Server cable management method and device, equipment and storage medium

Abstract

The invention discloses a server cable management method and device, equipment and a storage medium, and relates to the technical field of fault detection, and the method comprises the steps: constructing a cable connection relation graph of a target cable of a target server based on a preset graph neural network; the target cable is a cable embedded in the controller and the target sensor in advance, and a neural processing unit is integrated in the controller; generating a cable code of the target cable, and determining a cable state of the target cable through the target sensor by using the controller; when the cable state is abnormal, determining a fault path of the target server according to the cable code and the cable connection relation graph; and displaying the fault path by using a positioning display device of the target cable so as to manage the target server. The controller and the sensor are implanted into the cable body, the cable connection relation graph is constructed in combination with the neural network model, cable troubleshooting and life management are further achieved through cable coding, the troubleshooting cost is greatly reduced, and the intelligent management process of a server is improved.

Description

Technical Field

[0001] The present application relates to the field of fault detection technology, and in particular to a server cable management method, apparatus, device, and storage medium. Background Art

[0002] With the advent of the big data era, the demand for AI (Artificial Intelligence) servers is growing exponentially. AI servers require a large number of cables and complex layouts. Traditional manual troubleshooting is time-consuming and has a high maintenance cost. In addition, most AI machines are heavy, making troubleshooting difficult. Furthermore, they rely on paper labels to trace connection relationships, requiring an average of more than five operation and maintenance records to be reviewed, making management more complex. Summary of the Invention

[0003] The present application provides a server cable management method, apparatus, device and storage medium to at least solve the problems in the related art of manual troubleshooting of server failures, which are time-consuming, high maintenance costs and complex management.

[0004] This application provides a server cable management method, including:

[0005] Based on a preset graph neural network, a cable connection relationship map of the target cable of the target server is constructed; the target cable is a cable pre-embedded with a controller and target sensor, and the controller has an integrated neural processing unit;

[0006] Generate a cable code corresponding to each target cable, and use a controller to determine the cable status of each target cable through a target sensor;

[0007] Determine whether the cable status is abnormal. If the cable status is abnormal, determine the fault path to the target server based on the cable code and cable connection relationship map;

[0008] The fault path is displayed using a preset positioning display device on the target cable, so that the target server can be managed based on the fault path.

[0009] The present application also provides a server cable management device, comprising:

[0010] A graph construction module is used to construct a cable connection relationship graph of the target cable of the target server based on a preset graph neural network; the target cable is a cable pre-embedded with a controller and a target sensor, and the controller has an integrated neural processing unit;

[0011] A status determination module is used to generate a cable code corresponding to each target cable and determine the cable status of each target cable through a target sensor using a controller;

[0012] A path determination module is used to determine whether the cable status is abnormal and, if the cable status is abnormal, determine the fault path to the target server based on the cable code and cable connection relationship map;

[0013] The path display module is used to display the fault path using a preset positioning display device on the target cable, so as to manage the target server based on the fault path.

[0014] The present application 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 server cable management methods when executing the computer program.

[0015] The present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned server cable management methods are implemented.

[0016] After embedding the controller and sensor into the cable, this application builds a cable connection relationship map of the server based on the graph neural network, and generates cable codes corresponding to each target cable, so as to use the controller to determine the corresponding cable status through the sensor. When the cable status is abnormal, the fault path is determined according to the cable code and the cable connection relationship map, so as to use the positioning display device on the cable to display the fault path. Through this application, embedded hardware and neural networks are integrated, and by implanting intelligent perception and execution units in the cable body and combining neural network models, a management system covering the entire life cycle of the cable is constructed, realizing closed-loop optimization from the physical layer to the management layer, and combining the cable connection relationship map and cable coding to realize cable troubleshooting and life management. For servers with complicated, large, and difficult-to-disassemble cables, the troubleshooting cost is greatly reduced, and the intelligent management process of the server is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application, 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] Figure 1 A flow chart of a server cable management method provided in an embodiment of the present application;

[0019] Figure 2 A schematic diagram of a server cable management system provided in an embodiment of the present application;

