Liquid cooling control system and liquid cooling server

Abstract

This application provides a liquid cooling control system and a liquid cooling server, applicable to the field of liquid cooling server technology. The liquid cooling control system includes multiple liquid cooling node control units and a liquid cooling management control unit. Each liquid cooling node control unit is located within a corresponding server node and is configured to collect node data from the server node. The liquid cooling management control unit is connected via a CAN bus to a heat exchange unit, each liquid cooling node control unit, and the corresponding coolant pump and solenoid valve of each server node. The liquid cooling management control unit is configured to control the heat exchange unit and the corresponding coolant pump and solenoid valve of each server node based on the node data from the server node. The liquid cooling control system of this application communicates via a CAN bus, significantly increasing the number of server nodes that the liquid cooling control system can control and the communication distance, thus meeting the liquid cooling control requirements of large-scale data centers.

Description

Technical Field

[0001] This application relates to the field of liquid-cooled server technology, and in particular to a liquid-cooled control system and a liquid-cooled server. Background Technology

[0002] Liquid-cooled servers are a high-efficiency heat dissipation solution that uses liquid cooling technology to replace traditional air cooling. By directly or indirectly contacting the core heat-generating components of the server (such as CPU and GPU) with a coolant with a high specific heat capacity (such as mineral oil or fluorinated liquid), the heat is precisely and efficiently dissipated by utilizing the rapid heat conduction characteristics of the coolant.

[0003] Currently, during operation, a single server node can acquire key node data such as the temperature of corresponding components within the node and the temperature of the coolant entering and exiting the node. This collected node data is then transmitted to the liquid cooling management and control unit in the liquid cooling control center via Ethernet or I2C bus. Based on this node data, the liquid cooling management and control unit precisely controls the liquid cooling heat dissipation of the server node, ensuring stable operation within a suitable temperature range.

[0004] However, the communication cost of Ethernet communication increases with the number of nodes, and the communication stability and real-time performance are also easily affected. On the other hand, with the increase of the number of nodes, the cable length will become too long, or when there are too many nodes, the signal quality will be seriously affected, causing communication problems. Utility Model Content

[0005] This application provides a liquid cooling control system and a liquid cooling server to solve the technical problem that the existing liquid cooling control system of liquid cooling server cannot meet the liquid cooling control requirements of large-scale data centers.

[0006] According to a first aspect disclosed in this application, this application provides a liquid cooling control system, comprising:

[0007] Multiple liquid-cooled node control units are provided, each of which is located within a corresponding server node. The liquid-cooled node control units are configured to collect node data from the server node.

[0008] A liquid cooling management and control unit is provided, which is connected via a CAN bus to the heat exchange unit, each liquid cooling node control unit, and the coolant pump and solenoid valve corresponding to each server node. The liquid cooling management and control unit is configured to control the heat exchange unit and the coolant pump and solenoid valve corresponding to each server node according to the node data of the server node. The coolant pump is configured to control the flow rate of coolant in the server node, and the solenoid valve is configured to control the opening and closing of the coolant inlet in the server node.

[0009] In one feasible implementation, the liquid-cooled node control unit includes a liquid-cooled node controller and a substrate management controller, wherein the liquid-cooled node controller is connected to the substrate management controller and the liquid-cooled node controller is connected to the liquid-cooled management control unit;

[0010] The baseboard management controller is configured to collect node temperature data and node load from the node data;

[0011] The liquid-cooled node controller is configured to collect liquid level data from the node data.

[0012] In one feasible implementation, the liquid-cooled node controller is connected to a corresponding solenoid valve, and the liquid-cooled node controller is configured to control the corresponding solenoid valve based on the node data of the server node.

[0013] In one feasible implementation, the liquid-cooled node controller is connected to the substrate management controller via an I2C bus.

[0014] In one feasible implementation, the liquid-cooled node controller is connected to the corresponding coolant pump via a CAN bus, and the liquid-cooled node controller is configured to control the corresponding coolant pump based on the node data of the server node.

[0015] In one feasible implementation, the liquid-cooled node controller is connected to the heat exchange unit via a CAN bus, and the liquid-cooled node controller is configured to control the heat exchange unit based on node data from the server node.

[0016] In one feasible implementation, the node temperature data includes GPU temperature, CPU temperature, coolant inlet temperature, and coolant outlet temperature.

[0017] According to a second aspect disclosed in this application, this application provides a liquid-cooled server, including a liquid-cooled control system as described in any one of the first aspects.

[0018] In one feasible implementation, the liquid-cooled server includes a heat exchange unit, multiple server nodes, and a coolant pump and a solenoid valve corresponding to each server node, wherein the solenoid valve is disposed at the liquid inlet of the corresponding server node.

