Liquid cooling heat dissipation dynamic flow control method and device, medium and product

Abstract

The invention discloses a liquid cooling heat dissipation dynamic flow control method and device, a medium and a product, and relates to the technical field of liquid cooling heat dissipation. According to the scheme, the corresponding secondary liquid cooling distribution unit is arranged for each computing device in the liquid cooling heat dissipation system, the load information of each component of the corresponding computing device can be collected, the sum of the cooling liquid flow required by the corresponding computing device in the next collection period is determined accordingly, and the corresponding cooling liquid flow is controlled; and meanwhile, the secondary liquid cooling distribution unit reports the corresponding cooling liquid flow sum to the liquid cooling distribution unit, so that the liquid cooling distribution unit controls the total flow of the cooling liquid according to the cooling liquid flow sum corresponding to each computing device, and the overall regulation and control of the flow are realized. Compared with the prior art, more sensors do not need to be arranged and controlled, the cost is reduced, liquid cooling dynamic flow control can be achieved in combination with the specific service load condition, meanwhile, the cooling liquid flow can be dynamically adjusted according to the difference of heat dissipation pipelines of different device parts, and flow part-level refined control is achieved.

Description

Technical Field

[0001] The present invention relates to the field of liquid cooling and heat dissipation technology, and in particular to a liquid cooling and heat dissipation dynamic flow control method, equipment, medium and product. Background Art

[0002] Cold plate liquid cooling technology, with its efficient heat transfer characteristics through direct contact with the heat source, has become a key solution for high-density computing infrastructure. To achieve closed-loop control of coolant flow, multiple types of sensors are currently deployed to monitor system status in real time, combined with coordinated control of solenoid valves and pumps for dynamic regulation.

[0003] However, while this approach enables real-time status monitoring, the sensors experience delays in their temperature control response. Furthermore, after data is reported, the main control program also has a response cycle, making flow control unable to accurately achieve the preset target. Furthermore, the liquid cooling distribution units (CDUs) utilize a centralized liquid cooling distribution solution, which cannot provide precise control of liquid cooling heat dissipation at the component level, resulting in insufficiently precise heat dissipation. Furthermore, the need for more and more sophisticated sensors further increases the cost of cold plate liquid cooling, which already requires high costs, resulting in a low cost-performance ratio.

[0004] In view of the above, how to solve the current liquid cooling distribution method faced with sensor delay, response cycle of the main control program, inaccurate flow control, centralized liquid cooling distribution resulting in insufficient heat dissipation and high cost is an urgent problem that technical personnel in this field need to solve. Summary of the Invention

[0005] The present invention provides a method, device, medium and product for dynamic flow control of liquid cooling heat dissipation, so as to at least solve the problems faced by the current liquid cooling distribution method, such as sensor delay, response cycle of the main control program, inaccurate flow control, insufficient heat dissipation due to centralized liquid cooling distribution, and high cost.

[0006] The present invention provides a liquid cooling and heat dissipation dynamic flow control method, which is applied to a main control module of a secondary liquid cooling distribution unit; wherein the liquid cooling and heat dissipation system includes multiple secondary liquid cooling distribution units, each of which corresponds to each computing device; the secondary liquid cooling distribution unit also includes a flow control component; the main control module is communicatively connected with the corresponding flow control component and each component of the corresponding computing device; the flow control component and the cold plate pipeline of the corresponding computing device are connected to the liquid distributor via a pipeline; the method includes: Determining a collection period for each component of the corresponding computing device, and collecting load information of the corresponding component according to each collection period; Determine the coolant flow required by each component in the next acquisition cycle according to each load information; summing the coolant flow rates to determine the total coolant flow rate required by the corresponding computing device; The coolant flow of the corresponding flow control component is controlled according to the total coolant flow, and the total coolant flow is reported to the liquid cooling distribution unit so that the liquid cooling distribution unit can control the total coolant flow according to the total coolant flow corresponding to each computing device.

[0007] The present invention also provides an electronic device, comprising: a memory for storing a computer program; and a processor for implementing the steps of any of the above-mentioned liquid cooling dynamic flow control methods when executing the computer program.

[0008] The present invention also provides a computer-readable storage medium, in which a computer program is stored, wherein when the computer program is executed by a processor, the steps of any of the above-mentioned liquid cooling dynamic flow control methods are implemented.

[0009] The present invention also provides a computer program product, comprising a computer program, which implements the steps of any of the above-mentioned liquid cooling dynamic flow control methods when executed by a processor.

[0010] The beneficial effect of the present invention is that a corresponding secondary liquid cooling distribution unit is set for each computing device in the liquid cooling and heat dissipation system, which can collect the load information of each device component of the corresponding computing device, determine the total coolant flow required by the corresponding computing device in the next collection cycle based on each load information, and control the corresponding coolant flow; at the same time, the secondary liquid cooling distribution unit can report the corresponding coolant flow sum to the liquid cooling distribution unit, so that the liquid cooling distribution unit controls the total coolant flow according to the total coolant flow corresponding to each computing device, thereby realizing the overall regulation of the coolant flow. This solution does not require the deployment of more sensors for the liquid cooling and heat dissipation system, reduces costs, can realize dynamic liquid cooling flow control in combination with specific business load conditions, and can also dynamically adjust the coolant flow according to the differences in the heat dissipation pipelines of different devices, realizing refined flow component-level control.

[0011] In addition, the present invention also provides a liquid cooling dynamic flow control device, medium and product, with the same effect as above. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0013] Figure 1A schematic diagram of a liquid cooling and heat dissipation system including a secondary liquid cooling distribution unit provided in an embodiment of the present invention; Figure 2 A schematic structural diagram of a secondary liquid cooling distribution unit provided in an embodiment of the present invention; Figure 3 A flow chart of a method for dynamic flow control of liquid cooling provided by an embodiment of the present invention; Figure 4 A schematic diagram of a liquid cooling dynamic flow control device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0014] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0015] It should be noted that, in the description of the present invention, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. The terms "first," "second," etc., in the present invention are used to distinguish similar objects, and are not used to describe a particular order or precedence.

[0016] In order to enable those skilled in the art to better understand the solutions of the present invention, the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0017] Figure 1 Schematic diagram of a liquid cooling system including a secondary liquid cooling distribution unit provided in an embodiment of the present invention. Figure 1 As shown, the liquid cooling system includes multiple sub-liquid cooling distribution units (Sub-CDUs), each of which corresponds to a computing device. In this embodiment, there is no limitation on the computing device, and it can be, for example, a server, a host, etc.

[0018] Figure 2 This is a schematic diagram of the structure of the secondary liquid cooling distribution unit provided in an embodiment of the present invention. Figure 2As shown, the secondary liquid cooling distribution unit is composed of a main control module and a flow control component. The main control module is in communication with the corresponding flow control component and the various components of the corresponding computing device; the flow control component, the cold plate pipeline of the corresponding computing device, and the liquid distributor are connected by pipelines. In this embodiment, there is no restriction on the specific composition of the flow control component. For example, it can be composed of a micro water pump and a flow regulating valve. The method provided by the present invention is applied to the main control module of the secondary liquid cooling distribution unit. The following is a detailed description of the liquid cooling heat dissipation dynamic flow control method in conjunction with specific embodiments: Figure 3 This is a flow chart of a method for controlling the dynamic flow of liquid cooling provided by an embodiment of the present invention. Figure 3 As shown, the method includes: S10: Determine a collection period of each component of the corresponding computing device, and collect load information of the corresponding component according to each collection period.