[0020] Figure 3 A schematic diagram of the structure of a flexible electronic cable provided in an embodiment of the present application;

[0021] Figure 4 A schematic diagram of the fault location process provided in an embodiment of the present application;

[0022] Figure 5 A flow chart of a specific server cable management method provided in an embodiment of the present application;

[0023] Figure 6 A schematic diagram of a cable life management process provided in an embodiment of the present application;

[0024] Figure 7 A schematic diagram of a cable dynamic switching process provided in an embodiment of the present application;

[0025] Figure 8 This is a structural diagram of a server cable management device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0027] It should be noted that, in the description of this application, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device 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 device. The terms "first," "second," etc., in this application are used to distinguish similar objects, and are not used to describe a particular order or sequence.

[0028] AI servers have a large number of cables and complex layouts. Traditional manual troubleshooting takes a long time and has a high maintenance cost. Most AI machines are heavy, making troubleshooting difficult. Furthermore, they rely on paper labels to trace connection relationships, requiring extensive review of operation and maintenance records, making management complex. This application addresses systemic challenges in AI servers, such as the surge in the number of cables (more than 200 cables per machine), low fault location efficiency (average fault location time exceeds 30 minutes), high manual maintenance costs (more than 500 yuan per operation and maintenance hour), and uncontrollable hidden losses (42% of failures are caused by bending fatigue) caused by clustered computing deployment. By implanting intelligent sensing and execution units into the flexible cable itself and combining them with a neural network-driven predictive maintenance model, we build a digital management system covering the entire cable life cycle, achieving closed-loop optimization from the physical layer to the management layer.

[0029] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0030] The embodiment of the present application provides a server cable management method. The method is described in detail in conjunction with the execution process of the server cable management method. Figure 1 As shown, the above method includes:

[0031] Step S11: Based on a preset graph neural network, a cable connection relationship graph of the target cable of the target server is constructed; the target cable is a cable pre-embedded with a controller and a target sensor, and a neural processing unit is integrated in the controller.

[0032] In this embodiment, it should be pointed out that the server includes but is not limited to an AI server with a three-dimensional layout (interweaving between cabinet layers, equipment rooms, and liquid cooling pipes). Based on this embodiment, the controller and target sensor can be pre-embedded in the cables of the server, and the neural processing unit can be integrated into the controller. It is understandable that the above-mentioned controller can specifically be a BMC (Baseboard Management Controller). And as Figure 2 As shown, this embodiment can be implemented based on a corresponding server cable management system. The system comprises an embedded master control unit based on a low-power MCU (microcontroller unit) chip, an addressing unit driven by binary coding and neural networks, a fault location unit, a neural network-driven cable management unit, a multi-dimensional lifespan management and intelligent switching unit, and an intelligent energy consumption adjustment unit. The MCU main control chip is controlled by the server CPU (central processing unit) and communicates with the CPU in real time. It also interacts with each module and provides feedback on decision-making solutions. Furthermore, the neural network-driven cable management unit can construct a topology-aware neural network to learn cable connection relationships and load distribution in real time, automatically generating an optimal cabling solution and adapting to changes in server topology (such as GPU (Graphics Processing Unit) cluster expansion). These modules can interact with the MCU in real time, enabling the MCU to control and provide feedback on multiple abnormal scenarios.

[0033] And the target cable in this embodiment is as follows Figure 3As shown, a multilayer circuit structure can be integrated using a polyimide substrate. Electrodes are then printed onto the flexible substrate using inkjet printing technology. An MCU chip and sensor module are then integrated to form a flexible electronic cable. For example, a low-power MCU chip (such as the Guoxin Technology CCR4001S) can be integrated into each cable. This chip utilizes the RISC-V architecture's CRV4H core (230MHz) and a 0.3 TOPS@INT8 NPU (Neural Processing Unit) to collect 12 health parameters in real time, including the cable's ID (unique code), physical location, temperature, bend angle, and current load. Using dynamic voltage and frequency scaling (DVFS) technology, the chip achieves operating power consumption as low as 1.2mW and power consumption less than 0.1μW in sleep mode, meeting the battery life requirements of the cable's entire lifecycle (5-8 years). The chip's built-in EEPROM can store 2048 historical fault logs and supports remote firmware upgrades via OTA (Over-the-Air Technology). In this way, using flexible materials as the base and combining modern transfer technology to encapsulate sensor modules, MCUs and high-speed signal lines, flexible packaging can greatly reduce signal loss caused by cable bending, while achieving the functions of flexible electronics while maintaining high flexibility.

