Controlling data center cooling systems

• US 9,476,657 B1
• Filed: 11/20/2013
• Issued: 10/25/2016
• Est. Priority Date: 03/13/2013
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Tibbits, Andrew H., Pettis, Nathaniel Edward
• Primary Examiner(s): Laughlin, Nathan L
• Application Number:
  US14/085,066
• Time in Patent Office: 1,070 Days
• US Class Current: 1/1
• CPC Class Codes: F28D 1/024 with an air driving element
• CPC Class Codes: F28D 15/00 Heat-exchange apparatus with the intermediate heat-transfer medium in closed tubes passing into or through the conduit walls ; Heat-exchange apparatus employing intermediate heat-transfer medium or bodies F28D17/00, F28D19/00, F28D20/00 take precedence
• CPC Class Codes: F28D 2021/0019 Other heat exchangers for particular applications; Heat exchange systems not otherwise provided for
• CPC Class Codes: F28F 27/02 for controlling the distribution of heat-exchange media between different channels static flow control means in header boxes F28F9/026; arrangements of guide plates or guide vanes F28F9/22, F28F25/12
• CPC Class Codes: G05D 23/1917 using digital means
• CPC Class Codes: G05D 23/1932 to control the temperature of a plurality of spaces
• CPC Class Codes: H05K 7/2079 within rooms for removing heat from cabinets
• CPC Class Codes: H05K 7/208 Liquid cooling with phase change
• CPC Class Codes: H05K 7/20836 Thermal management, e.g. server temperature control

Abstract

Techniques for controlling a data center cooling system include polling a plurality of control devices associated with the data center cooling system for a respective state of each of the control devices; receiving, from each of the plurality of control devices, a response that includes the respective state; aggregating the responses from the plurality of control devices; executing a control algorithm that includes the aggregated responses as an input to the algorithm and an output that includes a setpoint of the plurality of control devices; and transmitting the output to the plurality of control devices.

Claims

1. A computer-implemented method for controlling a data center cooling system, the method comprising:

polling a plurality of control devices associated with the data center cooling system for a respective state of each of the control devices, the plurality of control devices comprising a plurality of control valves, each of the control valves in fluid communication with a cooling fluid source and at least one air-to-liquid heat exchanger mounted in a modular cooling apparatus, wherein polling a plurality of control devices associated with the data center cooling system comprises polling a plurality of controllers communicably coupled to the plurality of control valves by a proxy, the proxy executed on at least one rack-mounted computing device of the data center;

receiving, from each of the plurality of control devices, a response that comprises the respective state;

aggregating the responses from the plurality of control devices;

executing a control algorithm that comprises the aggregated responses as an input to the algorithm and an output that comprises a setpoint of the plurality of control devices; and

transmitting the output to the plurality of control devices.

2. The method of claim 1, wherein,

each respective state comprises a valve position of each respective control valve, and

the setpoint comprises a new valve position of each respective control valve.

3. The method of claim 1, wherein executing a control algorithm comprises:

determining a temperature of air leaving each respective modular cooling unit;

determining a temperature of the cooling fluid circulated to each respective modular cooling unit through the respective control valves;

determining an approach temperature that comprises a difference between the temperature of the air leaving each respective modular cooling unit and the temperature of the cooling fluid circulated to each respective modular cooling unit; and

determining the new valve position based on the determined approach temperature and an approach temperature setpoint.

4. The method of claim 1, further comprising:

partitioning the at least one rack-mounted computing device into an authenticated portion and an unauthenticated portion, the authenticated portion executing a plurality of processing jobs received from networked computing devices external to the data center; and

routing the responses and output through the unauthenticated portion.

5. The method of claim 1, further comprising:

executing a second control algorithm that comprises the aggregated responses as a second input to the second algorithm and a second output that comprises the setpoint of the plurality of control devices;

storing the second output; and

based on a failure to transmit the output to the plurality of control devices, transmit the stored second output to the plurality of control devices.

6. The method of claim 5, wherein the second output is stored in a database that is physically separated from the data center.

7. The method of claim 1, further comprising:

determining a failure mode; and

based on the failure mode, reverting the plurality of control devices to respective fail safe positions.

8. A system of one or more computers configured to perform operations comprising:

polling a plurality of control devices associated with the data center cooling system for a respective state of each of the control devices, the plurality of control devices comprising a plurality of control valves, each of the control valves in fluid communication with a cooling fluid source and at least one air-to-liquid heat exchanger mounted in a modular cooling apparatus, wherein polling a plurality of control devices associated with the data center cooling system comprises polling a plurality of controllers communicably coupled to the plurality of control valves by a proxy, the proxy executed on at least one rack-mounted computing device of the data center;

receiving, from each of the plurality of control devices, a response that comprises the respective state;

aggregating the responses from the plurality of control devices;

executing a control algorithm that comprises the aggregated responses as an input to the algorithm and an output that comprises a setpoint of the plurality of control devices; and

transmitting the output to the plurality of control devices.

9. The system of claim 8, wherein,

each respective state comprises a valve position of each respective control valve, and

the setpoint comprises a new valve position of each respective control valve.

