Heat dissipation control method and electronic device

Abstract

The application discloses a heat dissipation control method and electronic equipment, and relates to the technical field of heat dissipation. The method comprises the following steps: reading the identification information of each component in the electronic equipment; determining the first heat dissipation strategy corresponding to each component according to a preset global mapping table and executing the first heat dissipation strategy; monitoring the running state of the electronic equipment; when it is monitored that the electronic equipment enters a low-power consumption state, determining the in-place state and power supply state of each component; determining the current effective configuration of the electronic equipment based on the in-place state and power supply state of each component; determining the adjustment coefficient of each working component based on the current effective configuration; adjusting the first heat dissipation strategy corresponding to each working component according to the adjustment coefficient to determine the second heat dissipation strategy corresponding to each working component; and solving the technical problem that a control element depends on firmware to acquire component information through bus enumeration in an initialization stage, so that the technical effect of accurately and automatically matching the heat dissipation strategy under the whole life cycle and the whole hardware state of the electronic equipment is achieved.

Description

Technical Field

[0001] This application relates to the field of heat dissipation technology, and in particular to a heat dissipation control method and electronic device. Background Technology

[0002] In related technologies, the control elements of electronic devices rely on the main processor or firmware to obtain component identification information through bus enumeration during the initialization phase, and then match it with a preset heat dissipation strategy. This mechanism has significant limitations in dynamic and low-power scenarios: when the system is in a low-power state such as soft shutdown, the main processor is unavailable, and the control elements cannot identify specific components. They can only adopt a conservative strategy based on a uniform high temperature threshold or a fixed speed mode, which leads to excessive heat dissipation for high-temperature resistant components, increased energy consumption, or overheating risk due to insufficient heat dissipation when components suddenly experience high loads.

[0003] Therefore, in view of the shortcomings of related technical solutions, the present invention provides a heat dissipation control method. Summary of the Invention

[0004] This application provides a heat dissipation control method and an electronic device to at least solve the problem in the related art where the control elements of electronic devices rely on the main processor or firmware to obtain component identity information through bus enumeration during the initialization phase.

[0005] This application provides a heat dissipation control method, the method comprising: reading the identification information of each component in an electronic device; determining and executing a first heat dissipation strategy corresponding to each component according to a preset global mapping table; monitoring the operating status of the electronic device; determining the presence and power supply status of each component in response to detecting that the electronic device has entered a low-power state; determining the current effective configuration of the electronic device based on the presence and power supply status of each component, wherein the current effective configuration includes working components and power-off components; determining the adjustment coefficient of each working component based on the current effective configuration; adjusting the first heat dissipation strategy corresponding to each working component according to the adjustment coefficient to determine a second heat dissipation strategy corresponding to each working component; performing heat dissipation control on each working component according to the second heat dissipation strategy, and continuously monitoring the status of each working component.

[0006] This application also provides an electronic device, including: a memory for storing a computer program; and a processor for implementing the steps of any of the above-described heat dissipation control methods when executing the computer program.

[0007] This application also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of any of the above-described heat dissipation control methods.

[0008] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the above-described heat dissipation control methods.

[0009] This application achieves the following: by reading the identification information of each component within the electronic device, determining and executing the first heat dissipation strategy corresponding to each component based on a preset global mapping table; monitoring the operating status of the electronic device, and determining the presence and power supply status of each component when the electronic device enters a low-power state; determining the current effective configuration of the electronic device based on the presence and power supply status of each component, wherein the current effective configuration includes working components and power-off components; determining the adjustment coefficient of each working component based on the current effective configuration, adjusting the first heat dissipation strategy corresponding to each working component according to the adjustment coefficient, and determining the second heat dissipation strategy corresponding to each working component; performing heat dissipation control on each working component according to the second heat dissipation strategy, and continuously monitoring the status of each working component. Therefore, the control element can identify components independently of the host firmware by actively reading the identification information. Based on this, through the mechanisms of status awareness, configuration synchronization, dynamic weighting, and real-time hot-plug response, accurate and automatic matching of heat dissipation strategies throughout the entire lifecycle and all hardware states of the electronic device is achieved. Attached Figure Description

[0010] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A schematic flowchart illustrating a heat dissipation control method provided in an embodiment of this application; Figure 2 A schematic diagram of the global mapping table deployment process of a heat dissipation control method provided in an embodiment of this application; Figure 3 A schematic diagram of the global mapping table update process of a heat dissipation control method provided in an embodiment of this application; Figure 4 A schematic flowchart illustrating another heat dissipation control method provided in an embodiment of this application; Figure 5 A structural block diagram of a heat dissipation control device provided in an embodiment of this application; Figure 6 This is an internal structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0013] It should be noted that, in the description of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. The terms "first," "second," etc., in this application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0014] It should be noted that the terms "S1," "S2," etc., are used only for descriptive purposes and do not specifically refer to the order or sequence, nor are they intended to limit this application. They are merely for the convenience of describing the method of this application and should not be construed as indicating the sequential order of the steps. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0015] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0016] The embodiments of this application provide a heat dissipation control method, and the method is described in detail below in conjunction with the execution flow of the heat dissipation control method.

