白丝美女被狂躁免费视频网站,500av导航大全精品,yw.193.cnc爆乳尤物未满,97se亚洲综合色区,аⅴ天堂中文在线网官网

Intelligent in-rack pump or compressor unit for datacenter cooling systems

專利號
US12082382B2
公開日期
2024-09-03
申請人
Nvidia Corporation（US CA Santa Clara）
發(fā)明人
Ali Heydari
IPC分類
H05K7/20; G05B13/02
技術(shù)領(lǐng)域
least,in,at,embodiment,or,may,cooling,network,be,rack
地域： CA CA Santa Clara

摘要

Systems and methods for cooling a datacenter are disclosed. In at least one embodiment, a control unit within a rack has a pump or compressor unit to cause two-phase fluid to circulate through a cold plate associated with a computing device and to circulate through a heat exchanger associated with a rear door of a rack, so as to dissipate heat from a computing device through a heat exchanger by a control unit within a rack.

說明書

FIELD

At least one embodiment pertains to cooling systems, including systems and methods for operating those cooling systems. In at least one embodiment, such a cooling system can be utilized in a datacenter containing one or more racks or computing servers.

BACKGROUND

Datacenter cooling systems use fans to circulate air through server components. Certain supercomputers or other high capacity computers may use water or other cooling systems instead of air-cooling systems to draw heat away from the server components or racks of the datacenter to an area external to the datacenter. The cooling systems may include a chiller within the datacenter area, which may include area external to the datacenter itself. Further, the area external to the datacenter may include a cooling tower or other external heat exchanger that receives heated coolant from the datacenter and that disperses the heat by forced air or other means to the environment (or an external cooling medium). The cooled coolant is recirculated back into the datacenter. The chiller and the cooling tower together form a chilling facility.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates an exemplary datacenter cooling system subject to improvements described in at least one embodiment;

FIG. 2 illustrates server-level features associated with an intelligent in-rack pump or compressor unit for a datacenter cooling system, according to at least one embodiment;

