Liquid-cooled dummy load test device, system, and method

Abstract

The application discloses a liquid cooling dummy load testing device, system and method. The device comprises a case, a mainboard in the case and a plurality of simulated heat generating components on the mainboard. The mainboard comprises a plurality of partitions, each of which comprises a first fixing device and a first signal connecting device for connecting the simulated heat generating components, and a buried wire for power supply and signal transmission. Each simulated heat generating component comprises a second fixing device and a second signal connecting device for connecting the mainboard, a temperature sensor and a remotely controllable adjustable heating device. The simulated heat generating component is used for simulating a target component generating heat in a server, and the target component at least comprises a hard disk, a CPU, a GPU and a memory. The application solves the technical problem of lacking effective testing means for direct contact type liquid cooling cabinets in the related art.

Description

Technical Field

[0001] This application relates to the field of equipment testing technology, and more specifically, to a liquid-cooled dummy load testing device, system, and method. Background Technology

[0002] To reduce the PUE (Power Usage Effectiveness) of data centers, energy-saving technologies for data center HVAC systems are actively exploring liquid cooling. Based on whether the coolant is in direct contact with the heat-generating components of the server, liquid cooling in computer rooms is divided into direct contact liquid cooling and indirect contact liquid cooling. Among them, the mainstream direct contact liquid cooling in the industry is divided into two types: immersion type and spray type.

[0003] In direct-contact liquid cooling scenarios, on the one hand, servers need to be in direct contact with the coolant. Existing air-cooled server room dummy load testing equipment uses air cooling for heat dissipation and does not have the ability to simulate the heat generation of real equipment in a liquid environment. On the other hand, servers and racks (or liquid cooling boxes) need to be customized with a single rack power of over 20KW. The configuration of each server will affect the flow direction and flow rate of the coolant in the rack (or liquid cooling box). In the original air-cooled dummy load scenario, the method of simulating the heat generation of all rack servers through a single dummy load device is no longer suitable for server-level testing in direct-contact liquid-cooled racks (or liquid cooling boxes).

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a liquid-cooled dummy load testing device, system, and method to at least solve the technical problem of the lack of effective testing methods for direct-contact liquid-cooled cabinets in related technologies.

[0006] According to one aspect of the embodiments of this application, a liquid-cooled dummy load testing device is provided, comprising: a chassis, a motherboard located within the chassis, and a plurality of simulated heat-generating components located on the motherboard, wherein the motherboard includes a plurality of partitions, each partition including: a first fixing device and a first signal connection device for connecting to the simulated heat-generating components, and embedded wires for power supply and signal transmission; each simulated heat-generating component includes: a second fixing device and a second signal connection device for connecting to the motherboard, a temperature sensor, and a remotely controllable adjustable heat-generating device; wherein the simulated heat-generating components are used to simulate target components that generate heat in a server, the target components including at least: a hard disk, a CPU, a GPU, and memory.

[0007] Optionally, the chassis dimensions are standard server dimensions.

[0008] Optionally, a coolant drain pipe is provided at the bottom of the chassis, and a temperature sensor and a liquid flow rate sensor are provided at the interface of the coolant drain pipe; a liquid level sensor is provided inside the chassis.

[0009] Alternatively, the chassis may have no top cover and may be equipped with mounting ears.

[0010] Optionally, the embedded wire is connected to the first signal connection device in each partition, and the type of embedded wire includes at least one of the following: power supply line, sensor communication data line, and resistor control line; the chassis is provided with a wire outlet hole for the embedded wire.

[0011] Optionally, the motherboard is a rigid printed circuit board, and the motherboard is fixed to a base plate inside the chassis. The base plate is made of at least one of the following materials: bakelite board, mica board.

[0012] Optionally, both the first fixing device and the second fixing device are neodymium iron boron high-strength magnets.

[0013] Optionally, both the first signal connection device and the second signal connection device are metal contacts.

[0014] Optionally, a groove is provided around the first signal connection device, and a rubber ring is provided in the groove, with the upper surface of the rubber ring flush with the motherboard plane; a groove is provided around the second signal connection device, and a rubber ring is provided in the groove, with the upper surface of the rubber ring higher than the motherboard plane.

[0015] Optionally, the simulated heating component has a layered structure, wherein the temperature sensor is located on the upper layer of the second fixing device and the second signal connection device, and the adjustable heating device is located on the upper layer of the temperature sensor.