[0019] First, the main control module determines the collection period of each component of the corresponding computing device, and collects the load information of the corresponding component according to each collection period. It should be noted that the component in this embodiment refers to the component that requires liquid cooling, including but not limited to the central processing unit (CPU), memory, input / output (IO) devices and graphics processing unit (GPU) and other components; the collection period of each component is different, and it is necessary to comprehensively consider the change rules of the load of each component and adaptively change the collection period to provide more accurate load data. In this embodiment, there is no restriction on the specific method of determining the collection period of each component, which depends on the specific implementation situation. In addition, in this embodiment, there is no restriction on the specific load information that needs to be collected. For example, the load information that needs to be collected is shown in the following table: Table 1 Load information collection table

[0020] It is understandable that Table 1 merely provides an example of load information collection. In a specific implementation, more or less load information may be collected, depending on the specific implementation situation.

[0021] S11: Determine the coolant flow rate required by each component in the next corresponding acquisition cycle according to each load information.

[0022] S12: summing the coolant flow rates of the respective coolant flows to determine the total coolant flow rate required by the corresponding computing device.

[0023] Furthermore, based on the load information of each component, the coolant flow rate required by each component in the corresponding next acquisition cycle is calculated respectively, and the coolant flow rates are summed to determine the total coolant flow rate required by the corresponding computing device.

[0024] It should be noted that, in this embodiment, there is no limitation on the specific process of determining the cooling liquid flow rate required by each component in the corresponding next acquisition cycle, and the process depends on the specific implementation situation.

[0025] S13: Control the coolant flow of the corresponding flow control component according to the total coolant flow, and report the total coolant flow to the liquid cooling distribution unit so that the liquid cooling distribution unit can control the total coolant flow according to the total coolant flow corresponding to each computing device.

[0026] Finally, the main control module reports the total coolant flow of the corresponding computing devices to the liquid cooling distribution unit. At this time, the liquid cooling distribution unit will receive the total coolant flow corresponding to multiple computing devices. In order to execute the coolant flow control in the next cycle and achieve the purpose of fine-grained coolant flow control, the liquid cooling distribution unit will control the total coolant flow of the next cycle according to the total coolant flow corresponding to each computing device; at the same time, the main control module of each computing device will also control the coolant flow of the corresponding flow control component according to the corresponding total coolant flow, such as controlling the corresponding micro-water pump speed and flow regulating valve opening, so as to achieve fine-grained control of the coolant flow required by each component in the next cycle, and finally achieve the purpose of fine heat dissipation of each component, solve the problems of untimely flow response during liquid cooling, inconsistent heat dissipation modes of different components leading to heat dissipation failure, and component damage, ensure the stability and reliability of liquid-cooled data center production, and reduce the failure rate of components caused by liquid cooling.

[0027] It should be noted that, in this embodiment, the specific process of the liquid cooling distribution unit controlling the total cooling liquid flow rate according to the total cooling liquid flow rate corresponding to each computing device is not limited and depends on the specific implementation situation.

[0028] In this embodiment, a corresponding secondary liquid cooling distribution unit is provided for each computing device in the liquid cooling system. This unit can collect the load information of each component of the corresponding computing device, determine the total coolant flow required by the corresponding computing device in the next collection cycle based on the load information, and control the corresponding coolant flow accordingly. At the same time, the secondary liquid cooling distribution unit can report the corresponding coolant flow sum to the liquid cooling distribution unit, so that the liquid cooling distribution unit controls the total coolant flow according to the total coolant flow corresponding to each computing device, thereby achieving overall control of the coolant flow. This solution does not require the deployment of more sensors for the liquid cooling system, which reduces costs. It can achieve dynamic liquid cooling flow control based on specific business load conditions. It can also dynamically adjust the coolant flow according to the differences in the heat dissipation pipelines of different components, achieving refined flow control at the component level.

[0029] Based on the above embodiments, in some embodiments, determining the collection period of each component of the corresponding computing device includes: S101: Determine the initial acquisition period of each component.

[0030] Among them, each initial acquisition period is the same; the device components at least include a central processing unit, a memory, an input and output device and a graphics processing unit.

[0031] S102: Collecting initial load information of each component according to an initial collection period; S103: Determine the collection period of the corresponding component according to each initial load information.

[0032] To determine the collection period for each component, first identify the specific component type involved in liquid cooling. In this embodiment, these components include at least the CPU, memory, I / O devices, and GPU. Also, determine the initial collection period for each component. Note that the initial collection period is the same for each component. In this embodiment, there is no specific limit on the length of the initial collection period; for example, it can be 5 minutes.

[0033] Furthermore, initial load information of each component is collected based on the initial collection period, and then the collection period for the corresponding component is determined based on each initial load information. It should be noted that in order to comprehensively consider the changing patterns of the loads of each component and provide more accurate load data, the collection periods of each component determined based on each initial load information may not necessarily be the same.

[0034] Specifically, in order to determine the collection period of the corresponding device component according to each initial load information, the load standard deviation and the current load change rate of the corresponding device component are determined according to each initial load information; the basic period, the dynamic amplification factor, and the component weight value of each device component are obtained; and the collection period of each device component is determined according to the basic period, the dynamic amplification factor, the component weight value of each device component, the load standard deviation, and the current load change rate. The specific formula is as follows: ; in, is the acquisition period of the i-th device component (s); The default period is 1s, which must meet the minimum monitoring requirements. is the dynamic amplification factor, which is recommended to be 0.5~1.0 to balance the fluctuation sensitivity; is the load standard deviation of component i, specifically a normalized value of 0 or 1, reflecting historical volatility; is the current load change rate, specifically the absolute value; is the component weight value, see the following table for details: Table 2 Component weights and parameter reference table

[0035] As shown in Table 2 and the above formula, as the load standard deviation or preload change rate of a component increases, the numerator increases, the corresponding acquisition period shortens, and the component weight value reflects the priority. Furthermore, to achieve boundary control, a minimum period Tmin of 0.2s can be set to avoid excessive overhead, while a maximum period Tmax of 5s can be set to ensure real-time control.

[0036] In this embodiment, the initial load information of each component is collected according to the initial collection period, and the collection period of the corresponding component is determined based on each initial load information. The change law of the load of each component is comprehensively considered, and the collection period is adaptively changed to provide more accurate load data.

[0037] Based on the above embodiment, in some embodiments, determining the coolant flow required by each component in the corresponding next acquisition cycle according to each load information includes: S110: Determine the heat generated by each component in the next corresponding collection cycle according to each load information; S111: Determine the cooling liquid flow required by each component in the next collection cycle according to each heat amount.

[0038] In a specific implementation, in order to determine the coolant flow rate required by each component in the next acquisition cycle, it is first necessary to determine the heat generated by each component in the next acquisition cycle based on the load information, and then determine the coolant flow rate required by each component in the next acquisition cycle based on the heat. The specific process of determining the heat generated by each component in the next acquisition cycle is described in detail below with reference to a specific embodiment: (1) Determine the heat generated by the CPU during the next acquisition cycle; In some embodiments, determining the amount of heat generated by each component in a corresponding next acquisition cycle according to each load information includes: S112: Determine a first static power consumption, a first dynamic power consumption, a maximum utilization, a nonlinear correction coefficient, a core temperature, an ambient temperature, a reference temperature difference, a temperature amplification coefficient, and a first liquid cooling efficiency factor corresponding to the central processing unit.

