DATA CENTER THERMAL PERFORMANCE OPTIMIZATION USING DISTRIBUTED COOLING SYSTEMS

• US 20100076607A1
• Filed: 08/06/2009
• Published: 03/25/2010
• Est. Priority Date: 08/08/2008
• Status: Active Grant

• Current Assignee: Siemens Industry Incorporated (Siemens AG)
• Original Assignee: Siemens Industry Incorporated (Siemens AG)
• Inventors: Pienta, William Thomas, Ahmed, Osman
• Granted Patent: US 8,346,398 B2
• Time in Patent Office: Days
• US Class Current: 700/276
• CPC Class Codes: G06F 1/206 comprising thermal management
• CPC Class Codes: H05K 7/20745 within rooms for removing heat from cabinets, e.g. by air conditioning device
• CPC Class Codes: H05K 7/20836 Thermal management, e.g. server temperature control
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

The various embodiments described herein relate to systems and methods, circuits and devices for providing data center cooling while optimizing power usage effectiveness and/or compute power efficiency of the data center. The various embodiments can provide optimized thermal performance and can reduce power consumption of the data center by strategically locating sensor modules, preferably microsystems with MEMS technology, in the data center and using a processing circuit to acquire data from the sensors and to generate a control law for operating the air conditioning system efficiently. In particular the sensors are operable to measure and provide granular environmental data to further characterize the environmental conditions of the racks locally, and the data center as a whole. The processing circuit may also generate a profile of local racks and simulate a data center environment to develop and test control strategies for implementation in the actual data center.

Claims

1. An arrangement for optimizing power efficiency of a data center having

a plurality of racks having at least one server computer, comprising:

a plurality of sensors coupled to one or more of the plurality of racks and operable to measure at least one environmental parameter of said rack, each of the sensors configured to wirelessly transmit a measured environmental parameter; and

a processing circuit configured to receive information based on the at least one measured environmental parameter, the processing circuit further configured to;

generate a profile indicative of the at least one environmental parameter corresponding to said rack; and

generate a control law configured to operate one or more air conditioning systems in order to satisfy a thermal demand of the data center in an optimally power efficient manner based at least in part on information representative of the profile of said rack.

2. The arrangement of claim 1, wherein each of the sensors further comprises a MEMS sensor.

3. The arrangement of claim 1, wherein the at least one environmental parameter comprises at least one of temperature, relative humidity, pressure, and air flow.

4. The arrangement of claim 1, further comprising a memory operably connected to the processing circuit, the memory configured to store power efficiency characteristic information for said air conditioning unit.

5. The arrangement of claim 4, wherein the generated control law is based on at least one of the profile of said rack and the power efficiency characteristic information of said air conditioning unit.

6. The arrangement of claim 5, wherein the processing circuit is further configured to calculate total power consumption of said air conditioning system.

7. The arrangement of claim 1, wherein the processing circuit further comprises a virtual data center representative of said data center, wherein the virtual data center is configured to:

acquire the at least one environmental parameter;

generate a simulated control law for operating said air conditioning system based on the at least one environmental parameter; and

validate the simulated control law before implementation in said air conditioning system.

8. The arrangement of claim 7, wherein the virtual data center comprises an expert control system configured to optimize total power consumption of said air conditioning system for the generated profile of said rack.

9. The arrangement of claim 8, wherein the expert control system is selected from the group consisting of a fuzzy expert control system and a neural network.

10. The arrangement of claim 9, further comprising a memory operably connected to the processing circuit and configured to archive the optimized total power consumption of said air conditioning system for the generated profile of said rack.

11. A method for optimizing power efficiency of a data center having a plurality of racks having at least one server computer, comprising:

receiving at least one measured environmental parameter of one or more racks;

generating a profile indicative of the at least one measured environmental parameter of said rack;

generating a control law operable to satisfy a thermal demand of the data center in an optimally power efficient manner based at least in part on information representative of the profile of said rack; and

implementing the control law to operate one or more air conditioning systems.

12. The method of claim 11, wherein each of the sensors further comprises a MEMS sensor coupled to the racks and configured to wirelessly transmit a measured environmental parameter, the method further comprising wirelessly transmitting the at least one environmental parameter.

13. The method of claim 11, wherein the at least one measured environmental parameters comprises at least one of temperature, relative humidity, pressure, and air flow.

14. The method of claim 11, further comprising storing power efficiency characteristic information for said air conditioning system in a memory operably coupled to a processing circuit.

15. The method of claim 14, wherein the generated control law is based on at least one of the profile of said rack and the power efficiency characteristic information of said air conditioning system.

16. The method of claim 15, further comprising calculating total power consumption of said air conditioning system.

17. The method of claim 11, further comprising acquiring the at least one measured environmental parameter by a virtual data center comprising an expert control system, wherein the expert control system is configured to generate a simulated control law for operating said air conditioning system based on the acquired measured environmental parameter, and to validate the simulated control law before implementation in said air conditioning system.

18. The method of claim 17, wherein the expert control system is configured to optimize total power consumption of said air conditioning system for the generated profile of said rack.

19. The method of claim 17, wherein the expert control system is selected from the group consisting of a fuzzy expert control system and a neural network.

20. The method of claim 19, further comprising storing the optimized total power consumption of said air conditioning system for the generated profile of said rack in a memory operably coupled to a processing circuit.

21. An arrangement for optimizing power efficiency of a data center having a plurality of racks having at least one server computer, comprising:

a plurality of sensors coupled to one or more racks and operable to measure at least one environmental parameter of said rack, each sensor configured to wirelessly transmit a measured environmental parameter;

a rack controller coupled to said rack and associated with one or more sensors, the rack controller comprising a processing circuit configured to receive information based on the at least one measured environmental parameter from the respective sensors, the processing circuit further configured to generate a control law configured to operate one or more air conditioning systems in order to satisfy a local thermal demand of said rack in an optimally power efficient manner based at least in part on information representative of the at least one environmental parameter.

22. An arrangement for optimizing power efficiency of a data center having a plurality of racks having at least one server computer, comprising:

a plurality of sensors coupled to one or more racks and operable to measure at least one environmental parameter of said rack, each sensor configured to wirelessly transmit a measured environmental parameter; and

a processing circuit configured to receive information based on the at least one measured environmental parameter, the processing circuit further configured to generate a control law configured to operate a dedicated cooler configured to at least partially satisfy a local thermal demand of said rack, and a control law configured to operate one or more air conditioning systems in order to satisfy an overall thermal load of the data center, each in an optimally power efficient manner based at least in part on information representative of the at least one environmental parameter.