[0016] Optionally, a pressure sensor is also provided on the upper part of the adjustable heating device.

[0017] Optionally, the first fixing device is located at the four corners of the partition, the first signal connection device is located at the center of the partition, the second fixing device is located at the four corners of the bottom layer of the simulated heating component, and the second signal connection device is located at the center of the bottom layer of the simulated heating component.

[0018] According to another aspect of the embodiments of this application, a liquid-cooled dummy load testing system is also provided, including: a power supply module, a data acquisition module, a control module, and a liquid-cooled dummy load testing device. The power supply module is used to supply power to each simulated heat-generating component in the liquid-cooled dummy load testing device according to power supply commands issued by the control module. The data acquisition module is used to acquire test data from the liquid-cooled dummy load testing device and send the test data to the control module. The control module is used to perform K-means cluster analysis on the test data and determine a management strategy for the target server simulated by the liquid-cooled dummy load testing device based on the cluster analysis results.

[0019] Optionally, the control module is used to send power supply commands to the power supply module, including the set heating power for each simulated heating component; the power supply module is used to send the real-time heating power fed back by each simulated heating component to the control module.

[0020] Optionally, the acquisition module may include at least one of the following: a pressure acquisition submodule, a temperature acquisition submodule, a liquid level acquisition submodule, and a flow rate acquisition submodule.

[0021] Optionally, the control module is used to determine abnormal components in the target server based on the scatter plot corresponding to the cluster analysis results, and to issue an alarm message when an abnormal component is found; the control module is also used to determine the coolant flow direction control strategy for each component in the target server based on the standard deviation of the cluster analysis results.

[0022] According to another aspect of the embodiments of this application, a liquid-cooled dummy load testing method is also provided, comprising: acquiring temperature data during the testing process of the liquid-cooled dummy load testing device; performing K-means clustering analysis on the temperature data to obtain clustering analysis results; and determining a management strategy for the target server simulated by the liquid-cooled dummy load testing device based on the clustering analysis results.

[0023] Optionally, acquiring temperature data during the testing process of the liquid-cooled dummy load test device includes: acquiring temperature data of each simulated heat-generating component during the testing process of the liquid-cooled dummy load test device according to a preset acquisition frequency within the test cycle.

[0024] Optionally, K-means clustering analysis is performed on the temperature data to obtain clustering results, including: at each acquisition time, filtering out data whose temperature does not exceed a preset threshold; for each filtered temperature data, converting the temperature data into two-dimensional coordinates, where the horizontal axis of the two-dimensional coordinates is the distance between the position of the simulated heating component on the motherboard corresponding to the temperature data and the preset origin, and the vertical axis is the temperature corresponding to the temperature data; setting K=4, clustering all two-dimensional coordinates at the current acquisition time based on the four centroids obtained at the previous acquisition time and the preset clustering distance of each centroid, obtaining clustering results, and plotting the scatter plot corresponding to the clustering results. The clustering results include four new centroids, where at the initial acquisition time, the initial four centroids are determined to be the two-dimensional coordinates corresponding to the temperature data of the four simulated heating components used to simulate the hard drive, CPU, GPU, and memory; and determining the standard deviation of the four centroids based on all clustering results throughout the entire test cycle.

[0025] Optionally, a management strategy for the target server simulated by the liquid-cooled dummy load test device is determined based on the cluster analysis results, including: determining that the server component simulated by the simulated heat-generating component corresponding to the two-dimensional coordinate that did not appear in the cluster results during the entire test cycle has an abnormal temperature and issuing an alarm message; determining the coolant flow direction control strategy for each component in the target server based on the standard deviation, wherein, for the server component simulated by the simulated heat-generating component corresponding to the centroid with a larger standard deviation, the coolant flows through the server component more times and the flow rate is slower.

[0026] According to another aspect of the embodiments of this application, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored computer program, wherein the device where the non-volatile storage medium is located executes the above-described liquid-cooled dummy load test method by running the computer program.