[0039] S113: Determine the amount of heat generated by the central processing unit in a corresponding next acquisition cycle based on the first static power consumption, the first dynamic power consumption, the maximum utilization, the nonlinear correction coefficient, the core temperature, the ambient temperature, the reference temperature difference, the temperature amplification coefficient, the first liquid cooling efficiency factor, and the corresponding acquisition cycle corresponding to the central processing unit.

[0040] Specifically, to determine the amount of heat generated by the CPU during the next acquisition cycle, it is first necessary to determine the CPU's corresponding first static power consumption, first dynamic power consumption, maximum utilization, nonlinear correction coefficient, core temperature, ambient temperature, reference temperature difference, temperature amplification coefficient, and first liquid cooling efficiency factor. Based on the CPU's corresponding first static power consumption, first dynamic power consumption, maximum utilization, nonlinear correction coefficient, core temperature, ambient temperature, reference temperature difference, temperature amplification coefficient, first liquid cooling efficiency factor, and the corresponding acquisition cycle, the heat generated by the CPU during the next acquisition cycle is determined using the following formula: ; in, is the heat generated by the CPU during the corresponding acquisition period Ti (J); The first static power consumption (W) is related to the process and temperature, such as the power consumption caused by leakage current. is the first dynamic power consumption (W). is the maximum CPU utilization (100%); is the nonlinear correction coefficient corresponding to the CPU (1.1-1.5), reflecting the accelerated growth of power consumption under high load. is the CPU core temperature ( ); is the ambient temperature corresponding to the CPU ( ), specifically corresponding to the coolant inlet temperature (such as 40 ). is the reference temperature difference (such as 50 ), which is used to normalize the effect of temperature difference. and These are the temperature amplification coefficients corresponding to the CPU. The range is 0.2 to 0.5, The range of is 1.2 to 1.5, which represents the feedback enhancement effect of temperature difference on heat production; It is the first liquid cooling efficiency factor.

[0041] It should be noted that, in order to determine the first dynamic power consumption of the CPU, in a specific implementation, it is necessary to obtain the load capacitance, voltage, frequency, and current utilization of the CPU; based on the load capacitance, voltage, frequency, and current utilization of the CPU, the first dynamic power consumption of the CPU is determined using the following formula: ; in, is the first dynamic power consumption; is the load capacitance of the CPU; is the voltage of the CPU; is the CPU frequency, Current CPU utilization (user state + system state).

[0042] To determine the CPU core temperature, in a specific implementation, it is necessary to obtain the sensor temperature, heat transfer coefficient, and load intensity parameters set on the CPU. The CPU core temperature is determined based on the sensor temperature, heat transfer coefficient, and load intensity parameters. The specific formula is as follows: ; in, is the core temperature of the CPU, and These are the CPU sensor temperatures. and are the heat transfer coefficients, is the load intensity parameter, is the average temperature difference of the CPU load intensity (from zero load to full load).

[0043] To determine the first liquid cooling efficiency factor of the CPU, in a specific implementation, it is necessary to obtain the actual thermal resistance, design thermal resistance, and optimization parameters of the CPU. Based on the actual thermal resistance, design thermal resistance, and optimization parameters of the CPU, the first liquid cooling efficiency factor of the CPU is determined. The specific formula is as follows: ; in, The first liquid cooling efficiency factor for CPU, is the actual thermal resistance corresponding to the CPU ( ), is the design thermal resistance corresponding to the CPU, and These are the optimization parameters corresponding to the CPU. The range is 0.1 to 0.3, The range is 1.0 to 1.5.

[0044] (2) Determine the amount of heat generated by the memory corresponding to the next acquisition cycle; In some embodiments, determining the amount of heat generated by each component in a corresponding next acquisition cycle according to each load information includes: S114: Determine a second static power consumption, memory utilization, utilization amplification factor, second dynamic power consumption, page swap frequency, swap frequency nonlinear coefficient, basic swap power consumption, and second liquid cooling efficiency factor corresponding to the memory.

[0045] S115: Determine the amount of heat generated by the memory in the next acquisition cycle according to the second static power consumption, memory utilization, utilization amplification factor, second dynamic power consumption, page swap frequency, swap frequency nonlinear coefficient, basic swap power consumption, second liquid cooling efficiency factor, and corresponding acquisition cycle corresponding to the memory.

[0046] Specifically, to determine the amount of heat generated by the memory during the next acquisition cycle, it is first necessary to determine the memory's corresponding second static power consumption, memory utilization, utilization amplification factor, second dynamic power consumption, page swap frequency, swap frequency nonlinear coefficient, basic swap power consumption, and second liquid cooling efficiency factor. Further, based on the memory's corresponding second static power consumption, memory utilization, utilization amplification factor, second dynamic power consumption, page swap frequency, swap frequency nonlinear coefficient, basic swap power consumption, second liquid cooling efficiency factor, and the corresponding acquisition cycle, the amount of heat generated by the memory during the next acquisition cycle is determined using the following formula: ; in, is the heat generated by the memory during the corresponding acquisition period Ti (J). The second static power consumption (W) is related to the memory leakage current and is affected by temperature. For example, at high temperatures, the leakage current increases, resulting in increased power consumption. is the memory utilization; is the memory utilization magnification factor (0.1-0.3), which represents the accelerated growth effect of leakage current under high load; The second dynamic power consumption (W / GB / s) is related to the memory bandwidth Bi (GB / s) and reflects the energy consumption of data transmission. For example, the DDR4-3200 bandwidth of 34.1GB / s corresponds to a typical power consumption of approximately 0.35W / GB / s. The page swap frequency (times / s) is calculated using pswpin and pswpout in / proc / vmstat. is the nonlinear coefficient of the switching frequency (0.5-1.0), which represents the exponential power consumption growth caused by frequent switching. For example, when the switching frequency doubles, the power consumption growth becomes superlinear. Basic swap power consumption (W), reflecting the average energy consumption of a single page swap (e.g., 5-10W); It is the second liquid cooling efficiency factor.

[0047] To determine the memory utilization corresponding to the memory, in a specific implementation, it is necessary to obtain the total memory capacity, free memory capacity, and used memory capacity; based on the total memory capacity, free memory capacity, and used memory capacity, the memory utilization corresponding to the memory is determined. The specific formula is as follows: ; in, is the memory utilization corresponding to the memory, is the total capacity of the memory, is the free memory capacity of the memory, The used memory capacity of the memory.

[0048] In order to determine the second liquid-cooling heat dissipation efficiency factor corresponding to the memory, in a specific implementation, it is necessary to obtain the memory temperature, coolant inlet temperature, reference temperature difference, actual thermal resistance, design thermal resistance and correction coefficient corresponding to the memory; based on the memory temperature, coolant inlet temperature, reference temperature difference, actual thermal resistance, design thermal resistance and correction coefficient corresponding to the memory, determine the second liquid-cooling heat dissipation efficiency factor corresponding to the memory.

[0049] ; in, The second liquid cooling efficiency factor corresponding to the memory is is the memory temperature ( ), The ambient temperature corresponding to the memory, that is, the coolant inlet temperature (such as 40 ); is the reference temperature difference (50 ), is the actual thermal resistance corresponding to the memory ( ), is the design thermal resistance corresponding to the memory, 、 and These are the correction parameters corresponding to the memory. The range is 1.2 to 1.5, The range is 0.8 to 1.0, The range is 0.1 to 0.2.

[0050] (3) Determine the heat generated by the IO device during the next acquisition cycle; In some embodiments, determining the amount of heat generated by each component in a corresponding next acquisition cycle according to each load information includes: S116: Determine a third static power consumption, a third dynamic power consumption, and a third liquid cooling efficiency factor corresponding to the input / output device.