[0019] The outlet of the heat exchange unit is connected to the inlet of each server node, the outlet of the server node is connected to the inlet of the corresponding coolant pump, and the outlet of the coolant pump is connected to the inlet of the heat exchange unit.

[0020] In one feasible implementation, the liquid-cooled server further includes a rack in which multiple server nodes are disposed.

[0021] Compared with the prior art, this application has the following advantages:

[0022] This application provides a liquid cooling control system and a liquid cooling server. Each liquid cooling management control unit collects node data from each server node and sends this data to the liquid cooling management control unit. The liquid cooling management control unit then controls the heat exchange unit and the corresponding coolant pumps and solenoid valves of each server node based on this node data, thereby regulating the cooling efficiency of each server node and achieving efficient operation of the liquid cooling server. During this process, the liquid cooling management control unit communicates with the heat exchange unit, each liquid cooling node control unit, and the corresponding coolant pumps and solenoid valves of each server node via a CAN bus. Using a CAN bus for communication offers low hardware and software costs, strong scalability, flexible application, high stability, high real-time performance, and low cost, significantly increasing the number of server nodes that the liquid cooling control system can control and the communication distance, meeting the liquid cooling control requirements of large-scale data centers. Attached Figure Description

[0023] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0024] Figure 1 This is a schematic diagram of the structure of a liquid-cooled server provided in an embodiment of this application;

[0025] Figure 2 This is a schematic diagram of a liquid cooling control system provided in an embodiment of this application.

[0026] Explanation of reference numerals in the attached figures:

[0027] 100 - Server Nodes;

[0028] 101 - Solenoid valve;

[0029] 102-Baseboard Management Controller;

[0030] 103-Liquid-cooled node controller;

[0031] 200-Coolant Pump;

[0032] 300 - Heat exchange unit;

[0033] 400 - Liquid Cooling Management and Control Unit.

[0034] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0035] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0036] Liquid-cooled servers are a high-efficiency heat dissipation solution that uses liquid cooling technology to replace traditional air cooling. By directly or indirectly contacting the core heat-generating components of the server (such as CPU and GPU) with a coolant with a high specific heat capacity (such as mineral oil or fluorinated liquid), the heat is precisely and efficiently dissipated by utilizing the rapid heat conduction characteristics of the coolant. Compared with traditional air cooling, it can achieve a heat dissipation efficiency improvement of several times or even ten times.

[0037] Currently, during operation, a single server node in a liquid-cooled server can acquire key node data such as the temperature of corresponding components within the node and the temperature of the inlet and outlet coolant. For immersion liquid-cooled servers, in addition to the aforementioned temperature information, data also includes liquid level parameters and solenoid valve status information. The collected node data is then transmitted to the liquid cooling management and control unit in the liquid cooling control center via a network or I2C bus. After receiving the node data, the liquid cooling management and control unit will precisely control the liquid cooling heat dissipation of the server node according to certain strategies to ensure that the server node operates stably within a suitable temperature range.

[0038] However, when communicating via Ethernet, as the number of nodes increases, additional network modules and switching units are required, leading to a continuous increase in hardware and software analysis costs. Furthermore, the stability and real-time performance of network communication are easily affected by the network environment. Network congestion or failures can cause data transmission delays or losses, thus affecting the accuracy of liquid cooling control. While I2C communication avoids the complexity and high cost of network communication to some extent, when multiple nodes use I2C simultaneously, the increasing number of nodes leads to excessively long and numerous cables, resulting in a complex control structure and poor scalability. Alternatively, with too many nodes, simultaneous competition for bus resources can severely impact signal quality, causing communication problems and preventing the liquid cooling control center from obtaining data from each node in a timely and accurate manner, thereby affecting the normal operation of the liquid cooling server.

[0039] To address the aforementioned technical issues, this application proposes a liquid cooling control system and a liquid cooling server. The liquid cooling control system communicates via a CAN bus, which greatly increases the number of server nodes that the liquid cooling control system can control and the communication distance, thus meeting the liquid cooling control requirements of large-scale data centers.

[0040] The technical solutions of the liquid cooling control system and liquid cooling server provided in this application will be described in detail below through specific embodiments. It should be noted that the following embodiments may exist alone or in combination with each other, and the same or similar content may not be described again in different embodiments.

[0041] Below, in conjunction with Figure 1 This application describes a liquid-cooled server.

