Grid-friendly data center

• US 9,691,112 B2
• Filed: 05/31/2013
• Issued: 06/27/2017
• Est. Priority Date: 05/31/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation, Universiti Brunei Darussalam
• Original Assignee: International Business Machines Corporation, Universiti Brunei Darussalam
• Inventors: Arya, Vijay, Ganu, Tanuja Hrishikesh, Husain, Saiful A., Kumar, Ashkok Pon, Seetharamakrishnan, Devasenapathi Periagraharam, De Silva, Liyanage Chandratilak, Kalyanaraman, Shivkumar, Mylavarapu, Swarna Latha
• Primary Examiner(s): Huang, Miranda
• Assistant Examiner(s): Lopez Alvarez, Olvin
• Application Number:
  US13/907,084
• Publication Number:
  US 20140358309A1
• Time in Patent Office: 1,488 Days
• Field of Search: 700295
• 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
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 50/06 Energy or water supply

Abstract

Methods and arrangements for calculating a power budget. Data are received, the data including data relating to a current state of an electrical grid at a power-drawing location which draws at least one variable electrical load, the at least one variable electrical load being subject to variable pricing. At least one prediction is determined, the at least one prediction pertaining to a future state of the electrical grid and a future electrical load to be drawn at the power-drawing location. A power budget is thereupon calculated, with respect to the power-drawing location, based on the data and the at least one prediction. Other variants and embodiments are broadly contemplated herein.

Claims

1. A method of calculating a power budget, said method comprising:

utilizing at least one processor to execute computer code configured to perform the steps of;

receiving data, the data including data relating to a current state of an electrical grid at a power-drawing location which draws at least one variable electrical load, the at least one variable electrical load being subject to variable pricing;

determining at least one prediction relating to a future state of the electrical grid over a predetermined future time period, wherein the determining comprises identifying an estimated pricing function for the predetermined future time period, an estimated grid state function for the predetermined future time period, and a power sensitivity index value;

the determining comprising weighting the estimated pricing function and the estimated grid state function based on the power sensitivity index value and identifying the function having a higher weighted value and selecting, from the estimated pricing function and the estimated grid state function, a value, corresponding to the function having the higher weighted value, for the future state of the electrical grid;

identifying, based upon the predicted future state of the electrical grid, a time period from the predetermined future time period corresponding to an overload of the electrical grid, wherein the identifying comprises identifying a value for the overload that is greater than a predetermined value;

determining a future electrical load to be drawn at the power-drawing location for the time period from the predetermined future time period, wherein the future electrical load comprises a variable electrical load and a base electrical load at the power-drawing location;

identifying a correlation between the predicted future state of the electrical grid and the future electrical load, wherein the correlation identifies a degree of influence of the future electrical load to the predicted future state of the electrical grid; and

thereupon calculating a power budget, with respect to the power-drawing location, based on the correlation.

2. The method according to claim 1, wherein the power-drawing location comprises a data center.

3. The method according to claim 2, wherein the power drawing location comprises an air-conditioning system which cools the data center.

4. The method according to claim 1, wherein the data additionally include data relating to external temperatures at the power-drawing location.

5. The method according to claim 1, wherein the at least one prediction additionally relates to future external temperatures at the power-drawing location.

6. The method according to claim 1, wherein said determining of at least one prediction comprises predicting a future state of the electrical grid via receiving at least one of:

current voltage information, and current pricing information.

7. The method according to claim 6, wherein said determining of at least one prediction comprises predicting a future state of the electrical grid via receiving both of:

current voltage information, and current pricing information.

8. The method according to claim 1, wherein the power-drawing location comprises a location which draws non-deferrable electrical loads.

9. The method according to claim 8, wherein the power-drawing location comprises a data center.

10. An apparatus for calculating a power budget, said apparatus comprising:

at least one processor; and

a computer readable storage medium having computer readable program code embodied therewith and executable by the at least one processor, the computer readable program code comprising;

computer readable program code configured to receive data, the data including data relating to a current state of an electrical grid at a power-drawing location which draws at least one variable electrical load, the at least one variable electrical load being subject to variable pricing;

computer readable program code configured to determine at least one prediction relating to a future state of the electrical grid over a predetermined future time period, wherein the determining comprises identifying an estimated pricing function for the predetermined future time period, an estimated grid state function for the predetermined future time period, and a power sensitivity index value;

the determining comprising weighting the estimated pricing function and the estimated grid state function based on the power sensitivity index value and identifying the function having a higher weighted value and selecting, from the estimated pricing function and the estimated grid state function, a value, corresponding to the function having the higher weighted value, for the future state of the electrical grid;