[0017] S101: Read the identification information of each component in the electronic device, determine the first heat dissipation strategy corresponding to each component according to the preset global mapping table, and execute it.

[0018] Here, electronic devices can be servers, switches, storage devices, personal computers, and other devices with heat-generating components and control elements (BMCs).

[0019] Here, the executing entity is the control element. For example, when the electronic device is a server, the corresponding control element can be a BMC (Baseboard Management Controller), a dedicated microcontroller integrated on the server motherboard, responsible for out-of-band management, health monitoring, fan control, etc.

[0020] Among them, BMC is responsible for independently identifying hardware, matching heat dissipation strategies, and executing heat dissipation control.

[0021] This information can be read via the system bus.

[0022] The system bus can be I2C (Inter-Integrated Circuit, two-wire serial bus) or SMBus (System Management Bus).

[0023] The identification information can be the part number (PN, Product Number), the supplier identification code (VID, Vendor ID), and / or the device identification code (DID, Device ID).

[0024] The components may include OCP (Open Compute Project NIC), smart NIC (Network Interface Card) / DPU (Data Processing Unit), and some NVMe SSDs (Non-Volatile Memory Express Solid State Drives).

[0025] For example, the global mapping table may include a network card with PN as NIC-Intel-X710-AT2, VID as 0x8086, and DID as 0x15F2, whose first heat dissipation strategy is to issue an alarm when the ambient temperature reaches 70℃; and a network card with PN as NIC-Nvidia-MCX-5, VID as 0x15B3, and DID as 0x1019, whose first heat dissipation strategy is to issue an alarm when the ambient temperature reaches 85℃.

[0026] Specifically, assuming the electronic device is a server, the BMC directly reads the component serial number (PN) from the target component's storage chip (such as EEPROM) via the I2C / SMBus bus. The BMC queries the global mapping table in the local firmware using the PN. If the identification is successful, it obtains the corresponding VID / DID, and then loads the model-specific precise heat dissipation parameters (such as an alarm temperature threshold of 85°C) from the policy library, combining them with real-time temperature sensor data for heat dissipation control.

[0027] In one embodiment, the method further includes: If the component's identifier is not in the global mapping table, obtain the component's limited identity information, which includes at least one of the following: slot type, physical size, power consumption level indication information, and descriptive string; Determine the basic component category of the component based on the slot type and / or physical dimensions; Determine the estimated thermal load level of the component based on the current ambient temperature and / or power consumption level indication information; Using the basic component category and estimated heat load level as indexes, the corresponding initial fan speed control curve is matched from the preset gradient safety strategy library as the fourth heat dissipation strategy. During the operation of electronic devices, the temperature changes of components are continuously monitored, and the control parameters of the safe heat dissipation strategy are adjusted based on a predefined learning algorithm. The adjusted secure heat dissipation strategy is associated with limited identity information and persistently stored as a temporary strategy.

[0028] Here, limited identity information refers to a set of basic attributes that can be directly read through physical interfaces, standard bus protocols, or physical tags when the component cannot be accurately identified through a preset global mapping table.

[0029] Here, the gradient security policy library is a pre-built thermal strategy matrix or decision tree in the BMC firmware, indexed by multi-dimensional conditions. It is divided according to the basic component category (such as GPU, FPGA, general network card) and the estimated heat load level (such as low, medium, high). Each cell corresponds to a set of preset, relatively conservative but targeted fan speed control curves and temperature alarm thresholds.

[0030] The learning algorithm can be a simple rule engine (such as "if the temperature is lower than the threshold X% for three consecutive cycles, the speed will be reduced by Y%)" or a more complex model fitting (such as recording the "temperature-load-speed" data points and fitting the approximate thermal characteristic curve of the component).

[0031] In this way, not only is the handling of unknown components improved, but this handling is also transformed from costly safety redundancy into value-creating energy efficiency optimization and intelligent operation and maintenance.

[0032] S102: Monitor the operating status of electronic devices, and in response to detecting that the electronic devices have entered a low-power state, determine the presence and power supply status of each component.