[0034] Based on the above-mentioned cables, this embodiment can first construct a cable connection relationship map of the target cables of the target server based on a preset graph neural network. Specifically, this embodiment can use binary coding and a neural network-driven addressing unit to construct a topology management platform based on a graph neural network (GNN) to update the cable connection relationship map in real time. Among them, the model input layer contains an 8-dimensional feature vector (code, temperature, load, etc.), the hidden layer uses the GraphSAGE algorithm for feature aggregation, and the output layer outputs the optimal addressing path. When performing fault detection based on the above-mentioned relationship map, experimental data shows that the system's addressing speed is 8.7 times faster than traditional solutions, and the path planning energy consumption is reduced by 35%.

[0035] Step S12: Generate a cable code corresponding to each target cable, and use the controller to determine the cable status of each target cable through the target sensor.

[0036] In this embodiment, binary encoding and a neural network-driven addressing unit can be used to encode each target cable based on a preset fourth-order Gray code, generating a corresponding cable code for each target cable. After the cable code is generated, the cable code is displayed via a micro-display matrix on the target cable. In other words, this embodiment can first establish a Gray code identification system. For example, a cable can be uniquely encoded using a fourth-order Gray code (e.g., G4-0110), and the code segments can be dynamically displayed via a matrix of micro-LEDs (Light Emitting Diodes) on the cable surface. This encoding rule complies with the IEEE 11073 standard, ensuring compatibility with existing data center identification systems. This reduces the error rate of the neural network model learning topological relationships by 42%, due to the Gray code's characteristic of only one-bit difference between adjacent codes, thereby improving the accuracy of server fault monitoring.

[0037] Step S13: determine whether the cable status is abnormal, and if the cable status is abnormal, determine the fault path of the target server according to the cable code and the cable connection relationship map.

[0038] In this embodiment, the controller can be used to determine the temperature, bending angle, current and other conditions of each target cable through temperature sensors, bending angle sensors and current sensors. If the temperature and / or bending angle and / or current do not meet the preset threshold, the cable state is determined to be abnormal, and the corresponding fault type is determined based on the cable state. It is understandable that the above-mentioned sensors are not limited to temperature sensors, bending angle sensors and current sensors, and the sensor type integrated in the cable can be adjusted in real time according to actual needs. Accordingly, the controller can be used to send the cable code and cable state of the target cable to a preset neural network model, and then the preset neural network model is used to match the historical fault data corresponding to the cable state and cable code, and determine the fault type of the target cable based on the matching results, and determine the fault path of the target server based on the cable code and cable connection relationship map. In this way, the comprehensiveness of fault diagnosis is effectively improved by integrating physical, electrical and environmental parameters through multimodal sensor fusion, and monitoring various cable parameters through the MCU.

[0039] Specifically, the MCU chip in this embodiment monitors cable status in real time via an integrated temperature sensor (±0.1°C accuracy), a bend angle sensor (0.05° resolution), and a current sensor (16-bit ADC). When an anomaly is detected (such as a temperature >85°C or a bend angle >45°), a neural network model is immediately triggered to classify the fault type (short circuit, open circuit, poor contact, etc.) with an accuracy rate of 98.5%. It should be noted that this sensor module connects to the server motherboard via a USB (Universal Serial Bus) or I2C (Inter-Integrated Circuit) interface, requiring no additional power supply.

[0040] Step S14: Use the preset positioning display device on the target cable to display the fault path, so as to manage the target server based on the fault path.