computer readable program code configured to identify, based upon the predicted future state of the electrical grid, a time period from the predetermined future time period corresponding to an overload of the electrical grid, wherein the identifying comprises identifying a value for the overload that is greater than a predetermined value;

computer readable program code configured to determine a future electrical load to be drawn at the power-drawing location for the time period from the predetermined future time period, wherein the future electrical load comprises a variable electrical load and a base electrical load at the power-drawing location;

computer readable program code to identify a correlation between the predicted future state of the electrical grid and the future electrical load, wherein the correlation identifies a degree of influence of the future electrical load to the predicted future state of the electrical grid; and

computer readable program code configured to thereupon calculate a power budget, with respect to the power-drawing location, based on the correlation.

11. A computer program product for calculating a power budget, said computer program product comprising:

a non-transitory computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising;

computer readable program code configured to receive data, the data including data relating to a current state of an electrical grid at a power-drawing location which draws at least one variable electrical load, the at least one variable electrical load being subject to variable pricing;

computer readable program code configured to determine at least one prediction relating to a future state of the electrical grid over a predetermined future time period, wherein the determining comprises identifying an estimated pricing function for the predetermined future time period, an estimated grid state function for the predetermined future time period, and a power sensitivity index value;

the determining comprising weighting the estimated pricing function and the estimated grid state function based on the power sensitivity index value and identifying the function having a higher weighted value and selecting, from the estimated pricing function and the estimated grid state function, a value, corresponding to the function having the higher weighted value, for the future state of the electrical grid;

computer readable program code configured to identify, based upon the predicted future state of the electrical grid, a time period from the predetermined future time period corresponding to an overload of the electrical arid, wherein the identifying comprises identifying a value for the overload that is greater than a predetermined value;

computer readable program code configured to determine a future electrical load to be drawn at the power-drawing location for the time period from the predetermined future time period, wherein the future electrical load comprises a variable electrical load and a base electrical load at the power-drawing location;

computer readable program code to identify a correlation between the predicted future state of the electrical grid and the future electrical load, wherein the correlation identifies a degree of influence of the future electrical load to the predicted future state of the electrical grid; and

computer readable program code configured to thereupon calculate a power budget, with respect to the power-drawing location, based on the correlation.

12. The computer program product according to claim 11, wherein the power-drawing location comprises a data center.

13. The computer program product according to claim 12, wherein the power-drawing location comprises an air-conditioning system which cools the data center.

14. The computer program product according to claim 11, wherein the data additionally include data relating to external temperatures at the power-drawing location.

15. The computer program product according to claim 11, wherein the at least one prediction additionally relates to future external temperatures at the power-drawing location.

16. The computer program product according to claim 15, wherein said computer readable program code is configured to predict a future state of the electrical grid via receiving at least one of:

current voltage information, and current pricing information.

17. The computer program product according to claim 16, wherein said computer readable program code is configured to predict a future state of the electrical grid via receiving both of current voltage information, and current pricing information.

18. The computer program product according to claim 11, wherein the power-drawing location comprises a location which draws non-deferrable electrical loads.

19. The computer program product according to claim 18, wherein the power-drawing location comprises a data center.

20. A method comprising:

measuring, with respect to a data center, current grid load conditions, current energy pricing and external temperature, the data center including a plurality of computing nodes and a cooling system for cooling the computer nodes;

predicting future loads for a predetermined future time period with respect to the data center, wherein the future loads comprise a variable electrical load and a base electrical load at the data center;

predicting future grid load conditions for the predetermined future time period, wherein the future grid load conditions comprise an estimated future pricing function, an estimated grid state function, and a power sensitivity index value;

the predicting future grid load conditions comprising weighting the estimated future pricing function and the estimated grid state function based on the power sensitivity index value and identifying the function having a higher weighted value and selecting, from the estimated future pricing function and the estimated grid state function, a value, corresponding to the function having the higher weighted value, for a predicted future state of the electrical grid;

identifying, based upon the predicted future state of the electrical grid, a time period from the predetermined future time period corresponding to an overload of the electrical grid, wherein the identifying comprises identifying a value for the overload that is greater than a predetermined value;

determining, for the time period from the predetermined future time period, a correlation between the predicted future loads and the predicted future state of the electrical grid, wherein the correlation identifies a degree of influence of the predicted future loads to the predicted future state of the electrical grid; and

calculating a future power budget with respect to the data center, based on said correlation.