權(quán)利要求

1
What is claimed is:1. A datacenter cooling system, comprising: a control unit within a rack and comprising a pump or compressor unit to cause two-phase fluid to flow from a cold plate coupled with at least one computing device to a heat exchanger incorporated with a rear door of the rack for dissipation of heat from the at least one computing device through the heat exchanger, the control unit further comprising an input to receive sensor inputs associated with a temperature from the at least one computing device, secondary fluid, or the two-phase fluid, wherein one or more neural networks of the control unit are trained to infer a change in coolant state has occurred based in part on the temperature and on prior temperatures in the rear door heat exchanger, the control unit to enable or disable the pump or compressor unit.2. The datacenter cooling system of claim 1, further comprising:at least one processor within the control unit to determine a temperature associated with the at least one computing device or a secondary coolant and to enable the pump or compressor unit of the control unit to provide cooling along with or instead of a secondary cooling loop.3. The datacenter cooling system of claim 1, further comprising:at least one flow controller associated with the control unit, the at least one flow controller to be enabled based in part on a cooling requirement for a secondary coolant or for the at least one computing device.4. The datacenter cooling system of claim 1, further comprising:the cold plate having ports for microchannels to support flow of a secondary coolant and for an evaporator section to support flow of the two-phase fluid.5. The datacenter cooling system of claim 1, further comprising:at least one processor within the control unit to receive sensor inputs from sensors associated with the at least one computing device or a secondary coolant, the at least one processor to enable the pump or compressor unit to support cooling for the computing device using the two-phase fluid instead of a secondary cooling loop.6. The datacenter cooling system of claim 5, further comprising: the one or more neural networks to receive the sensor inputs and to infer a cooling requirement for the at least one computing device.7. The datacenter cooling system of claim 1, further comprising:at least one processor to cause at least one flow controller to enable flow of the two-phase fluid through the pump or compressor unit and to prevent flow of a secondary coolant to a secondary cooling loop.8. The datacenter cooling system of claim 1, wherein the two-phase fluid is a refrigerant or an engineered fluid.9. The datacenter cooling system of claim 1, further comprising:at least one flow controller associated with the pump or compressor unit and with a secondary cooling loop, the at least one flow controller to receive input from the control unit, to support flow of the two-phase fluid through the pump or compressor unit, and to prevent flow of a secondary coolant to a secondary cooling loop.10. The datacenter cooling system of claim 1, further comprising:at least one processor within the control unit to enable a first mode of a datacenter cooling system to provide cooling by an action of the pump or compressor unit, and the at least one processor to enable a second mode of the datacenter cooling system to provide cooling from a secondary cooling loop associated with a primary cooling loop and a chilling facility.11. A processor within a control unit of a rack, the processor comprising one or more circuits to determine a cooling requirement for at least one computing device, the control unit further comprising a pump or compressor unit to cause, upon input from the processor, two-phase fluid to flow from a cold plate coupled with the at least one computing device to a heat exchanger integrated with a rear door of the rack for dissipation of heat from the at least one computing device through the heat exchanger, the processor further comprising an input to receive sensor inputs associated with a temperature from the at least one computing device, secondary coolant, or the two phase fluid, wherein one or more neural networks of the processor are trained to infer a change in coolant state has occurred based in part on the temperature and on prior temperatures in the rear door heat exchanger, the one or more circuits to enable or disable the pump or compressor unit.12. The processor of claim 11, further comprising:an output to provide signals for at least one flow controller to enable flow of the two-phase fluid through the pump or compressor unit and to prevent flow of a secondary coolant to a secondary cooling loop.13. The processor of claim 11, further comprising:an input to receive sensor inputs from sensors associated with the at least one computing device, a rack, a secondary coolant, or the two-phase fluid, the processor to determine a first cooling requirement associated with a secondary cooling loop and a second cooling requirement associated with the pump or compressor unit, based in part on the sensor inputs.14. The processor of claim 13, further comprising: the one or more neural networks to receive the sensor inputs and to infer the first cooling requirement and the second cooling requirement.15. The processor of claim 11, further comprising: the one or more neural networks to infer a failure of a secondary cooling loop, the one or more circuits to cause at least one flow controller to activate the pump or compressor unit.16. A processor of a control unit within a rack, the processor comprising one or more circuits to train one or more neural networks to infer, from sensor inputs of sensors associated with at least one computing device that a cooling requirement exists, the processor to cause a response to the cooling requirement, the control unit comprising a pump or compressor unit to enable two-phase fluid to flow from a cold plate coupled with the at least one computing device to a heat exchanger integrated with a rear door of the rack for dissipation of heat from the at least one computing device through the heat exchanger, the processor further comprising an input to receive the sensor inputs associated with a temperature from the at least one computing device, secondary coolant, or the two-phase fluid, the one or more neural networks trained to infer a change in coolant state has occurred based in part on the temperature and on prior temperatures in the rear door heat exchanger, the one or more circuits to enable or disable the pump or compressor unit.17. The processor of claim 16, further comprising:an output to provide signals for at least one flow controller to enable flow of the two-phase fluid through the pump or compressor unit and to prevent flow of a secondary coolant to a secondary cooling loop of a datacenter cooling system.18. The processor of claim 16, further comprising:the one or more neural networks to receive the sensor inputs and to be trained to infer a first cooling requirement associated with a secondary cooling loop and a second cooling requirement associated with the pump or compressor unit based in part on an analysis of prior sensor inputs and prior cooling requirements.19. The processor of claim 16, further comprising:an output to provide signals to cause one or more of the pump or compressor unit or a secondary cooling loop to be adjusted to address different cooling requirements.20. A processor of a control unit within a rack, the processor comprising one or more circuits, the one or more circuits to comprise one or more neural networks to infer, from sensor inputs of sensors associated with at least one computing device that a cooling requirement exists, the processor to cause a response to the cooling requirement, the control unit comprising a pump or compressor unit to enable two-phase fluid to flow from a cold plate coupled with the at least one computing device to a heat exchanger integrated with a rear door of the rack for dissipation of heat from the at least one computing device through the heat exchanger, the processor further comprising an input to receive the sensor inputs associated with a temperature from the at least one computing device, secondary coolant, or the two-phase fluid, the one or more neural networks trained to infer a change in coolant state has occurred based in part on the temperature and on prior temperatures in the rear door heat exchanger, the one or more circuits to enable or disable the pump or compressor unit.21. The processor of claim 20, further comprising:an output to provide signals for at least one flow controller to enable flow of the two-phase fluid through the pump or compressor unit and to prevent flow of a secondary coolant to a secondary cooling loop of a datacenter cooling system.22. The processor of claim 20, further comprising:the one or more neural networks to receive the sensor inputs and to be trained to infer a first cooling requirement associated with a secondary cooling loop and a second cooling requirement associated with the pump or compressor unit based in part on an analysis of prior sensor inputs and prior cooling requirements.23. The processor of claim 20, further comprising:an output to provide signals to cause one or more of the pump or compressor unit, or a secondary cooling loop to be adjusted to address different cooling requirements.24. A method for datacenter cooling system, comprising:providing a control unit within a rack to comprise a pump or compressor unit, the control unit comprising an input to receive sensor inputs associated with a temperature from at least one computing device, secondary fluid, or two-phase fluid;determining cooling requirements for the at least one computing device of a rack by one or more neural networks trained to infer a change in coolant state has occurred based in part on the temperature and on prior temperatures in a rear door heat exchanger;enabling at least one cold plate to absorb heat from the at least one computing device using the two-phase fluid; andenabling, by the control unit, the pump or compressor unit to cause the two-phase fluid to flow from the at least one cold plate to a heat exchanger integrated with a rear door of the rack for dissipation of heat from the at least one computing device through the heat exchanger.25. The method of claim 24, further comprising:determining, using at least one processor, a temperature associated with the at least one computing device in the rack;determining a first cooling requirement or a second cooling requirement using the temperature; andcausing, based in part on the first cooling requirement or the second cooling requirement, the pump or compressor unit or a secondary cooling loop to cool the at least one computing device.26. The method of claim 25, further comprising:receiving, in at least one processor, sensor inputs from sensors associated with the at least one computing device, the rack, a secondary coolant, or the two-phase fluid; anddetermining, using the at least one processor, the first cooling requirement and the second cooling requirement based in part on the sensor inputs.27. The method of claim 24, further comprising:enabling the pump or compressor unit to dissipate heat to a hot aisle within the datacenter.28. The method of claim 24, further comprising:receiving, by at least one processor, sensor inputs from sensors associated with the at least one computing device;determining, by the at least one processor, a change in a coolant state based in part on the sensor inputs; andcausing, based in part on the change in the coolant state, the pump or compressor unit to cause cooling of the at least one computing device.
微信群二維碼
意見反饋