate the PUE of the electrical system using the base-line PUE model.

8. The system for making real-time predictions about PUE of an electrical system, as recited in claim 7, wherein the analytics server is further configured to:

create a calibrated PUE model of the electrical system; and

calculate the PUE of the electrical system using the calibrated PUE model.

9. The system for making real-time predictions about PUE of an electrical system, as recited in claim 8, wherein the analytics server is further configured to:

compare the PUE calculated from the calibrated PUE model with the PUE calculated from the base-line PUE model to validate the base-line model.

10. The system for making real-time predictions about PUE of an electrical system, as recited in claim 9, wherein the base-line PUE model is validated if the PUE values from the calibrated model and base-line model are within a predetermined range.

11. The system for making real-time predictions about PUE of an electrical system, as recited in claim 9, wherein the analytics server is further configured to:

update the calibrated PUE model with a first modification to the electrical system; and

calculate the PUE of the electrical system using the updated calibrated PUE model.

12. The system for making real-time predictions about PUE of an electrical system, as recited in claim 11, wherein the analytics server is further configured to:

update the calibrated PUE model with a second modification to the electrical system; and

calculate the PUE of the electrical system using the updated calibrated PUE model.

13. A system for making real-time predictions about data center infrastructure efficiency (DCiE) of an electrical system, the system comprising:

a data acquisition component communicatively connected to one or more sensors configured to acquire real-time data output from the electrical system of a data center, wherein the real-time data output comprises real-time data for one or more transformers in the data center and real-time data for one or more power supplies providing power to information technology (IT) equipment in the data center; and

an analytics server communicatively connected to the data acquisition component, the analytics server comprising

a virtual system modeling engine configured to generate predicted data output for the electrical system using a virtual system model of the electrical system,

an analytics engine configured to monitor the real-time data output and the predicted data output of the electrical system, and validate the virtual system model using the real-time data output, and

a DCiE simulation engine configured to receive one or more modifications for the data center, modify a copy of the validated virtual system model based on the received one or more modifications, and use the modified copy of the validated virtual system model to forecast the DCiE for the data center with the one or more modifications, wherein the DCiE comprises a ratio of a measure of load associated with the IT equipment in the data center to a total power dedicated solely to the data center.

14. The system for making real-time predictions about DCiE of an electrical system, as recited in claim 13, further comprising a client terminal communicatively connected to the analytics server and configured to communicate the forecasted DCiE.

15. The system for making real-time predictions about DCiE of an electrical system, as recited in claim 13, wherein the forecasted DCiE is communicated by way of graphics on a display interfaced with the client terminal.

16. The system for making real-time predictions about DCiE of an electrical system, as recited in claim 13, wherein the forecasted DCiE is communicated by way of text on a display interfaced with the client terminal.

17. The system for making real-time predictions about DCiE of an electrical system, as recited in claim 13, wherein the analytics server is configured to:

create a base-line DCiE model of the electrical system; and

calculate the DCiE of the electrical system using the base-line DCiE model.

18. The system for making real-time predictions about DCiE of an electrical system, as recited in claim 17, wherein the analytics server is further configured to:

create a calibrated DCiE model of the electrical system; and

calculate the DCiE of the electrical system using the calibrated DCiE model.

19. The system for making real-time predictions about DCiE of an electrical system, as recited in claim 18, wherein the analytics server is further configured to:

compare the DCiE calculated from the calibrated DCiE model with the DCiE calculated from the base-line DCiE model to validate the base-line model.

20. The system for making real-time predictions about DCiE of an electrical system, as recited in claim 19, wherein the base-line DCiE model is validated if the DCiE values from the calibrated model and base-line model are within a predetermined range.

21. The system for making real-time predictions about DCiE of an electrical system, as recited in claim 19, wherein the analytics server is further configured to:

update the calibrated DCiE model with a first modification to the electrical system; and

calculate the DCiE of the electrical system using the updated calibrated DCiE model.

22. The system for making real-time predictions about DCiE of an electrical system, as recited in claim 21, wherein the analytics server is further configured to:

update the calibrated DCiE model with a second modification to the electrical system; and

calculate the DCiE of the electrical system using the updated calibrated DCiE model.