[0027] In this embodiment, a liquid-cooled dummy load testing device is provided based on the layout of heat-generating components within a real server. This device simulates components with a motherboard layout and heat generation that closely resemble those of a real server, ensuring maximum fidelity in replicating the actual server. The flexibility of simulating heat-generating component placement and its pluggable installation method gives the testing device high flexibility and reusability, improving the efficiency of batch testing. Furthermore, K-means clustering analysis of temperature data during the liquid-cooled dummy load testing process accurately locates anomaly monitoring points and provides early warnings. This solution effectively addresses the technical problem of lacking effective testing methods for direct-contact liquid-cooled server racks in related technologies. Attached Figure Description

[0028] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0029] Figure 1 This is a schematic diagram of an optional liquid-cooled dummy load testing device according to an embodiment of this application;

[0030] Figure 2 This is a schematic diagram of another optional liquid-cooled dummy load testing device according to an embodiment of this application;

[0031] Figure 3 This is a schematic diagram of an optional simulated heating component according to an embodiment of this application;

[0032] Figure 4 This is a schematic diagram of an optional liquid-cooled dummy load testing system according to an embodiment of this application;

[0033] Figure 5This is a schematic diagram of the structure of a computer terminal according to an embodiment of this application;

[0034] Figure 6 This is a flowchart illustrating an optional liquid-cooled dummy load testing method according to an embodiment of this application. Detailed Implementation

[0035] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0036] It should be noted that the terms "first," "second," etc., used in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0037] To better understand the embodiments of this application, the following is a translation and explanation of some nouns or terms that appear in the description of the embodiments of this application:

[0038] Liquid-cooled dummy load: This is a device used to test the performance of liquid-cooled systems. A liquid-cooled system is a heat dissipation system that uses liquid as the heat transfer medium to dissipate heat from high-power electronic devices. A liquid-cooled dummy load simulates the thermal load of high-power electronic devices, testing the heat dissipation effect and stability of the liquid-cooled system by flowing liquid coolant through the dummy load.

[0039] K-means clustering algorithm: It is an unsupervised learning algorithm based on distance metric, used to divide a dataset into K non-overlapping clusters. The basic idea of ​​the algorithm is to first randomly select K initial cluster centers, then iteratively assign each sample point to the nearest cluster center, and then update the position of the cluster center according to the assignment result, until the cluster center no longer changes or the maximum number of iterations is reached.

[0040] Example 1

[0041] According to an embodiment of this application, a liquid-cooled dummy load testing device is first provided, such as... Figure 1 As shown, the device includes at least: a chassis 11, a motherboard 12 located inside the chassis, and multiple analog heating components 13 (1 to m) located on the motherboard. The motherboard includes multiple partitions 12 (1 to n). Each partition 121 includes: a first fixing device 1211 for connecting to the analog heating components, a first signal connection device 1212, and embedded wires 1213 for power supply and signal transmission. The other partitions 12 (2 to n) have the same structure as partition 121. Figure 1 Not shown; the simulated heating component 131 includes: a second fixing device 1311 and a second signal connection device 1312 for connection to the motherboard, a temperature sensor 1313, a remotely controllable adjustable heating device 1314, and other simulated heating components 13 (2~m) with the same structure as the simulated heating component 131. Figure 1 Not shown. Among them, the simulated heat-generating component 13 (1~m) is used to simulate the target components that generate heat in the server, and the target components include at least: hard disk, CPU, GPU and memory.

[0042] exist Figure 1 Based on the liquid-cooled dummy load test device shown, Figure 2 A schematic diagram of a more specific liquid-cooled dummy load testing device is shown below, in conjunction with... Figure 2 The specific structure and function of each component in the liquid-cooled dummy load testing device provided in this application example are described.

[0043] In the diagram, 1 is the chassis, 2 is the motherboard, which is divided into multiple independent partitions, 3 is the mounting bracket, 4 is the liquid level sensor, 5 is the coolant drain pipe, 6 is the cable outlet hole, and the simulated hard drive, simulated CPU, simulated GPU, and simulated memory are simulated heat-generating components used to simulate the hard drive, CPU, GPU, and memory, respectively.

[0044] Optionally, the chassis can be made of galvanized steel sheet, with standard server dimensions, and supports three heights: 1 / 2 / 4U, allowing the chassis to be pushed parallel into the rack or tray.

[0045] Optionally, a coolant drain pipe is provided at the bottom of the chassis, allowing the coolant to flow out by gravity after heat exchange. The coolant drain pipe can also be equipped with a quick-connect fitting, allowing it to connect to the main drain pipe of the cabinet. Additionally, a temperature sensor and a liquid flow rate sensor are installed at the interface of the coolant drain pipe. A liquid level sensor is also installed inside the chassis to monitor the coolant level in a timely manner.