[0051] S117: Determine the amount of heat generated by the input / output device in a corresponding next acquisition cycle according to the third static power consumption, the third dynamic power consumption, the third liquid cooling efficiency factor, and the corresponding acquisition cycle corresponding to the input / output device.

[0052] Specifically, to determine the amount of heat generated by the IO device during the next acquisition cycle, it is first necessary to determine the third static power consumption, third dynamic power consumption, and third liquid cooling efficiency factor corresponding to the IO device. Further, based on the third static power consumption, third dynamic power consumption, third liquid cooling efficiency factor, and the corresponding acquisition cycle, the amount of heat generated by the IO device during the next acquisition cycle is determined using the following formula: ; in, In the corresponding next collection cycle The heat generated by the internal IO devices, is the third static power consumption, is the third dynamic power consumption, It is the third liquid cooling efficiency factor.

[0053] To determine the third static power consumption corresponding to the IO device, in a specific implementation, it is necessary to obtain the basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference, and temperature coefficient corresponding to the IO device. The third static power consumption corresponding to the IO device is determined based on the basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference, and temperature coefficient corresponding to the IO device. The specific formula is as follows: ; in, is the third static power consumption corresponding to the IO device, is the basic static power consumption of the IO device (W), is the core temperature of the IO device, is the coolant inlet temperature corresponding to the IO device, is the reference temperature difference, It is the temperature coefficient corresponding to the IO device.

[0054] To determine the third dynamic power consumption of an IO device, in a specific implementation, it is necessary to obtain the power consumption coefficient, read / write rate, input / output queue depth, device busy rate, and busy rate nonlinear coefficient corresponding to the IO device. The third dynamic power consumption corresponding to the IO device is determined based on the power consumption coefficient, read / write rate, input / output queue depth, device busy rate, and busy rate nonlinear coefficient corresponding to the IO device. The specific formula is as follows: ; in, The third dynamic power consumption of the IO device (W), is the power consumption coefficient of the IO device (W / MB / s), is the read and write rate of the IO device (MB / s), The input and output queue depth of the IO device (such as the maximum queue depth of NVMe SSD is 64K). The device busy rate of the IO device (%util), The nonlinear coefficient of the busy rate of the I / O device (0.5~1.0).

[0055] To determine the third liquid cooling efficiency factor for an I / O device, the actual thermal resistance, design thermal resistance, and correction factor of the IO device must be obtained. The third liquid cooling efficiency factor for the IO device is determined based on the actual thermal resistance, design thermal resistance, correction factor, core temperature, coolant inlet temperature, and reference temperature difference. The formula is as follows: ; in, The third liquid cooling efficiency factor of the IO device is is the actual thermal resistance corresponding to the IO device, is the design thermal resistance, 、 and are correction coefficients, The range is 0.1 to 0.3, The range is 1.0 to 1.5, The range is 0.05 to 0.1.

[0056] (4) Determine the heat generated by the GPU in the next acquisition cycle; In some embodiments, determining the amount of heat generated by each component in a corresponding next acquisition cycle according to each load information includes: S118: Determine a fourth static power consumption, a fourth dynamic power consumption, and a fourth liquid cooling efficiency factor corresponding to the graphics processing unit.

[0057] S119: Determine the amount of heat generated by the graphics processing unit in a corresponding next acquisition cycle according to the fourth static power consumption, the fourth dynamic power consumption, the fourth liquid cooling efficiency factor, and the corresponding acquisition cycle corresponding to the graphics processing unit.

[0058] Specifically, to determine the amount of heat generated by the GPU during the next acquisition cycle, it is first necessary to determine the fourth static power consumption, fourth dynamic power consumption, and fourth liquid cooling efficiency factor corresponding to the GPU. Further, based on the fourth static power consumption, fourth dynamic power consumption, fourth liquid cooling efficiency factor, and the corresponding acquisition cycle, the amount of heat generated by the GPU during the next acquisition cycle is determined using the following formula: ; in, To correspond to the heat generated by the GPU in the next acquisition cycle, is the fourth static power consumption corresponding to the GPU, is the fourth dynamic power consumption, It is the fourth liquid cooling efficiency factor.

[0059] To determine the fourth static power consumption corresponding to the GPU, in a specific implementation, it is necessary to obtain the basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference, and temperature coefficient corresponding to the GPU; based on the basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference, and temperature coefficient corresponding to the GPU, the fourth static power consumption corresponding to the input and output devices is determined. The specific formula is as follows: ; in, is the fourth static power consumption corresponding to the GPU, is the basic static power consumption of the GPU (W), is the core temperature of the GPU, is the coolant inlet temperature corresponding to the GPU, is the reference temperature difference, is the temperature coefficient corresponding to the GPU.

[0060] In order to determine the fourth dynamic power consumption corresponding to the GPU, in a specific implementation, it is necessary to obtain the video memory utilization, computing unit utilization, task queue length, video memory nonlinear coefficient, task queue nonlinear coefficient, and power consumption coefficient corresponding to the GPU; based on the video memory utilization, computing unit utilization, task queue length, video memory nonlinear coefficient, task queue nonlinear coefficient, and power consumption coefficient corresponding to the GPU, the fourth dynamic power consumption corresponding to the input and output devices is determined. The specific formula is as follows: ; in, is the fourth dynamic power consumption corresponding to the GPU, The GPU memory utilization (used video memory / total video memory) reflects the data transmission load. The GPU computing unit utilization (%), obtained through nvidia-smi or ROCmSMI. is the GPU task queue length (number of cores to be processed), which indicates the degree of task accumulation. It is the nonlinear coefficient of GPU memory, reflecting the accelerated growth of power consumption under high load. is the nonlinear coefficient of the GPU task queue, which represents the exponential growth of power consumption when the queue is saturated; is the power consumption coefficient of the GPU (W), which needs to be calibrated through actual measurement.

[0061] In order to determine the fourth liquid cooling efficiency factor of the GPU, in a specific implementation, it is necessary to obtain the actual thermal resistance, design thermal resistance, correction factor, actual coolant flow rate, design flow rate, and flow compensation factor corresponding to the GPU; based on the actual thermal resistance, design thermal resistance, correction factor, actual coolant flow rate, design flow rate, and flow compensation factor corresponding to the GPU, the fourth liquid cooling efficiency factor corresponding to the GPU is determined using the following formula: ; in, It is the fourth liquid cooling efficiency factor of GPU. is the actual thermal resistance of the GPU ( ), affected by the cold plate flow channel design; is the GPU's designed thermal resistance ( ), cold plate optimization target value. and are correction coefficients, The range is 0.1 to 0.3, The range is 1.0 to 1.5. The actual coolant flow rate (L / min) corresponding to the GPU is obtained through the CDU sensor. The design flow rate (L / min) corresponding to the GPU meets the maximum heat load requirement. is the flow compensation coefficient, which characterizes the inhibitory effect of flow reduction on heat dissipation efficiency, and its value range is 0.05 to 0.1.

[0062] In summary, the coolant flow rates of the CPU, memory, IO devices, and GPU corresponding to the next acquisition cycle are obtained respectively. 、 、 and , taking into account the specific load conditions of each component, the accurate acquisition of the corresponding heat is achieved, so that the coolant flow required by each component in the corresponding next acquisition cycle can be determined based on the heat.