[0042] Figure 1 This is a schematic diagram of the structure of a liquid-cooled server provided in an embodiment of this application. (See attached diagram.) Figure 1 In some embodiments, the liquid-cooled server includes a heat exchange unit 300, multiple server nodes 100, and a coolant pump 200 and a solenoid valve 101 corresponding to each server node 100. The solenoid valve 101 is disposed at the inlet of the corresponding server node 100. The outlet of the heat exchange unit 300 is connected to the inlet of each server node 100, the outlet of the server node 100 is connected to the inlet of the corresponding coolant pump 200, and the outlet of the coolant pump 200 is connected to the inlet of the heat exchange unit 300.

[0043] Specifically, server node 100 is the basic computing unit of the liquid-cooled server, consisting of a motherboard, CPU, GPU, memory, cold plate, and power supply. The server achieves efficient heat dissipation through indirect contact between the cold plate and the coolant to support high-density computing tasks.

[0044] Specifically, the heat exchange unit 300 is the cooling capacity distribution unit of the liquid-cooled server. It is responsible for accurately distributing the cryogenic coolant to each server node 100, removing heat from the server nodes 100 through heat exchange, and regulating the coolant temperature to ensure the stable operation of the server nodes 100. The heat exchange unit 300 can be a shell-and-tube heat exchanger, a coaxial heat exchanger, a plate heat exchanger, etc.

[0045] Specifically, the coolant pump 200 is responsible for powering the circulation of coolant within the server node 100, thereby regulating the coolant flow rate to optimize the heat transfer efficiency of the coolant.

[0046] Specifically, the solenoid valve 101 is the safety valve and regulating valve of the server node 100. The solenoid valve 101 can dynamically adjust the coolant flow of the server node 100 to prevent the server node 100 from being too cold or too hot, thereby improving the cooling utilization efficiency.

[0047] In the liquid cooling process of the liquid-cooled server, the low-temperature coolant of the heat exchange unit 300 enters the server node 100 through the pipeline to cool the server node 100. During the cooling process, the coolant absorbs heat and rises in temperature. The high-temperature coolant is then exchanged with the heat exchange unit 300 again through the pipeline and becomes a low-temperature coolant to participate in the cooling cycle again, thereby achieving the cyclic cooling of the server node 100.

[0048] Optionally, the liquid-cooled server also includes a rack, in which multiple server nodes 100 are housed.

[0049] The rack provides robust physical support and organized installation space for multiple server nodes 100.

[0050] See Figure 2 In some embodiments, the liquid-cooled server further includes the following liquid-cooling control system.

[0051] In this embodiment, during the cyclic cooling process of the liquid-cooled server, the heat exchange unit 300, coolant pump 200, and solenoid valve 101 are controlled by the liquid cooling control system. The liquid cooling control system communicates via a CAN bus, which greatly increases the number of server nodes 100 that the liquid cooling control system can control and the communication distance, thus meeting the liquid cooling control requirements of large-scale data centers.

[0052] Specifically, the liquid-cooled server can be a single-phase liquid-cooled server or a multi-phase liquid-cooled server.

[0053] Figure 2 This is a schematic diagram of a liquid cooling control system provided in an embodiment of this application. (See attached diagram.) Figure 2In some embodiments, the liquid cooling control system includes multiple liquid cooling node control units and a liquid cooling management control unit 400. Each liquid cooling node control unit is located within a corresponding server node 100 and is configured to collect node data from the server node 100. The liquid cooling management control unit 400 is connected to the heat exchange unit 300, each liquid cooling node control unit, and the coolant pump 200 and solenoid valve 101 corresponding to each server node 100 via a CAN bus. The liquid cooling management control unit 400 is configured to control the heat exchange unit 300 and the coolant pump 200 and solenoid valve 101 corresponding to each server node 100 based on the node data of the server node 100. The coolant pump 200 is configured to control the flow rate of the coolant within the server node 100, and the solenoid valve 101 is configured to control the opening and closing of the coolant inlet within the server node 100.

[0054] In this embodiment, each liquid cooling management control unit 400 collects node data from each server node 100 and sends the node data to the liquid cooling management control unit 400. The liquid cooling management control unit 400 then controls the heat exchange unit 300 and the corresponding coolant pump 200 and solenoid valve 101 of each server node 100 based on the node data, thereby regulating the cooling efficiency of each server node 100 and achieving efficient operation of the liquid-cooled server. During this process, the liquid cooling management control unit 400 communicates with the heat exchange unit 300, each liquid cooling node control unit, and the corresponding coolant pump 200 and solenoid valve 101 of each server node 100 via a CAN bus. Using a CAN bus for communication offers low hardware and software costs, strong scalability, flexible application, high stability, high real-time performance, and low cost, significantly increasing the number of server nodes 100 that the liquid cooling control system can control and the communication distance, meeting the liquid cooling control requirements of large-scale data centers.