[0046] Optionally, to meet different application environments, the chassis has no top cover and is equipped with mounting ears. In the spray-type liquid cooling test environment, the test device can be directly pushed horizontally into the rack for placement. The chassis does not have a top cover and can directly use the top cover of a real server or adopt an open structure to ensure that the coolant outflow position is consistent with the real usage scenario. In the immersion-type liquid cooling test environment, the server can be vertically immersed in the coolant by relying on the mounting ears for fixation.

[0047] Optionally, the motherboard is a rigid printed circuit board (PCB). To ensure the stability of the motherboard under the flushing of coolant, the motherboard can be fixed to the base plate inside the chassis. The base plate must have the characteristics of high strength, non-conductivity, and high temperature resistance. Its materials can be: bakelite board or mica board.

[0048] To more realistically simulate the heat-generating components in a server, the simulated heat-generating components should be the same size as the actual heat-generating components in a server. Correspondingly, the size of the independent partitions within the motherboard can also be set with reference to the size of the main heat-generating components in a server, such as hard drives, CPUs, GPUs, and memory. For example, they can be set to 76mm*56.5mm or 70mm*100mm to make the simulated heat-generating components fit the motherboard partitions more closely.

[0049] Optionally, the simulated heating component has a layered structure, wherein the temperature sensor is located on the upper layer of the second fixing device and the second signal connection device, and the adjustable heating device is located on the upper layer of the temperature sensor; the heating device can be made of materials such as alloy heating wire or ceramic heating element, and the device has a remote adjustment function, which can adjust the heat generation through the control system; the temperature sensor can be a patch-type thermally coupled temperature sensor matrix, which is installed on the lower part of the heating device, and can measure the cooling effect of the heating device.

[0050] Optionally, in the simulated heating component, a pressure sensor is also provided on the upper layer of the adjustable heating device. This pressure sensor can be a patch-type pressure sensor matrix to ensure further zoning and fault tolerance of pressure measurement. Figure 3 The structure of an optional simulated heating element is shown, wherein, from bottom to top, the elements are: L1 layer - second fixing device and second circuit connection device, L2 layer - temperature sensor, L3 layer - heating device, and L4 layer - pressure sensor.

[0051] When fixing the simulated heating component and the motherboard, in order to avoid the impact of opening holes in the motherboard or adding component fixing bases on the flow of coolant, the simulated heating component and the motherboard can be connected by strong magnets. Specifically, the first fixing device and the second fixing device can be neodymium iron boron strong magnets.

[0052] Optionally, to ensure normal signal data transmission, the first and second signal connection devices can be metal contacts, which are mainly used for connection between power supply and signal lines. The embedded wires in the motherboard are connected to the first signal connection device in each partition, thus forming metal contacts within each independent partition. The embedded wires include at least one of the following: power supply lines, sensor communication data lines, and resistor control lines. The chassis is provided with wire exit holes for the embedded wires.

[0053] Optionally, to prevent coolant from flowing in and affecting the metal contacts, a groove is provided around the first signal connection device, and a rubber ring is provided in the groove, with the upper surface of the rubber ring flush with the motherboard plane; a groove is also provided around the second signal connection device, and a rubber ring is provided in the groove. The rubber ring can be installed in a semi-recessed and convex manner, that is, the upper surface of the rubber ring is higher than the motherboard plane, to ensure that the simulated heating component fits in close contact with the rubber ring of the motherboard partition.

[0054] Optionally, the first fixing device can be set at the four corners of the partition, the first signal connection device can be set at the center of the partition, the second fixing device is also set at the four corners of the bottom layer of the simulated heating component, and the second signal connection device is set at the center of the bottom layer of the simulated heating component.

[0055] The liquid-cooled dummy load testing device in this embodiment simulates the physical distribution of heat-generating components such as the CPU and GPU inside a server. This makes the dummy load simulation components extremely close to the real server in terms of motherboard layout and heat generation. It ensures that the dummy load server's impact on the coolant temperature, flow direction, and flow rate during the test is basically consistent with that of the real server, thus replicating the real server to the greatest extent. The simulated heat-generating components are installed on the motherboard in a pluggable manner, and the position of the components can be flexibly adjusted, giving the testing device extremely high flexibility and reusability. This effectively reduces testing costs and improves the efficiency of batch testing.