[0063] Based on the above embodiment, in some embodiments, determining the cooling liquid flow required by each component in the corresponding next acquisition cycle according to each heat includes: S121: Determine the theoretical coolant flow rate required by each component in the next acquisition cycle according to the heat and thermodynamic energy conservation law.

[0064] S122: Obtaining the flow redundancy coefficient, pipeline resistance correction coefficient, pipeline total resistance value, and design resistance target value.

[0065] S123: Determine the actual coolant flow required by each component in the next acquisition cycle according to the flow redundancy coefficient, the pipeline resistance correction coefficient, the total pipeline resistance value, the design resistance target value, and each theoretical coolant flow rate.

[0066] S124: Setting the target speed of the water pump and the opening of the flow control valve of the flow control component according to the actual coolant flow rate.

[0067] S125: Determine the coolant flow required by each component in the next acquisition cycle according to the actual coolant flow, the target water pump speed, and the flow control valve opening.

[0068] After obtaining each heat, the theoretical coolant flow rate required by each component in the corresponding next acquisition cycle is determined based on each heat and the law of conservation of thermodynamic energy. Specifically, based on the law of conservation of thermodynamic energy, the theoretical coolant flow rate satisfy: ; in, is the theoretical coolant flow rate, is the coolant density; is the specific heat capacity of the coolant; is the maximum allowable temperature rise; is the collection period corresponding to the device component.

[0069] The theoretical coolant flow rate is further corrected for redundancy and system losses. Specifically, the flow redundancy coefficient, pipeline resistance correction coefficient, total pipeline resistance value, and design resistance target value are obtained. Based on the flow redundancy coefficient, pipeline resistance correction coefficient, total pipeline resistance value, design resistance target value, and each theoretical coolant flow rate, the actual coolant flow rate required by each component in the corresponding next acquisition cycle is determined. The specific formula is as follows: ; in, is the actual coolant flow rate, is the flow redundancy coefficient, is the pipeline resistance correction coefficient, is the total resistance of the pipeline, is the design resistance target value.

[0070] Then adjust the relationship between the pump speed and flow rate in the flow control component. The pump flow rate is proportional to the speed, and the formula is as follows: ; in, is the pump flow rate, is the rated flow rate, is the rated speed, is the current speed. In this embodiment, it is necessary to ensure , and the working point is in the high-efficiency area of the pump (70%-110% rated flow).

[0071] At the same time, the actual flow rate is adjusted by the valve opening of the flow control valve ,satisfy: ; in, For valve characteristic functions (such as linear or equal percentage characteristics), V needs to be dynamically adjusted to match In this way, the target speed of the water pump and the opening of the flow control valve of the flow control component can be set according to the actual coolant flow rate.

[0072] Finally, considering the above factors, the required coolant flow rate of each component in the next acquisition cycle is determined according to the actual coolant flow rate, the target speed of the water pump, and the opening of the flow control valve. The specific formula is as follows: ; in, is the coolant flow rate required by the device components in the corresponding next acquisition cycle, is the actual coolant flow rate, The target speed of the water pump must satisfy ; The opening of the flow control valve is dynamically adjusted by temperature feedback.

[0073] In this embodiment, by setting the water pump speed and the flow regulating valve opening to control the flow rate in the next cycle time to maintain the coolant flow, the accuracy and timeliness of the liquid cooling heat dissipation of the cold plates of each component can be guaranteed.

[0074] Correspondingly, the coolant flow required by each component is summarized, and the total coolant flow required by the computing equipment in the next cycle is calculated and sent to the CDU. The formula is as follows: ; in, is the total coolant flow corresponding to a single computing device, The coolant flow rate required for each component (CPU, GPU, memory, IO devices, etc.) in the next acquisition cycle.

[0075] It should also be noted that after determining the total coolant flow corresponding to a single computing device After that, it is necessary to determine the overall dynamic collection cycle of the computing device based on the next collection cycle / load change rules corresponding to each component (CPU, GPU, memory, IO devices, etc.) It is understandable that when the computing device is in the next dynamic collection cycle When the corresponding components (CPU, GPU, memory, IO devices, etc.) are in the corresponding next collection cycle In this embodiment, for the dynamic acquisition cycle The specific determination process is not limited.

[0076] Based on the above embodiment, in some embodiments, the liquid cooling distribution unit controls the total cooling liquid flow rate according to the total cooling liquid flow rate corresponding to each computing device, including: S131: Obtaining a reference period, a global redundancy coefficient, a computing device priority coefficient, and a dynamic collection period corresponding to each computing device; S132: Determine the total coolant flow rate according to the reference period, the global redundancy coefficient, the computing device priority coefficient, the dynamic collection period corresponding to each computing device, and the total coolant flow rate.

[0077] The CDU calculates the total coolant flow required for the next cycle based on the dynamic flow requirements reported by different computing devices within the cycle. This can be combined with dynamic load prediction, redundancy design, and real-time feedback mechanisms. The specific formula is as follows: ; in, is the total coolant flow rate, is the dynamic collection period corresponding to the i-th computing device (s), predicted in real time by load fluctuation rate (such as GPU memory utilization change rate). The reference period (e.g., 1 second) is used to standardize traffic demands at different periods. The flow rate required by each computing device in the next dynamic collection cycle, that is, the total coolant flow rate; The global redundancy factor (10% to 20% is recommended) compensates for the cumulative effect of load fluctuations on multiple servers. Prioritize critical servers (0.1-0.3) to reserve additional traffic for high-priority tasks (such as AI training clusters).

[0078] It should be noted that the above formula can be used to calculate the total coolant flow rate corresponding to each computing device. According to its cycle Adjusted to base period .like , that is, high-frequency monitoring scenarios, need to increase instantaneous traffic to cope with rapid load changes, such as a sudden increase in AI training tasks; if , i.e. low-frequency monitoring scenario, can reduce instantaneous flow, but requires extending the cooling duration.

[0079] In this embodiment, the total flow of the CDU in the next cycle is dynamically calculated to meet the cooling needs of multiple computing devices, thereby achieving refined control of the cooling flow of each computing device.

[0080] Based on the above embodiment, in some embodiments, after determining the total flow rate of the coolant, the method further includes: S133: Obtain the flow deviation record, response sensitivity coefficient, reference period, global redundancy coefficient and reference flow in the historical period.

[0081] S134: Determine the redundant flow rate according to the flow deviation record, the response sensitivity coefficient, the reference period, the global redundancy coefficient, the reference flow rate, and the dynamic collection period corresponding to each computing device and the total coolant flow rate.

[0082] S135: Redundancy correction is performed on the total coolant flow according to the redundant flow.

[0083] In order to avoid the problem of insufficient coolant flow when the load suddenly increases and the coolant flow needs to be increased instantaneously, after determining the total coolant flow, it is necessary to further obtain the flow deviation records, response sensitivity coefficient, benchmark cycle, global redundancy coefficient and benchmark flow in the historical period. Based on the flow deviation records, response sensitivity coefficient, benchmark cycle, global redundancy coefficient, benchmark flow and the collection cycle and coolant flow sum corresponding to each computing device, the redundant flow is determined. The specific formula is as follows: ; ; in, For redundant traffic; It is the flow deviation record in the historical period. is the response sensitivity coefficient. is the dynamic collection period corresponding to the i-th computing device (s); is the benchmark period, used to standardize the flow requirements of different periods; The required flow rate of each computing device in the next cycle; is the global redundancy coefficient (10%~20% is recommended, 20%) to compensate for the superposition effect of load fluctuations on multiple servers. is the benchmark flow, the average standard flow during the benchmark period.