[0041] In this embodiment, a pre-set positioning display device (such as an RGB LED positioning light strip) on the target cable can be used to display the fault path, allowing target servers to be managed based on the fault path. Specifically, this embodiment can employ an RGB LED positioning light strip (with one node deployed every centimeter) and utilize PWM (Pulse Width Modulation) modulation technology to illuminate the fault path within milliseconds. This system utilizes a three-dimensional spatial positioning algorithm, combined with Gray code encoding and the cable's physical coordinate system, to achieve a positioning error of <±1cm, enabling more accurate determination of the server's cable fault path. Furthermore, in the aforementioned technical solution, this embodiment utilizes a lightweight neural network model to reduce computing resource usage and is compatible with existing server management chips. The MCU is also a low-power chip, effectively reducing the resource burden on the server.

[0042] In a specific embodiment, Figure 4As shown, after the AI server is powered on, the X86 server cable intelligent management system based on the MCU chip and neural network model starts up. After startup, it begins to scan the cables at every location on the entire machine and calls the binary code and neural network-driven addressing unit to encode each cable. After the encoding is completed, troubleshooting is performed. If a feedback fault is encountered, the fault location process is initiated. Specifically, if the system detects an abnormal voltage at a node, it triggers the MCU chip to report the ID and sensor data. The network model then matches historical fault patterns and uses positioning light strips to highlight the fault path based on the matching results, allowing operation and maintenance personnel to quickly locate the fault point. If the match is successful, manual intervention is initiated. By simulating the neural network algorithm through adaptive neural networks and matching historical fault patterns, the system can adapt to dynamic environments through online learning and reduce false alarm rates.

[0043] It should also be pointed out that the use of remote communication modules such as Bluetooth / zigbee / wifi can realize clustered management of server cables. For example, a control center is deployed in the data center management server to realize visualization of the cable status of the entire cluster and cross-cabinet fault linkage processing through WiFi or 5G network. An industrial-grade gateway (such as Zigbee / WiFi gateway) is used to obtain the data of the communication nodes in the cabinet and upload it to the control center. At the same time, control instructions are issued to the communication nodes of each server. The above communication nodes can establish a connection with the cable MCU via Zigbee or Bluetooth. In this way, in specific applications, when a cable detects that the bending stress exceeds the standard, the system automatically checks the load of other cables on the same route. If multiple nodes are found to be abnormal, power derating or business migration is immediately triggered to avoid cascading failures. Or, when multiple cables in the cluster are close to the life threshold, the system automatically generates a purchase work order to reduce the cost of manual intervention.

[0044] In this embodiment, a controller and sensor can be embedded in the cable, and a graph neural network can be used to construct a server's cable connection relationship map. Then, a cable code corresponding to each target cable is generated. The controller then uses the sensor to determine the corresponding cable status. When the cable status is abnormal, the fault path is determined based on the cable code and the cable connection relationship map, and the fault path is displayed using the positioning display device on the cable. In this way, the cable integrates embedded hardware and neural networks, embeds a low-power MCU chip into the flexible base cable, and combines it with multiple sensors to monitor the cable status. The MCU can intelligently monitor the cable status throughout its lifecycle, achieving closed-loop optimization from the physical layer to the management layer. Furthermore, the cable connection relationship map and cable code are combined to implement cable troubleshooting and life management. When a fault occurs, the fault point can be quickly identified and illuminated. This significantly reduces troubleshooting costs for servers with complex, large, and difficult-to-disassemble cables. Furthermore, this embodiment utilizes general-purpose hardware and lightweight algorithms to achieve low cost and high compatibility, adapting to various X86 server architectures and improving the server's fault detection accuracy and intelligent management process.

[0045] Based on the previous embodiment, it can be seen that the present application can realize cable troubleshooting by implanting a controller and a yellowing device in the cable body and building a cable connection relationship map in combination with a neural network model. Next, the cable management process will be described in detail in this embodiment. Figure 5 As shown, the embodiment of the present application provides a specific server cable management method, including:

[0046] Step S21: Using a controller to determine the cable status of each target cable through a target sensor, and if the cable status is normal, determining the remaining life of the target cable based on the cable status.