[0056] Example 2

[0057] Based on the liquid-cooled dummy load testing device provided in Example 1, this application provides a complete liquid-cooled dummy load testing system, such as... Figure 4 As shown, the system includes: a power supply module 41, a data acquisition module 42, a control module 43, and a liquid-cooled dummy load testing device 44. The power supply module 41 supplies power to each simulated heat-generating component in the liquid-cooled dummy load testing device according to the power supply command issued by the control module 43. The data acquisition module 42 collects the test data of the liquid-cooled dummy load testing device and sends the test data to the control module. The control module 43 performs K-means cluster analysis on the test data and determines the management strategy for the target server simulated by the liquid-cooled dummy load testing device 44 based on the cluster analysis results.

[0058] Optionally, the control module can send a power supply command to the power supply module, which includes the set heating power for each simulated heating component; the power supply module will send the real-time heating power fed back by each simulated heating component back to the control module; the control module can monitor the operating status of each simulated heating component by comparing the real-time heating power and the set heating power, and when an abnormality is found in a simulated heating component, it can issue a prompt message to remind the test personnel to replace the simulated heating component.

[0059] Optionally, corresponding to the various sensors in the liquid-cooled dummy load testing device, the acquisition module includes: a pressure acquisition submodule, a temperature acquisition submodule, a liquid level acquisition submodule, and a flow rate acquisition submodule. The acquisition module can send the test data to the control module, which monitors the testing process of the liquid-cooled dummy load testing device and analyzes the test data.

[0060] Optionally, the control module can identify abnormal components in the target server based on the scatter plot corresponding to the cluster analysis results and issue an alarm message when an abnormal component is detected; it can also determine the coolant flow control strategy for each component in the target server based on the standard deviation of the cluster analysis results.

[0061] Specifically, the acquisition module first obtains temperature data during the testing process of the liquid-cooled dummy load test device. During the test cycle, the acquisition module collects temperature data of each simulated heat-generating component of the liquid-cooled dummy load test device according to a preset acquisition frequency, and sends the temperature data to the control module.

[0062] When performing K-means clustering analysis on temperature data, the following method can be used: At each acquisition time, filter the acquired temperature data whose temperature does not exceed a preset threshold, where the preset threshold can be set according to actual conditions; for each filtered temperature data, convert the temperature data into two-dimensional coordinates, where the horizontal axis of the two-dimensional coordinates is the distance between the position of the simulated heating component on the motherboard corresponding to the temperature data and the preset origin, and the vertical axis is the temperature corresponding to the temperature data; set K=4, and cluster all two-dimensional coordinates at the current acquisition time based on the four centroids obtained at the previous acquisition time and the preset clustering distance of each centroid, to obtain the clustering results, and draw a scatter plot corresponding to the clustering results. The clustering results include four new centroids, where at the initial acquisition time, the initial four centroids are determined as the two-dimensional coordinates corresponding to the temperature data of the four simulated heating components used to simulate hard drives, CPUs, GPUs, and memory; determine the standard deviation of the four centroids based on all clustering results throughout the entire test cycle.

[0063] After clustering is completed, the control module can determine that the server component simulated by the simulated heat-generating component corresponding to the two-dimensional coordinate that did not appear in the clustering results during the entire test cycle has an abnormal temperature and issue an alarm message; based on the standard deviation, it determines the coolant flow control strategy for each component in the target server. Among them, for the server component simulated by the simulated heat-generating component corresponding to the centroid with a larger standard deviation, the coolant flows through the server component more times and the flow rate is slower.

[0064] In this embodiment, the control module is the core of the entire system. On the one hand, it is connected to the acquisition module to obtain test data during the testing process. On the other hand, it is connected to the power module to send control commands to the liquid-cooled dummy load test device and obtain real-time power data, thereby monitoring the entire testing process. At the same time, combined with the K-means algorithm, the abnormal temperature field area is located based on the scatter plot mapped by the clustering results and an early warning information is issued. The final coolant flow control strategy is determined based on the standard deviation of the clustering analysis results, thereby solving the technical problem of lacking effective testing methods for direct contact liquid-cooled cabinets in related technologies.