[0055] Specifically, CAN communication is a highly reliable, distributed, real-time serial communication protocol designed specifically for automotive and industrial environments. It employs two-wire differential signaling (CAN_H and CAN_L) for transmission, supports multiple master nodes freely competing for bus control, ensures high-priority messages are sent in real-time through a priority arbitration mechanism, and possesses strong anti-electromagnetic interference capabilities and error detection / recovery functions (such as CRC checksum and acknowledgment mechanisms). Its data frames are concise and efficient (maximum 8-byte payload), with a transmission rate of up to 1 Mbps (within 40 meters), and it is widely used in automotive electronics (such as engine control and body networks), industrial automation, and intelligent devices.

[0056] Therefore, the CAN bus was adopted as the communication and control bus. It offers advantages such as low cost, fewer cables, strong anti-interference capability, good real-time performance, support for multi-master systems, error detection and handling mechanisms, high flexibility, simple software operation, and long signal transmission distance. Up to 30 nodes can be connected, and at a rate of 40 kbits / s, the cabling length can reach 1000 meters. This better meets the requirements of current large-scale data centers for liquid-cooled server control in terms of reliability, stability, scalability, and low cost. Specifically, the devices using CAN bus communication also include a CAN controller and a CAN transceiver.

[0057] Specifically, the liquid cooling management control unit 400 acts as the central control center, receiving node data such as temperature, liquid level, and node load from each server node 100. It then executes liquid cooling control actions according to a specific control strategy, such as sending commands to the coolant pump 200 to increase or decrease the flow rate, and sending commands to the liquid cooling node control units of the server node 100 to open or close the solenoid valve 101. The liquid cooling management control unit 400 can be a programmable logic controller (PLC), a microcontroller (MCU), or a dedicated industrial control computer.

[0058] Specifically, all server nodes 100 are connected to a CAN bus. CAN network communication is carried out through frame data. All nodes can send or receive frame data. The frame data contains ID information. Different server nodes 100 are assigned a fixed ID. Server nodes 100 can receive frame information containing their own ID, thereby realizing the sending and receiving of information between different nodes.

[0059] See Figure 2 In some embodiments, the liquid-cooled node control unit includes a liquid-cooled node controller 103 and a substrate management controller 102. The liquid-cooled node controller 103 is connected to the substrate management controller 102 and is also connected to the liquid-cooled management control unit 400. The substrate management controller 102 is configured to collect node temperature data and node load data from the node data. The liquid-cooled node controller 103 is configured to collect liquid level data from the node data.

[0060] In this embodiment, the Baseboard Management Controller 102 (BMC) is an independent microcontroller embedded on the server motherboard, used for remote monitoring and management of server hardware status, providing out-of-band management capabilities, and ensuring efficient and stable server operation. The Liquid Cooling Node Controller 103 is a control element in the liquid-cooled server responsible for monitoring and managing the liquid cooling process of the server node 100, ensuring that the server operates in a highly efficient and stable state.

[0061] Optionally, node temperature data includes GPU temperature, CPU temperature, coolant inlet temperature, and coolant outlet temperature.

[0062] Among them, the cooling status of server node 100 can be comprehensively and accurately reflected by GPU temperature, CPU temperature, coolant inlet temperature and coolant outlet temperature.

[0063] Specifically, node load is a key indicator for measuring the utilization of its processing capacity. It is usually reflected by a combination of parameters such as the number of currently active tasks, CPU utilization, GPU utilization, memory usage, and disk I / O throughput. The dynamic changes in node load directly affect the heat generation of electronic components in server node 100.

[0064] Specifically, the liquid-cooled node controller 103 can be a programmable logic controller (PLC) or a microcontroller (MCU).

[0065] See Figure 2 Optionally, the liquid-cooled node controller 103 is connected to the corresponding solenoid valve 101, and the liquid-cooled node controller 103 is configured to control the corresponding solenoid valve 101 according to the node data of the server node 100.

[0066] By connecting the liquid-cooled node controller 103 to the solenoid valve 101, the liquid-cooled node controller 103 can also directly control the solenoid valve 101 within the node, thereby improving the control efficiency of the solenoid valve 101.

[0067] See Figure 2 Optionally, the liquid-cooled node controller 103 is connected to the baseboard management controller 102 via an I2C bus.