[0047] In this embodiment, the controller can be used to determine the cable status of each target cable through the target sensor, and the remaining life of the target cable can be determined based on the cable status when the cable status is normal. Specifically, multi-dimensional life management and intelligent switch units can be used to integrate sensors such as temperature, bending angle, and current load, and predict the remaining life of the cable based on the LSTM (long short-term memory) algorithm (error <15%). Figure 6 As shown in the figure, if the system encounters no feedback faults during troubleshooting, it initiates the lifespan management process. For example, a temperature sensor continuously collects cable data and uses an LSTM model to calculate the remaining lifespan based on the collected data (e.g., 896 hours remaining, which is less than a preset threshold).

[0048] Step S22: If the remaining life is less than the preset cable life threshold, a corresponding warning message and a cable replacement list are generated, and the warning message is reported to the target server using the controller.

[0049] In this embodiment, if the remaining life of the cable is less than the preset cable life threshold, a corresponding warning message and a cable replacement list are generated, and the controller is used to report the warning message to the target server. Figure 6 As shown, the maintenance platform can push alerts and generate replacement work orders, and transmit specific values back to the MCU, which then reports to the host server for processing. Accordingly, if the remaining life is not less than the preset threshold, the real-time status of the cable is continuously monitored.

[0050] Step S23: monitor the real-time load of the computing node of the target server. If the real-time load is greater than the preset load threshold, use the controller to send the corresponding cable switching instruction to the target server, so that the target server adjusts the current target cable to the preset low power mode according to the cable switching instruction, and adjusts the corresponding standby cable of the target cable to the working mode.

[0051] In this embodiment, the real-time load of the computing node of the target server can also be monitored. If the real-time load is greater than the preset load threshold, the controller is used to send a corresponding cable switching instruction to the target server, so that the target server adjusts the current target cable to the preset low-power mode according to the cable switching instruction, and adjusts the corresponding spare cable of the target cable to the working mode. Specifically, multi-dimensional life management and intelligent switching units can be used to set intelligent switching modules for high-frequency use locations (such as the connection between PCIE (Peripheral Component Interconnect Express) devices and GPUs), and realize physical plug-and-unplug switching through signal modulation to reduce losses. Figure 7 As shown in the figure, if the PCIE at a specific position of the cable needs to be switched on and off and the power consumption is adjusted, the dynamic switching process is started: if the GPU load is detected to exceed the threshold, the controller sends a switching instruction to the intelligent switch, and the signal can be transmitted along the spare cable through QAM-16 modulation, while the original cable enters standby mode.

[0052] In this embodiment, a preset neural network model can also be used to predict the load fluctuations of each target cable, and based on the load fluctuations, it can be determined whether the target cable meets the preset cable switching conditions; if the target cable meets the preset cable switching conditions, the controller can be used to send the corresponding cable switching instructions to the target server. In other words, the energy consumption and loss optimization unit can use the neural network to predict load fluctuations and dynamically adjust the cable working mode. For example: during the GPU idle period, the system switches the backup cable to the working state, reducing the current load of the main cable (from 120A to 85A), while reducing the wear of the connector gold plating caused by frequent plugging and unplugging (the annual loss rate is reduced from 18% to 3%).

[0053] Based on the above technical solution, this embodiment utilizes an MCU chip to simulate a neural network for cable management. This intelligent, learnable model enables intelligent troubleshooting, significantly reducing operation and maintenance costs. This not only facilitates troubleshooting during testing and use, but also intelligently manages the entire cable lifecycle and enables intelligent, precise control, making the AI server cable system controllable and intelligent while reducing overall power consumption. Furthermore, the entire system intelligently implements cable troubleshooting, lifecycle management, energy optimization, and intelligent point-to-point switch control, making it extremely valuable for AI servers, which have complex, bulky, and difficult-to-install cables. This significantly reduces troubleshooting costs and enhances intelligent management processes.