[0065] Example 3

[0066] Based on the liquid-cooled dummy load test system provided in Embodiment 2, this application embodiment provides a liquid-cooled dummy load test method implemented by a control module. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than that shown here.

[0067] The control module can be a mobile terminal, a computer terminal, or a similar computing device. Figure 5 A hardware block diagram of a computer terminal (or mobile device) for implementing a liquid-cooled dummy load testing method is shown. Figure 5 As shown, a computer terminal 50 (or mobile device 50) may include one or more processors 502 (shown as 502a, 502b, ..., 502n in the figure) 502 (processor 502 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 504 for storing data, and a transmission device 506 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 5 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 50 may also include... Figure 5The more or fewer components shown, or having the same Figure 5 The different configurations shown.

[0068] It should be noted that the aforementioned one or more processors 502 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 50 (or mobile device). As described in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0069] The memory 504 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the liquid-cooled dummy load testing method in this embodiment. The processor 502 executes various functional applications and data processing by running the software programs and modules stored in the memory 504, thereby implementing the above-mentioned application vulnerability detection method. The memory 504 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 504 may further include memory remotely located relative to the processor 502, and these remote memories can be connected to the computer terminal 50 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0070] The transmission device 506 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 50. In one example, the transmission device 506 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 506 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0071] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 50 (or mobile device).

[0072] Under the above operating environment, this application provides a liquid-cooled dummy load testing method, such as... Figure 6 As shown, the method includes the following steps:

[0073] Step S602: Obtain temperature data during the test process of the liquid-cooled dummy load test device.

[0074] Optionally, temperature data during the testing process of the liquid-cooled dummy load test device can be obtained in the following way: during the test cycle, temperature data of each simulated heat-generating component during the testing process of the liquid-cooled dummy load test device are collected according to a preset acquisition frequency.

[0075] Step S604: Perform K-means cluster analysis on the temperature data to obtain the cluster analysis results.

[0076] To address the shortcomings of the K-means algorithm, such as the difficulty in determining the K value, sensitivity to outliers, and dependence of clustering results on the initialization of cluster centers, this application optimizes the algorithm in conjunction with a sham load testing scenario. Specifically, when performing K-means clustering analysis on temperature data, the following steps are included:

[0077] Step S6042: At each acquisition time, filter the acquired temperature data whose temperature does not exceed a preset threshold T. The preset threshold can be set according to the actual situation.

[0078] Step S6044: For each filtered temperature data, convert the temperature data into a two-dimensional coordinate (i, j), where the horizontal coordinate i is the distance between the position of the simulated heating component corresponding to the temperature data on the motherboard and the preset origin, and the vertical coordinate j is the temperature corresponding to the temperature data.

[0079] Step S6046: Periodically perform K-means clustering analysis, setting K=4. At the beginning of clustering, considering the characteristic that the server's largest heat-generating components are usually concentrated in four locations: CPU, GPU, hard drive, and memory, at the initial data acquisition time, determine the initial four centroids as the two-dimensional coordinates k1 corresponding to the temperature data of the four simulated heat-generating components used to simulate the hard drive, CPU, GPU, and memory. s (i k1 j k1 ), k2 s (i k2 j k2 ), k3 s (i k3 j k3 ), k4 s (i k4 j k4Based on the size of the heating components, the radius ranges d1, d2, d3, and d4 for each cluster are determined. Then, for each new round of clustering, all two-dimensional coordinates at the current acquisition time are clustered using the four centroids obtained from the previous acquisition time and the preset clustering distance for each centroid, yielding the clustering results. Each time the calculation stabilizes, four new centroids k1 in the clustering results are recorded. e k2 e k3 e k4 e The coordinates are used to plot a scatter plot, which visually displays the clustering results.

[0080] Step S6048: Determine the four centroids k1 based on all clustering results throughout the entire testing period. e k2 e k3 e k4 e The standard deviation.

[0081] Step S606: Determine the management strategy for the target server simulated by the liquid-cooled dummy load test device based on the cluster analysis results.

[0082] Throughout the entire dummy load test cycle, points not appearing in the clustering results can be identified; these points are the locations of temperature field anomalies. Simultaneously, multiple sets of k1 values ​​obtained during the entire test process... e k2 e k3 e k4 e Performing standard deviation analysis to measure the temperature field stability of server heat-generating components can provide guidance for optimizing coolant flow direction and velocity.