[0084] Finally, the total coolant flow is corrected based on the redundant flow, thereby avoiding the problem of insufficient coolant flow when the load suddenly increases and the coolant flow needs to be increased instantaneously.

[0085] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method.

[0086] Figure 4 A schematic diagram of a liquid cooling and heat dissipation dynamic flow control device provided by an embodiment of the present invention. The device is applied to the main control module of a secondary liquid cooling distribution unit; wherein the liquid cooling and heat dissipation system includes multiple secondary liquid cooling distribution units, each of which corresponds to a computing device; the secondary liquid cooling distribution unit also includes a flow control component; the main control module is in communication with the corresponding flow control component and the various components of the corresponding computing device; the flow control component, the cold plate pipeline of the corresponding computing device, and the liquid distributor are connected via pipelines; Figure 4 As shown, the device includes: The collection module 10 is used to determine the collection period of each component of the corresponding computing device and collect the load information of the corresponding component according to each collection period; The first determining module 11 is used to determine the coolant flow rate required by each component in the next acquisition cycle according to each load information; A second determining module 12 is configured to sum the coolant flow rates to determine the total coolant flow rate required by the corresponding computing device; The control module 13 is used to control the coolant flow of the corresponding flow control component according to the total coolant flow, and report the total coolant flow to the liquid cooling distribution unit so that the liquid cooling distribution unit can control the total coolant flow according to the total coolant flow corresponding to each computing device.

[0087] In some embodiments, the acquisition module 10 includes: The first determining submodule is configured to determine an initial acquisition period of each device component; wherein each initial acquisition period is the same; the device component includes at least a central processing unit, a memory, an input / output device, and a graphics processing unit; A first acquisition submodule is configured to acquire initial load information of each component according to an initial acquisition period; The second determining submodule is configured to determine a collection period of a corresponding component according to each initial load information.

[0088] In some embodiments, the second determining submodule includes: a third determining submodule, configured to determine a load standard deviation and a current load change rate of a corresponding component according to each initial load information; A first acquisition submodule is used to obtain a basic period, a dynamic amplification factor and a component weight value of each device component; The fourth determination submodule is configured to determine the collection period of each component according to the basic period, the dynamic amplification factor, the component weight value of each component, the load standard deviation, and the current load change rate.

[0089] In some embodiments, the first determining module 11 includes: a fifth determining submodule, configured to determine the amount of heat generated by each component in a corresponding next acquisition cycle according to each load information; The sixth determining submodule is configured to determine the cooling liquid flow required by each component in the next corresponding acquisition cycle according to each heat amount.

[0090] In some embodiments, the fifth determining submodule includes: A CPU parameter determination module, configured to determine a first static power consumption, a first dynamic power consumption, a maximum utilization rate, a nonlinear correction coefficient, a core temperature, an ambient temperature, a reference temperature difference, a temperature amplification coefficient, and a first liquid cooling efficiency factor corresponding to the central processing unit; A CPU heat determination module is configured to determine the heat generated by the central processing unit in a corresponding next collection cycle based on the first static power consumption, the first dynamic power consumption, the maximum utilization rate, the nonlinear correction coefficient, the core temperature, the ambient temperature, the reference temperature difference, the temperature amplification coefficient, the first liquid cooling efficiency factor, and the corresponding collection cycle corresponding to the central processing unit; A memory parameter determination module is used to determine a second static power consumption, memory utilization, utilization amplification factor, second dynamic power consumption, page swap frequency, swap frequency nonlinear coefficient, basic swap power consumption, and second liquid cooling efficiency factor corresponding to the memory; A memory heat determination module is configured to determine the amount of heat generated by the memory in a corresponding next collection cycle based on the second static power consumption, memory utilization, utilization amplification factor, second dynamic power consumption, page swap frequency, swap frequency nonlinear coefficient, basic swap power consumption, second liquid cooling efficiency factor, and corresponding collection cycle corresponding to the memory; An IO device parameter determination module, configured to determine a third static power consumption, a third dynamic power consumption, and a third liquid cooling efficiency factor corresponding to an input / output device; an IO device heat determination module, configured to determine the amount of heat generated by the input / output device in a corresponding next collection cycle based on the third static power consumption, the third dynamic power consumption, the third liquid cooling efficiency factor, and the corresponding collection cycle corresponding to the input / output device; A GPU parameter determination module, configured to determine a fourth static power consumption, a fourth dynamic power consumption, and a fourth liquid cooling efficiency factor corresponding to the graphics processing unit; The GPU heat determination module is used to determine the heat generated by the graphics processing unit in the next corresponding acquisition cycle based on the fourth static power consumption, fourth dynamic power consumption, fourth liquid cooling efficiency factor and corresponding acquisition cycle corresponding to the graphics processing unit.

[0091] In some embodiments, the CPU parameter determination module includes: The second acquisition submodule is used to obtain the load capacitance, voltage, frequency and current utilization rate of the central processing unit; a seventh determining submodule, configured to determine a first dynamic power consumption of the central processing unit according to a load capacitance, a voltage, a frequency, and a current utilization rate of the central processing unit; The third acquisition submodule is used to obtain the sensor temperature, heat transfer coefficient, load intensity parameter and load intensity average temperature difference set on the central processing unit; an eighth determination submodule, configured to determine a core temperature of the central processing unit based on the sensor temperature, the heat transfer coefficient, the load intensity parameter, and the load intensity average temperature difference; The fourth acquisition submodule is used to obtain the actual thermal resistance, designed thermal resistance and optimization parameters of the central processing unit; The ninth determination submodule is configured to determine a first liquid cooling efficiency factor of the central processing unit according to the actual thermal resistance, the designed thermal resistance, and the optimization parameters of the central processing unit.

[0092] In some embodiments, the memory parameter determination module includes: The fifth acquisition submodule is used to obtain the total capacity, free memory capacity and used memory capacity of the memory; a tenth determining submodule, configured to determine a memory utilization rate corresponding to the memory according to the total capacity, free memory capacity, and used memory capacity of the memory; A sixth acquisition submodule is used to obtain the memory temperature, coolant inlet temperature, reference temperature difference, actual thermal resistance, design thermal resistance and correction coefficient corresponding to the memory; The eleventh determination submodule is used to determine the second liquid cooling heat dissipation efficiency factor corresponding to the memory according to the memory temperature corresponding to the memory, the coolant inlet temperature, the reference temperature difference, the actual thermal resistance, the design thermal resistance and the correction coefficient.

[0093] In some embodiments, the IO device parameter determination module includes: A seventh acquisition submodule is used to obtain basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference and temperature coefficient corresponding to input and output devices; a twelfth determining submodule, configured to determine a third static power consumption corresponding to the input / output device according to the basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference, and temperature coefficient corresponding to the input / output device; An eighth acquisition submodule is used to obtain the power consumption coefficient, read / write rate, input / output queue depth, device busy rate, and busy rate nonlinear coefficient corresponding to the input / output device; A thirteenth determination submodule is configured to determine a third dynamic power consumption corresponding to the input / output device according to the power consumption coefficient, read / write rate, input / output queue depth, device busy rate, and busy rate nonlinear coefficient corresponding to the input / output device; A ninth acquisition submodule is used to obtain the actual thermal resistance, design thermal resistance and correction coefficient corresponding to the input and output devices; The fourteenth determination submodule is used to determine the third liquid cooling heat dissipation efficiency factor corresponding to the input and output devices according to the actual thermal resistance, design thermal resistance, correction coefficient, core temperature, coolant inlet temperature and reference temperature difference corresponding to the input and output devices.