[0033] Here, the low-power state can be S5, the soft shutdown state, in which the BIOS (Basic Input / Output System) is not started.

[0034] In related technologies, the BMC relies on the host BIOS to enumerate component identification information through PCIe (Peripheral Component Interconnect Express) during the boot phase, and then matches it with the preset cooling strategy. In S5 state, the BMC cannot obtain the VID / DID information of the component. Therefore, the BMC can only perform heat dissipation control on the component according to the default cooling strategy of the unknown model.

[0035] The in-situ status detection methods may include detection pins, I2C / SMBus device enumeration (detection via device address), PCIe / USB bus enumeration, power supply or load detection, mechanical switches / micro switches, GPIO (General-Purpose Input / Output) status reading (with pull-up / pull-down resistors), and BMC / IPMI (Intelligent Platform Management Interface) active querying, etc.

[0036] The power supply status detection methods may include power rail voltage detection, power enable / PG signal detection, load current detection, communication bus enumeration and response, and dedicated power supply status pins.

[0037] S103: Based on the in-situ status and power supply status of each component, determine the current effective configuration of the electronic device, wherein the current effective configuration includes working components and power-off components.

[0038] Here, a working component is a hardware component that is recognized by the system, successfully powered on, in normal operating condition, and accessible and usable by the operating system or management controller (such as BMC).

[0039] Here, the power-off component is in place, but is currently not powered, not activated, or actively shut down.

[0040] Specifically, when an electronic device enters the ACPI S5 soft shutdown state, the BMC first triggers a hardware presence detection (such as by detecting the PRSNT signal of the PCIe slot) and a power supply detection. The detection results are compared with the stored runtime configuration snapshot to update the current effective configuration used for decision-making, and this configuration is then permanently stored again, overwriting the old snapshot.

[0041] Specifically, if a component is found to be missing, it will be removed from the current valid configuration; if an unknown component is found to be inserted, it will be marked as pending verification and a security policy will be applied.

[0042] S104: Based on the current effective configuration, determine the adjustment coefficient of each working component, adjust the first heat dissipation strategy corresponding to each working component according to the adjustment coefficient, and determine the second heat dissipation strategy corresponding to each working component.

[0043] The heat dissipation strategy is dynamically adjusted for each working component.

[0044] S105: Perform heat dissipation control on each working component according to the second heat dissipation strategy, and continuously monitor the status of each working component.

[0045] It should be noted that this application enables electronic devices to match the most suitable heat dissipation strategy for each in-situ component throughout their entire life cycle and under all hardware conditions, significantly improving the accuracy and safety of heat dissipation control. At the same time, through a dynamic weighting mechanism, it can calculate heat dissipation requirements based on the actual load and configuration of the effective working components, thereby reducing fan energy consumption and operating noise.

[0046] In some specific implementations, the method further includes: Obtain the first, second, and third identifiers of each component; Establish a relational database that uses the first identifier as an index and associates it with the second and third identifiers; Based on the model of the electronic device, relevant mapping relationships are extracted from the association database to generate a global mapping table corresponding to the model; The firmware containing the global mapping table is burned into the electronic device.

[0047] Here, the first identifier can be the component serial number. This is a unique string code that the manufacturer prints on the physical label of the hardware and stores in the hardware's own non-volatile memory (such as EEPROM). It is a physical, directly readable identifier, and is the "identity credential" that the BMC can directly obtain when it is separated from the BIOS.

[0048] Here, the second identifier can be a supplier identification code. This is a unique code assigned by an industry standards organization (such as PCI-SIG) to identify the hardware manufacturer. It is usually used in conjunction with a third identifier to uniquely identify the device model at the software and driver levels.

[0049] Here, the third identifier can be the device identification code. This is an identifier assigned by the hardware supplier to its different product models. Together with the second identifier, it forms a globally unique model ID card within the industry standard system.

[0050] Here, the relational database is a centralized database established and maintained in the R&D, production, or operations backend. Its core data structure uses a first identifier as the key index, and each record is associated with a corresponding second and third identifier. For example: (PN: "NIC-Intel-X710-AT2", VID: 0x8086, DID: 0x15F2).

[0051] Here, the global mapping table is a subset of mapping relationships for all hardware components supported or potentially used by a specific electronic device model, filtered and extracted from the relational database.

[0052] Specifically, after the firmware is burned into the memory of the electronic device, the global mapping table contained therein becomes part of the BMC firmware, which can be directly accessed and queried by the BMC software after the device is powered on.