[0083] Optionally, when determining the management strategy for the target server simulated by the liquid-cooled dummy load test device based on the cluster analysis results: it can be determined that the server component simulated by the simulated heat-generating component corresponding to the two-dimensional coordinate that did not appear in the cluster results throughout the entire test cycle has an abnormal temperature and an alarm message is issued; it can also be determined based on the standard deviation to determine the coolant flow control strategy for each component in the target server, wherein, for the server component simulated by the simulated heat-generating component corresponding to the centroid with a larger standard deviation, the coolant flows through the server component more times and the flow rate is slower.

[0084] By combining the K-means clustering algorithm with the heat-generating components of the server through the above steps, abnormal monitoring points outside the cluster can be accurately located, early warning information can be issued for abnormal temperature field points, and by measuring the temperature field stability of the heat-generating components of the server, a control strategy for optimizing the flow direction and velocity of the coolant can be provided, thereby solving the technical problem of lacking effective testing methods for direct contact liquid-cooled cabinets in related technologies.

[0085] Example 4

[0086] According to an embodiment of this application, a non-volatile storage medium is also provided, which includes a stored computer program, wherein the device containing the non-volatile storage medium executes the liquid-cooled dummy load test method in Embodiment 3 by running the computer program.

[0087] Specifically, the device containing the non-volatile storage medium executes the following steps by running the computer program: acquiring temperature data during the testing process of the liquid-cooled dummy load test device; performing K-means cluster analysis on the temperature data to obtain the cluster analysis results; and determining the management strategy for the target server simulated by the liquid-cooled dummy load test device based on the cluster analysis results.

[0088] According to an embodiment of this application, a processor is also provided for running a computer program, wherein the computer program executes the liquid-cooled dummy load test method in Embodiment 1 during runtime.

[0089] Specifically, the computer program executes the following steps during runtime: acquiring temperature data during the testing process of the liquid-cooled dummy load test device; performing K-means cluster analysis on the temperature data to obtain the cluster analysis results; and determining the management strategy for the target server simulated by the liquid-cooled dummy load test device based on the cluster analysis results.

[0090] According to an embodiment of this application, an electronic device is also provided, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the liquid-cooled dummy load test method of Embodiment 1 through the computer program.

[0091] Specifically, the processor is configured to execute the following steps via a computer program: acquire temperature data during the testing process of the liquid-cooled dummy load test device; perform K-means cluster analysis on the temperature data to obtain the cluster analysis results; and determine the management strategy for the target server simulated by the liquid-cooled dummy load test device based on the cluster analysis results.

[0092] The sequence numbers of the above embodiments are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0093] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0094] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0095] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0096] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0097] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0098] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A liquid-cooled dummy load testing system, characterized in that, include: The system includes a power supply module, a data acquisition module, a control module, and a liquid-cooled dummy load testing device. The liquid-cooled dummy load testing device includes: a chassis, a motherboard located inside the chassis, and multiple simulated heat-generating components located on the motherboard. The motherboard includes multiple partitions, each partition including: a first fixing device and a first signal connection device for connecting to the simulated heat-generating components, and embedded wiring for power supply and signal transmission. Each simulated heat-generating component includes: a second fixing device and a second signal connection device for connecting to the motherboard, a temperature sensor, and a remotely controllable adjustable heating device. The simulated heat-generating components are used to simulate target components that generate heat in a server, and the target components include at least: a hard drive, a CPU, a GPU, and memory. The power module is used to supply power to each simulated heating component in the liquid-cooled dummy load test device according to the power supply command issued by the control module. The acquisition module is used to acquire test data from the liquid-cooled dummy load test device and send the test data to the control module. The control module is used to perform K-means clustering analysis on the test data, determine the abnormal components in the target server simulated by the liquid-cooled dummy load test device based on the scatter plot corresponding to the clustering analysis results, issue alarm information when abnormal components are found, and determine the coolant flow direction control strategy for each component in the target server based on the standard deviation of the clustering analysis results.

2. The system according to claim 1, characterized in that, The control module is used to send a power supply command to the power module, which includes a set heating power for each of the simulated heating components. The power module is used to send the real-time heating power fed back by each of the simulated heating components to the control module.