[0054] like Figure 8 As shown, an embodiment of the present application further provides a server cable management device, comprising:

[0055] A graph construction module 11 is configured to construct a cable connection relationship graph of a target cable of a target server based on a preset graph neural network; the target cable is a cable pre-embedded with a controller and a target sensor, and the controller has an integrated neural processing unit;

[0056] A state determination module 12 is configured to generate a cable code corresponding to each target cable and determine the cable state of each target cable through a target sensor using a controller;

[0057] A path determination module 13 is used to determine whether the cable status is abnormal, and if the cable status is abnormal, determine the fault path of the target server based on the cable code and the cable connection relationship map;

[0058] The path display module 14 is configured to display the fault path using a preset positioning display device on the target cable, so as to manage the target server based on the fault path.

[0059] In this embodiment, the controller and sensors can be embedded in the cables, and a graph neural network can be used to construct a server's cable connection relationship map, as well as generate cable codes corresponding to each target cable. The controller can then use the sensors to determine the corresponding cable status. When the cable status is abnormal, the fault path is determined based on the cable code and the cable connection relationship map, and the positioning display device on the cable can be used to display the fault path. In this way, a management system covering the entire cable lifecycle is constructed in conjunction with a neural network model, achieving closed-loop optimization from the physical layer to the management layer. Furthermore, the cable connection relationship map and cable codes are combined to achieve cable troubleshooting and life management. This greatly reduces troubleshooting costs for servers with complex, large, and difficult-to-disassemble cables, and improves the intelligent management process of the server.

[0060] In some specific embodiments, the state determination module specifically includes:

[0061] A code generating unit, configured to encode each target cable based on a preset fourth-order Gray code to obtain a cable code;

[0062] Furthermore, the state determination module further includes:

[0063] The code display unit is used to display the cable code through the micro display matrix of the target cable.

[0064] In some specific embodiments, the state determination module specifically includes:

[0065] A state determination unit, configured to determine the cable state of each target cable using a temperature sensor, a bending angle sensor, and a current sensor using a controller; the cable state includes the temperature, bending angle, and current of the target cable;

[0066] Correspondingly, the path determination module specifically includes:

[0067] The fault determination submodule is configured to determine that the cable state is abnormal if the temperature and / or bending angle and / or current do not meet a preset threshold, and to determine a corresponding fault type based on the cable state.

[0068] In some specific embodiments, the fault determination submodule specifically includes:

[0069] A state sending unit, configured to send the cable code and cable state of the target cable to a preset neural network model using a controller;

[0070] The data matching unit is used to match the cable status and the historical fault data corresponding to the cable code using a preset neural network model, and determine the fault type of the target cable based on the obtained matching result.

[0071] In some specific embodiments, the path determination module specifically includes:

[0072] a lifespan determination unit, configured to determine a remaining lifespan of a target cable based on the cable status if the cable status is normal;

[0073] The information reporting unit is used to generate corresponding warning information and cable replacement list if the remaining life is less than the preset cable life threshold, and use the controller to report the warning information to the target server.

[0074] In some specific embodiments, the server cable management device further includes:

[0075] Load monitoring module, used to monitor the real-time load of the computing nodes of the target server;

[0076] The first instruction sending module is used to use the controller to send the corresponding cable switching instruction to the target server if the real-time load is greater than the preset load threshold, so that the target server adjusts the current target cable to the preset low power mode according to the cable switching instruction, and adjusts the corresponding backup cable of the target cable to the working mode.

[0077] In some specific embodiments, the server cable management device further includes:

[0078] A fluctuation prediction module is used to predict the load fluctuation of each target cable using a preset neural network model;

[0079] The second instruction sending module is used to determine whether the target cable meets the preset cable switching condition based on the load fluctuation; if the target cable meets the preset cable switching condition, the controller is used to send a corresponding cable switching instruction to the target server.

[0080] For the description of the features in the embodiment corresponding to the server cable management device, reference can be made to the relevant description of the embodiment corresponding to the server cable management method, which will not be repeated here.

[0081] 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.

[0082] An embodiment of the present application further provides an electronic device, including 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 of any of the above server cable management method embodiments.

[0083] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above-mentioned server cable management method embodiments when running.