[0068] Since the liquid-cooled node controller 103 and the baseboard management controller 102 are located within the same server node 100 and are relatively close, I2C communication is more suitable for this low-speed, short-distance, and low-complexity communication scenario than CAN communication. By connecting the liquid-cooled node controller 103 and the baseboard management controller 102 through the I2C bus, efficient, low-speed, and low-cost bidirectional data transmission and control between the liquid-cooled node controller 103 and the baseboard management controller 102 can be achieved.

[0069] See Figure 2 Optionally, the liquid-cooled node controller 103 is connected to the corresponding coolant pump 200 via a CAN bus, and the liquid-cooled node controller 103 is configured to control the corresponding coolant pump 200 according to the node data of the server node 100.

[0070] Since the CAN bus supports multi-master systems, in addition to the single-master control mode where the liquid cooling management control unit 400 is the only master, the liquid cooling node controller 103 can also be connected to the coolant pump 200 via the CAN bus, thus making the liquid cooling node controller 103 also a control master. For example, when the load of the server node 100 decreases and the CPU or GPU power decreases, the liquid cooling node controller 103 of the server node 100 can actively send a command to the coolant pump 200 to control the coolant pump 200 to reduce the coolant flow rate of the node, thereby reducing the overall system power consumption. Moreover, it does not need to go through the liquid cooling management control unit 400 as an intermediary, which greatly reduces the control delay and improves the control efficiency.

[0071] See Figure 2 Optionally, the liquid-cooled node controller 103 is connected to the heat exchange unit 300 via a CAN bus, and the liquid-cooled node controller 103 is configured to control the heat exchange unit 300 based on the node data of the server node 100.

[0072] The liquid cooling node controller 103 and the heat exchange unit 300 are connected via a CAN bus, which enables the liquid cooling node controller 103 to directly control the heat exchange unit 300, thereby reducing control delay and improving control efficiency.

[0073] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0074] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0075] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "joining," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0076] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0077] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0078] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the appended claims.

[0079] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A liquid cooling control system, characterized in that, include: Multiple liquid-cooled node control units are provided, each of which is located within a corresponding server node. The liquid-cooled node control units are configured to collect node data from the server node. A liquid cooling management and control unit is provided, which is connected via a CAN bus to the heat exchange unit, each liquid cooling node control unit, and the coolant pump and solenoid valve corresponding to each server node. The liquid cooling management and control unit is configured to control the heat exchange unit and the coolant pump and solenoid valve corresponding to each server node according to the node data of the server node. The coolant pump is configured to control the flow rate of coolant in the server node, and the solenoid valve is configured to control the opening and closing of the coolant inlet in the server node.

2. The liquid cooling control system according to claim 1, characterized in that, The liquid-cooled node control unit includes a liquid-cooled node controller and a substrate management controller, wherein the liquid-cooled node controller is connected to the substrate management controller, and the liquid-cooled node controller is connected to the liquid-cooled management control unit; The baseboard management controller is configured to collect node temperature data and node load from the node data; The liquid-cooled node controller is configured to collect liquid level data from the node data.

3. The liquid cooling control system according to claim 2, characterized in that, The liquid-cooled node controller is connected to a corresponding solenoid valve, and the liquid-cooled node controller is configured to control the corresponding solenoid valve according to the node data of the server node.

4. The liquid cooling control system according to claim 2, characterized in that, The liquid-cooled node controller is connected to the substrate management controller via an I2C bus.

5. The liquid cooling control system according to claim 2, characterized in that, The liquid-cooled node controller is connected to the corresponding coolant pump via a CAN bus, and the liquid-cooled node controller is configured to control the corresponding coolant pump based on the node data of the server node.

6. The liquid cooling control system according to claim 2, characterized in that, The liquid-cooled node controller is connected to the heat exchange unit via a CAN bus, and the liquid-cooled node controller is configured to control the heat exchange unit based on the node data of the server node.

7. The liquid cooling control system according to any one of claims 2-6, characterized in that, The node temperature data includes GPU temperature, CPU temperature, coolant inlet temperature, and coolant outlet temperature.

8. A liquid-cooled server, characterized in that, Includes the liquid cooling control system as described in any one of claims 1-7.

9. The liquid-cooled server according to claim 8, characterized in that, The liquid-cooled server includes a heat exchange unit, multiple server nodes, and a coolant pump and a solenoid valve corresponding to each server node. The solenoid valve is located at the liquid inlet of the corresponding server node. The outlet of the heat exchange unit is connected to the inlet of each server node, the outlet of the server node is connected to the inlet of the corresponding coolant pump, and the outlet of the coolant pump is connected to the inlet of the heat exchange unit.

10. The liquid-cooled server according to claim 8, characterized in that, The liquid-cooled server also includes a rack, in which multiple server nodes are housed.

Images