3. The system according to claim 1, characterized in that, The acquisition module includes at least one of the following: pressure acquisition submodule, temperature acquisition submodule, liquid level acquisition submodule, and flow rate acquisition submodule.

4. The system according to claim 1, characterized in that, The chassis dimensions are those of a standard server.

5. The system according to claim 1, characterized in that, The bottom of the chassis is equipped with a coolant drain pipe, and a temperature sensor and a liquid flow rate sensor are installed at the interface of the coolant drain pipe; a liquid level sensor is installed inside the chassis.

6. The system according to claim 1, characterized in that, The chassis has no top cover and is equipped with mounting ears.

7. The system according to claim 1, characterized in that, The embedded wire is connected to the first signal connection device in each of the partitions, and the type of the embedded wire includes at least one of the following: power supply line, sensor communication data line, and resistance control line; the chassis is provided with a wire outlet hole for the embedded wire.

8. The system according to claim 1, characterized in that, The motherboard is a rigid printed circuit board, and the motherboard is fixed to the base plate inside the chassis. The base plate is made of at least one of the following materials: bakelite board and mica board.

9. The system according to claim 1, characterized in that, Both the first fixing device and the second fixing device are neodymium iron boron high-strength magnets.

10. The system according to claim 9, characterized in that, Both the first signal connection device and the second signal connection device have metal contacts.

11. The system according to claim 10, characterized in that, The first signal connection device is surrounded by a groove, in which a rubber ring is placed, and the upper surface of the rubber ring is flush with the motherboard plane; the second signal connection device is surrounded by a groove, in which a rubber ring is placed, and the upper surface of the rubber ring is higher than the motherboard plane.

12. The system according to claim 10, characterized in that, The simulated heating component has a layered structure, wherein the temperature sensor is located on the upper layer of the second fixing device and the second signal connection device, and the adjustable heating device is located on the upper layer of the temperature sensor.

13. The system according to claim 12, characterized in that, A pressure sensor is also installed on the upper layer of the adjustable heating device.

14. The system according to claim 12, characterized in that, The first fixing device is located at the four corners of the partition, the first signal connection device is located at the center of the partition, the second fixing device is located at the four corners of the bottom layer of the simulated heating component, and the second signal connection device is located at the center of the bottom layer of the simulated heating component.

15. A liquid-cooled dummy load testing method, applied to the liquid-cooled dummy load testing system according to any one of claims 1 to 14, characterized in that, include: Acquire temperature data during the testing process of the liquid-cooled dummy load testing device; K-means clustering analysis is performed on the temperature data to obtain clustering results, including: at each acquisition time, filtering out data whose temperature does not exceed a preset threshold; for each filtered temperature data, converting the temperature data into two-dimensional coordinates, where the horizontal axis of the two-dimensional coordinates is the distance between the position of the simulated heating component corresponding to the temperature data on the motherboard and the preset origin, and the vertical axis is the temperature corresponding to the temperature data; setting K=4, clustering all two-dimensional coordinates at the current acquisition time based on the four centroids obtained at the previous acquisition time and the preset clustering distance for each centroid, obtaining clustering results, and plotting a scatter plot corresponding to the clustering results, the clustering results include four new centroids, wherein at the initial acquisition time, the initial four centroids are determined to be the two-dimensional coordinates corresponding to the temperature data of the four simulated heating components used to simulate hard disk, CPU, GPU and memory; the standard deviation of the four centroids is determined based on all clustering results throughout the entire test cycle; Based on the cluster analysis results, a management strategy for the target server simulated by the liquid-cooled dummy load testing device is determined, including: determining that the server component simulated by the simulated heat-generating component corresponding to the two-dimensional coordinates that did not appear in the cluster results throughout the entire testing cycle has an abnormal temperature and issuing an alarm message; and determining a coolant flow control strategy for each component in the target server based on the standard deviation, wherein, for the server component simulated by the simulated heat-generating component corresponding to the centroid with a larger standard deviation, the coolant flows through the server component more times and at a slower flow rate.

16. The method according to claim 15, characterized in that, Acquire temperature data during the testing process of the liquid-cooled dummy load test device, including: During the test period, temperature data of each simulated heat-generating component of the liquid-cooled dummy load test device are collected according to a preset acquisition frequency.

17. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored computer program, wherein the device containing the non-volatile storage medium executes the liquid-cooled dummy load test method of claim 15 or 16 by running the computer program.

Images