[0094] In some embodiments, the GPU parameter determination module includes: a tenth acquisition submodule, configured to acquire basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference, and temperature coefficient corresponding to the graphics processing unit; A fifteenth determining submodule is configured to determine a fourth static power consumption corresponding to the input / output device based on the basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference, and temperature coefficient corresponding to the graphics processing unit; An eleventh acquisition submodule is used to obtain the video memory utilization, computing unit utilization, task queue length, video memory nonlinear coefficient, task queue nonlinear coefficient and power consumption coefficient corresponding to the graphics processing unit; A sixteenth determining submodule is configured to determine a fourth dynamic power consumption corresponding to the input / output device based on a video memory utilization rate, a computing unit utilization rate, a task queue length, a video memory nonlinear coefficient, a task queue nonlinear coefficient, and a power consumption coefficient corresponding to the graphics processing unit; A twelfth acquisition submodule is used to obtain the actual thermal resistance, the designed thermal resistance, the correction coefficient, the actual coolant flow rate, the designed flow rate and the flow compensation coefficient corresponding to the graphics processing unit; The seventeenth determination submodule is used to determine the fourth liquid cooling efficiency factor corresponding to the graphics processing unit according to the actual thermal resistance, design thermal resistance, correction coefficient, actual coolant flow, design flow and flow compensation coefficient corresponding to the graphics processing unit.

[0095] In some embodiments, the sixth determining submodule includes: An eighteenth determination submodule is used to determine the theoretical coolant flow rate required by each component in the next acquisition cycle according to the heat and thermodynamic energy conservation law; The thirteenth acquisition submodule is used to obtain the flow redundancy coefficient, the pipeline resistance correction coefficient, the total pipeline resistance value and the design resistance target value; A nineteenth determination submodule is used to determine the actual coolant flow rate required by each component in the next acquisition cycle according to the flow redundancy coefficient, the pipeline resistance correction coefficient, the total pipeline resistance value, the design resistance target value and each theoretical coolant flow rate; A setting module is used to set the target speed of the water pump and the opening of the flow control valve of the flow control component according to the actual coolant flow rate; The twentieth determination submodule is used to determine the coolant flow required by each component in the corresponding next acquisition cycle according to each actual coolant flow, the target speed of the water pump and the opening of the flow control valve.

[0096] In some embodiments, the liquid cooling distribution unit controls the total coolant flow rate based on the sum of the coolant flow rates corresponding to each computing device, including: obtaining a reference period, a global redundancy coefficient, a computing device priority coefficient, and a dynamic acquisition period corresponding to each computing device; determining the total coolant flow rate based on the reference period, the global redundancy coefficient, the computing device priority coefficient, the dynamic acquisition period corresponding to each computing device, and the sum of the coolant flow rates.

[0097] In some embodiments, the liquid cooling distribution unit also obtains the flow deviation records, response sensitivity coefficient, reference period, global redundancy coefficient and reference flow within the historical period; determines the redundant flow based on the flow deviation records, response sensitivity coefficient, reference period, global redundancy coefficient, reference flow and the dynamic collection period and coolant flow sum corresponding to each computing device; and performs redundant correction on the total coolant flow based on the redundant flow.

[0098] For the description of the features in the embodiment corresponding to the liquid cooling dynamic flow control device, please refer to the relevant description of the embodiment corresponding to the liquid cooling dynamic flow control method, and will not be repeated here.

[0099] An embodiment of the present invention further provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any of the above-mentioned liquid cooling dynamic flow control method embodiments.

[0100] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program, wherein the computer program is configured to execute the steps of any of the above-mentioned liquid cooling dynamic flow control method embodiments when running.

[0101] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0102] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of any of the above-mentioned liquid cooling dynamic flow control method embodiments are implemented.

[0103] An embodiment of the present invention also provides another computer program product, including a non-volatile computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of any of the above-mentioned liquid cooling dynamic flow control method embodiments.

[0104] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0105] The above is a detailed introduction to the liquid cooling heat dissipation dynamic flow control method, equipment, medium and product provided by the present invention. This article uses specific examples to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present invention, the present invention can also be improved and modified in several ways, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. A liquid cooling dynamic flow control method, characterized in that: A main control module is applied to a secondary liquid cooling distribution unit; wherein the liquid cooling heat dissipation system includes a plurality of said secondary liquid cooling distribution units, each of said secondary liquid cooling distribution units corresponding to each computing device; said secondary liquid cooling distribution unit also includes a flow control component; said main control module is in communication with said flow control component and each component of said computing device; said flow control component and a cold plate pipeline of said computing device are connected to a liquid distributor via a pipeline; said method comprises: Determining a collection period for each component of the corresponding computing device, and collecting load information of the corresponding component according to each collection period; Determining the coolant flow required by each component in the corresponding next acquisition cycle according to each load information; summing the coolant flow rates to determine the total coolant flow rate required by the corresponding computing device; The coolant flow of the corresponding flow control component is controlled according to the total coolant flow, and the total coolant flow is reported to the liquid cooling distribution unit so that the liquid cooling distribution unit can control the total coolant flow according to the total coolant flow corresponding to each computing device.

2. The liquid cooling heat dissipation dynamic flow control method according to claim 1, characterized in that: Determine the collection period of each component of the corresponding computing device, including: Determining an initial acquisition period for each of the device components; wherein each of the initial acquisition periods is the same; the device components at least include a central processing unit, a memory, an input and output device, and a graphics processing unit; Collecting initial load information of each of the device components according to the initial collection period; The acquisition period of the corresponding device component is determined according to each of the initial load information.

3. The liquid cooling heat dissipation dynamic flow control method according to claim 2, characterized in that: Determining the acquisition period of the corresponding device component according to each of the initial load information includes: Determining the load standard deviation and current load change rate of the corresponding component according to each of the initial load information; Obtaining a basic period, a dynamic amplification factor, and a component weight value of each of the device components; The acquisition period of each of the device components is determined according to the basic period, the dynamic amplification coefficient, the component weight value of each of the device components, the load standard deviation and the current load change rate.

4. The liquid cooling heat dissipation dynamic flow control method according to claim 2, characterized in that: Determining the coolant flow required by each component in the corresponding next acquisition cycle according to each load information includes: Determining the amount of heat generated by each component in the corresponding next acquisition cycle according to each load information; The cooling liquid flow required by each component in the corresponding next collection cycle is determined based on each heat amount.

5. The liquid cooling heat dissipation dynamic flow control method according to claim 4, characterized in that: Determining the amount of heat generated by each component in the corresponding next acquisition cycle according to each load information includes: Determine a first static power consumption, a first dynamic power consumption, a maximum utilization, a nonlinear correction coefficient, a core temperature, an ambient temperature, a reference temperature difference, a temperature amplification coefficient, and a first liquid cooling efficiency factor corresponding to the central processing unit; Determining the amount of heat generated by the central processing unit in a corresponding next acquisition cycle according to the first static power consumption, the first dynamic power consumption, the maximum utilization rate, the nonlinear correction coefficient, the core temperature, the ambient temperature, the reference temperature difference, the temperature amplification factor, the first liquid cooling efficiency factor, and the corresponding acquisition cycle corresponding to the central processing unit; Determine a second static power consumption, memory utilization, utilization amplification factor, second dynamic power consumption, page swap frequency, swap frequency nonlinear coefficient, basic swap power consumption, and second liquid cooling efficiency factor corresponding to the memory; Determining the amount of heat generated by the memory in a corresponding next collection cycle based on the second static power consumption, memory utilization, utilization amplification factor, second dynamic power consumption, page swap frequency, swap frequency nonlinear coefficient, basic swap power consumption, second liquid cooling efficiency factor, and the corresponding collection cycle corresponding to the memory; Determine a third static power consumption, a third dynamic power consumption, and a third liquid cooling efficiency factor corresponding to the input and output devices; determining the amount of heat generated by the input / output device in a corresponding next acquisition cycle according to the third static power consumption, the third dynamic power consumption, the third liquid cooling efficiency factor, and the corresponding acquisition cycle corresponding to the input / output device; Determining a fourth static power consumption, a fourth dynamic power consumption, and a fourth liquid cooling efficiency factor corresponding to the graphics processing unit; The amount of heat generated by the graphics processing unit in the next corresponding acquisition cycle is determined according to the fourth static power consumption, the fourth dynamic power consumption, the fourth liquid cooling efficiency factor, and the corresponding acquisition cycle of the graphics processing unit.