[0084] 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.

[0085] An embodiment of the present application 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 server cable management method embodiments are implemented.

[0086] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above-mentioned server cable management method embodiments are implemented.

[0087] 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 this application.

[0088] The above is a detailed introduction to the server cable management method, device, equipment and storage medium provided by this application. This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of this application, several improvements and modifications can be made to this application, and these improvements and modifications also fall within the scope of protection of the claims of this application.

Claims

1. A server cable management method, characterized in that: include: Based on the preset graph neural network, a cable connection relationship graph of the target cable of the target server is constructed; The target cable is a cable pre-embedded with a controller and a target sensor, and the controller is integrated with a neural processing unit; generating a cable code corresponding to each of the target cables, and determining a cable status of each of the target cables using the target sensor by using the controller; determining whether the cable status is abnormal, and if the cable status is abnormal, determining a fault path of the target server according to the cable code and the cable connection relationship map; The fault path is displayed using a preset positioning display device on the target cable, so as to manage the target server based on the fault path.

2. The server cable management method according to claim 1, characterized in that: Generating a cable code corresponding to each target cable includes: Encoding each of the target cables based on a preset fourth-order Gray code to obtain the cable code; Furthermore, after generating the cable codes corresponding to the target cables, the method further includes: The cable code is displayed via a micro display matrix of the target cable.

3. The server cable management method according to claim 2, characterized in that: The determining the cable status of each target cable by using the controller through the target sensor includes: Determining the cable status of each target cable using the controller through a temperature sensor, a bending angle sensor, and a current sensor; the cable status includes the temperature, bending angle, and current of the target cable; Accordingly, determining whether the cable status is abnormal includes: If the temperature and / or the bending angle and / or the current do not meet a preset threshold, the cable state is determined to be abnormal, and a corresponding fault type is determined based on the cable state.

4. The server cable management method according to claim 3, characterized in that: The determining a corresponding fault type based on the cable status includes: Using the controller, sending the cable code and the cable status of the target cable to a preset neural network model; The preset neural network model is used to match the cable status with the historical fault data corresponding to the cable code, and the fault type of the target cable is determined based on the obtained matching result.

5. The server cable management method according to claim 1, wherein: The determining whether the cable status is abnormal includes: If the cable status is normal, determining the remaining life of the target cable based on the cable status; If the remaining life is less than a preset cable life threshold, corresponding warning information and a cable replacement list are generated, and the controller is used to report the warning information to the target server.

6. The server cable management method according to any one of claims 1 to 5, characterized in that: Also includes: Monitoring the real-time load of the computing nodes of the target server; If the real-time load is greater than the preset load threshold, the controller is used to send a corresponding cable switching instruction to the target server, so that the target server adjusts the current target cable to the preset low power consumption mode according to the cable switching instruction, and adjusts the corresponding backup cable of the target cable to the working mode.

7. The server cable management method according to claim 6, characterized in that: Also includes: Predicting the load fluctuation of each target cable using a preset neural network model; determining whether the target cable meets a preset cable switching condition based on the load fluctuation; If the target cable meets the preset cable switching condition, the controller is used to send a corresponding cable switching instruction to the target server.

8. A server cable management device, characterized in that: include: A graph construction module, used to construct a cable connection relationship graph of the target cable of the target server based on a preset graph neural network; The target cable is a cable pre-embedded with a controller and a target sensor, and the controller is integrated with a neural processing unit; a state determination module, configured to generate a cable code corresponding to each of the target cables, and determine the cable state of each of the target cables using the target sensor by using the controller; a path determination module, configured to determine whether the cable status is abnormal, and if the cable status is abnormal, determine the fault path to the target server based on the cable code and the cable connection relationship map; A path display module is used to display the fault path using a preset positioning display device on the target cable, so as to manage the target server based on the fault path.

9. An electronic device, characterized in that: include: memory for storing computer programs; A processor is configured to implement the steps of the server cable management method according to any one of claims 1 to 7 when executing the computer program.

10. 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 server cable management method according to any one of claims 1 to 7 are implemented.

Images