10. The system of claim 8, wherein executing a control algorithm comprises:

determining a temperature of air leaving each respective modular cooling unit;

determining a temperature of the cooling fluid circulated to each respective modular cooling unit through the respective control valves;

determining an approach temperature that comprises a difference between the temperature of the air leaving each respective modular cooling unit and the temperature of the cooling fluid circulated to each respective modular cooling unit; and

determining the new valve position based on the determined approach temperature and an approach temperature setpoint.

11. The system of claim 8, wherein the operations further comprise:

partitioning the at least one rack-mounted computing device into an authenticated portion and an unauthenticated portion, the authenticated portion executing a plurality of processing jobs received from networked computing devices external to the data center; and

routing the responses and output through the unauthenticated portion.

12. The system of claim 8, wherein the operations further comprise:

executing a second control algorithm that comprises the aggregated responses as a second input to the second algorithm and a second output that comprises the setpoint of the plurality of control devices;

storing the second output; and

based on a failure to transmit the output to the plurality of control devices, transmit the stored second output to the plurality of control devices.

13. The system of claim 12, wherein the second output is stored in a database that is physically separated from the data center.

14. The system of claim 8, wherein the operations further comprise:

determining a failure mode; and

based on the failure mode, reverting the plurality of control devices to respective fail safe positions.

15. A non-transitory computer storage medium encoded with a computer program, the program comprising instructions that when executed by one or more computers cause the one or more computers to perform operations comprising:

polling a plurality of control devices associated with the data center cooling system for a respective state of each of the control devices, the plurality of control devices comprising a plurality of control valves, each of the control valves in fluid communication with a cooling fluid source and at least one air-to-liquid heat exchanger mounted in a modular cooling apparatus, wherein polling a plurality of control devices associated with the data center cooling system comprises polling a plurality of controllers communicably coupled to the plurality of control valves by a proxy, the proxy executed on at least one rack-mounted computing device of the data center;

receiving, from each of the plurality of control devices, a response that comprises the respective state;

aggregating the responses from the plurality of control devices;

executing a control algorithm that comprises the aggregated responses as an input to the algorithm and an output that comprises a setpoint of the plurality of control devices; and

transmitting the output to the plurality of control devices.

16. The non-transitory computer storage medium of claim 15, wherein,

each respective state comprises a valve position of each respective control valve, and

the setpoint comprises a new valve position of each respective control valve.

17. The non-transitory computer storage medium of claim 15, wherein executing a control algorithm comprises:

determining a temperature of air leaving each respective modular cooling unit;

determining a temperature of the cooling fluid circulated to each respective modular cooling unit through the respective control valves;

determining an approach temperature that comprises a difference between the temperature of the air leaving each respective modular cooling unit and the temperature of the cooling fluid circulated to each respective modular cooling unit; and

determining the new valve position based on the determined approach temperature and an approach temperature setpoint.

18. The non-transitory computer storage medium of claim 15, wherein the operations further comprise:

partitioning the at least one rack-mounted computing device into an authenticated portion and an unauthenticated portion, the authenticated portion executing a plurality of processing jobs received from networked computing devices external to the data center; and

routing the responses and output through the unauthenticated portion.

19. The non-transitory computer storage medium of claim 15, wherein the operations further comprise:

executing a second control algorithm that comprises the aggregated responses as a second input to the second algorithm and a second output that comprises the setpoint of the plurality of control devices;

storing the second output; and

based on a failure to transmit the output to the plurality of control devices, transmit the stored second output to the plurality of control devices.

20. The non-transitory computer storage medium of claim 19, wherein the second output is stored in a database that is physically separated from the data center.

21. The non-transitory computer storage medium of claim 15, wherein the operations further comprise:

determining a failure mode; and

based on the failure mode, reverting the plurality of control devices to respective fail safe positions.

22. A computer-implemented method for controlling a data center cooling system, the method comprising:

polling a plurality of control devices associated with the data center cooling system for a respective state of each of the control devices, the plurality of control devices comprising a plurality of control valves, each of the control valves in fluid communication with a cooling fluid source and at least one air-to-liquid heat exchanger mounted in a modular cooling apparatus;

receiving, from each of the plurality of control devices, a response that comprises the respective state;

aggregating the responses from the plurality of control devices;

executing a control algorithm that comprises the aggregated responses as an input to the algorithm and an output that comprises a setpoint of the plurality of control devices;

transmitting the output to the plurality of control devices;

executing a second control algorithm that comprises the aggregated responses as a second input to the second algorithm and a second output that comprises the setpoint of the plurality of control devices;

storing the second output; and

based on a failure to transmit the output to the plurality of control devices, transmit the stored second output to the plurality of control devices.

23. The method of claim 22, wherein polling a plurality of control devices associated with the data center cooling system comprises polling a plurality of controllers communicably coupled to the plurality of control valves by a proxy, the proxy executed on at least one rack-mounted computing device of the data center.

24. The method of claim 22, wherein,

each respective state comprises a valve position of each respective control valve, and

the setpoint comprises a new valve position of each respective control valve.

25. The method of claim 22, wherein executing a control algorithm comprises:

determining a temperature of air leaving each respective modular cooling unit;

determining a temperature of the cooling fluid circulated to each respective modular cooling unit through the respective control valves;

determining an approach temperature that comprises a difference between the temperature of the air leaving each respective modular cooling unit and the temperature of the cooling fluid circulated to each respective modular cooling unit; and

determining the new valve position based on the determined approach temperature and an approach temperature setpoint.