6. The liquid cooling heat dissipation dynamic flow control method according to claim 5, characterized in that: Determine the first dynamic power consumption, core temperature, and first liquid cooling efficiency factor of the central processing unit, including: Get the CPU's load capacitance, voltage, frequency, and current utilization; determining a first dynamic power consumption of the central processing unit based on a load capacitance, a voltage, a frequency, and a current utilization rate of the central processing unit; Obtaining the sensor temperature, heat transfer coefficient, load intensity parameter and load intensity average temperature difference set on the central processing unit; Determine the core temperature of the CPU based on the sensor temperature, heat transfer coefficient, load intensity parameter, and load intensity average temperature difference; Obtain the actual thermal resistance, design thermal resistance and optimization parameters of the CPU; The first liquid cooling efficiency factor of the CPU is determined based on the actual thermal resistance, the designed thermal resistance and the optimization parameters of the CPU.

7. The liquid cooling heat dissipation dynamic flow control method according to claim 5, characterized in that: Determine the memory utilization and the second liquid cooling efficiency factor corresponding to the memory, including: Get the total memory capacity, free memory capacity and used memory capacity; Determine the memory utilization rate based on the total memory capacity, free memory capacity, and used memory capacity; Get the memory temperature, coolant inlet temperature, reference temperature difference, actual thermal resistance, design thermal resistance and correction factor corresponding to the memory; Determine the second liquid cooling efficiency factor corresponding to the memory based on the memory temperature, coolant inlet temperature, reference temperature difference, actual thermal resistance, design thermal resistance, and correction factor corresponding to the memory.

8. The liquid cooling heat dissipation dynamic flow control method according to claim 5, characterized in that: Determining a third static power consumption, a third dynamic power consumption, and a third liquid cooling efficiency factor corresponding to the input and output devices includes: Obtain the basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference, and temperature coefficient corresponding to the input and output devices; determining a third static power consumption corresponding to the input / output device according to the basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference, and temperature coefficient corresponding to the input / output device; Obtain the power consumption coefficient, read / write rate, input / output queue depth, device busy rate, and busy rate nonlinear coefficient corresponding to the input / output devices; Determining a third dynamic power consumption corresponding to the input / output device according to a power consumption coefficient, a read / write rate, an input / output queue depth, a device busy rate, and a busy rate nonlinear coefficient corresponding to the input / output device; Obtain the actual thermal resistance, design thermal resistance and correction coefficient corresponding to the input and output devices; Determine a third liquid cooling efficiency factor corresponding to the input and output devices based on the actual thermal resistance, design thermal resistance, correction coefficient, core temperature, coolant inlet temperature, and reference temperature difference corresponding to the input and output devices.

9. The liquid cooling heat dissipation dynamic flow control method according to claim 5, characterized in that: Determining a fourth static power consumption, a fourth dynamic power consumption, and a fourth liquid cooling efficiency factor corresponding to the graphics processing unit includes: Obtain the basic static power consumption, core temperature, coolant inlet temperature, reference temperature difference, and temperature coefficient corresponding to the graphics processing unit; determining a fourth static power consumption corresponding to the input / output device according to a basic static power consumption, a core temperature, a coolant inlet temperature, a reference temperature difference, and a temperature coefficient corresponding to the graphics processing unit; Obtaining the memory utilization, computing unit utilization, task queue length, memory nonlinear coefficient, task queue nonlinear coefficient, and power consumption coefficient corresponding to the graphics processing unit; Determining a fourth dynamic power consumption corresponding to the input / output device according to a video memory utilization rate, a computing unit utilization rate, a task queue length, a video memory nonlinear coefficient, a task queue nonlinear coefficient, and a power consumption coefficient corresponding to the graphics processing unit; Obtaining actual thermal resistance, design thermal resistance, correction factor, actual coolant flow rate, design flow rate, and flow rate compensation coefficient corresponding to the graphics processing unit; A fourth liquid cooling efficiency factor corresponding to the graphics processing unit is determined according to the actual thermal resistance, the designed thermal resistance, the correction coefficient, the actual coolant flow rate, the designed flow rate, and the flow compensation coefficient corresponding to the graphics processing unit.

10. The liquid cooling heat dissipation dynamic flow control method according to claim 5, characterized in that: Determining the cooling liquid flow required by each component in the corresponding next acquisition cycle according to each heat, including: Determining the theoretical coolant flow rate required by each component in the corresponding next collection cycle according to the heat and thermodynamic energy conservation law; Obtain flow redundancy coefficient, pipeline resistance correction coefficient, pipeline total resistance value and design resistance target value; Determining the actual coolant flow rate required by each component in the next acquisition cycle according to the flow redundancy coefficient, the pipeline resistance correction coefficient, the total pipeline resistance value, the design resistance target value, and each theoretical coolant flow rate; setting the target speed of the water pump and the opening of the flow regulating valve of the flow control component according to the actual coolant flow; The coolant flow rate required by each component in the corresponding next collection cycle is determined based on the actual coolant flow rate, the target water pump speed and the flow control valve opening.

11. The liquid cooling dynamic flow control method according to any one of claims 1 to 10, characterized in that: The liquid cooling distribution unit controls the total cooling liquid flow rate according to the total cooling liquid flow rate corresponding to each computing device, including: Obtaining the benchmark period, global redundancy coefficient, computing device priority coefficient, and dynamic collection period corresponding to each computing device; The total coolant flow rate is determined according to the reference period, the global redundancy coefficient, the computing device priority coefficient, the dynamic collection period corresponding to each computing device, and the total coolant flow rate.

12. The liquid cooling heat dissipation dynamic flow control method according to claim 11, characterized in that: After determining the total coolant flow rate, the method further includes: Obtain flow deviation records, response sensitivity coefficients, reference periods, global redundancy coefficients, and reference flow rates within historical periods; Determine the redundant flow rate according to the flow deviation record, the response sensitivity coefficient, the reference period, the global redundancy coefficient, the reference flow rate, the dynamic collection period corresponding to each computing device, and the sum of the coolant flow rates; Redundancy correction is performed on the total coolant flow according to the redundant flow.

13. An electronic device, characterized in that: include: memory for storing computer programs; A processor is configured to implement the steps of the liquid cooling dynamic flow control method according to any one of claims 1 to 12 when executing the computer program.

14. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the liquid cooling dynamic flow control method according to any one of claims 1 to 12 are implemented.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the liquid cooling dynamic flow control method according to any one of claims 1 to 12 are implemented.

Images