[0053] Specifically, methods for obtaining component identifiers can include manual entry and physical prototype verification. Manual entry involves maintenance personnel adding (PN, VID, DID) mapping records directly to the central database based on technical documentation for components with clearly defined specifications. Physical prototype verification involves inserting a component into a general-purpose verification prototype for components requiring physical verification. Upon power-on, the prototype's BIOS acquires and reports the component information. The system automatically captures this (PN, VID, DID) information, compares it with the central database, and completes the addition or verification. If verification fails, an alarm is triggered, and manual verification is notified.

[0054] For example, assuming the electronic device is a server, Figure 2 This is a schematic diagram of the global mapping table deployment process in an embodiment of this application, such as... Figure 2 As shown, the global mapping table deployment process in this application includes: constructing a global mapping table to start the BMC component process; selecting a new component information entry method; in response to manual entry, extracting PN, VID, and DID data according to the component specification document; in response to physical prototype verification, inserting the new component into a general verification prototype, powering on the prototype, and having the BIOS perform standard enumeration to capture PN, VID, and DID, and receiving the data; collecting the entered data and verifying whether PN exists in the mapping table database; if not, adding a new record: PN-VID / DID; if it exists, verifying whether VID / DID is consistent; if yes, the record verification passes; if no, the record conflicts and triggers a manual review alarm; generating a new version of the mapping table data file based on the record; compiling BMC firmware for a specific server model; compiling and linking the mapping table file into the firmware; releasing BMC firmware containing the new version of the mapping table; mass-producing and flashing it to all servers of that model; and completing the global mapping table deployment.

[0055] In this way, by establishing a central database and flashing the firmware, the consistency and feasibility of the solution at the product level are ensured, avoiding the chaos of each electronic device learning on its own, and ensuring that it has precise heat dissipation capabilities from the factory.

[0056] In one embodiment, if new components are introduced after the product has left the factory, the method further includes: Receive the first, second, and third identifiers of the new identification component identified and sent by the firmware via the out-of-band management interface; Compare the first identifier with the global mapping table; If the first identifier does not exist in the global mapping table, the identity information of the newly identified component is added to the global mapping table as a new mapping relationship.

[0057] Here, the out-of-band management interface refers to a dedicated management channel independent of the electronic device's main operating system and CPU. This interface allows the BMC to communicate and exchange data with external management software or other internal firmware (such as BIOS) even when the host operating system is not running (such as when the device is powered off or starting up) or is malfunctioning.

[0058] The out-of-band management interface can be IPMI.

[0059] Here, firmware is the Basic Input / Output System (BIOS) or Unified Extensible Firmware Interface (UEFI) that runs on the main processor of the electronic device and is responsible for hardware initialization and enumeration during the initial system startup.

[0060] Among them, the newly identified parts can be parts that are inserted for the first time after the last BMC update of the mapping table, or parts that existed before but were not learned by the BMC for some reason.

[0061] The new mapping relationship refers to a complete association record of first identifier -> (second identifier, third identifier).

[0062] Specifically, the BMC receives information including the Part Serial Number (PN), Vendor ID (VID), and Device ID (DID) from the BIOS via an out-of-band management interface (such as IPMI). The BMC uses the received PN as a reference in its locally stored global mapping table. If the PN does not exist in the mapping table, a new (PN, VID, DID) mapping is added to the table and persistently stored in the locally stored global mapping table, while simultaneously recording a new device record for subsequent maintenance and verification; otherwise, it is ignored.

[0063] For example, assuming the electronic device is a server, Figure 3 This is a schematic diagram of the global mapping table update process in an embodiment of this application, as shown below. Figure 3As shown, the global mapping table update process in this application includes: the server powers on and starts up; the BIOS completes hardware enumeration; the BIOS reports component information through interfaces such as IPMI; the BMC receives the reported information; the BMC extracts the PN from the information as a query key; the BMC queries the locally stored global mapping table; it determines whether the PN already exists in the mapping table; if so, the record already exists, the reported information is ignored, and no operation is performed; if not, it is identified as a new component, and the new mapping relationship is added to the global mapping table; the updated global mapping table is persistently stored in non-volatile memory; a "new device learning" log is recorded for maintenance and verification; the global mapping table has been updated.

[0064] In this way, based on the static knowledge base, BMC is allowed to learn and record newly emerging components during operation, dynamically expanding its recognition capabilities.

[0065] In some specific implementations, Based on the current effective configuration, the adjustment coefficients for each working component are determined. According to these adjustment coefficients, the first heat dissipation strategy corresponding to each working component is adjusted, and the second heat dissipation strategy corresponding to each working component is determined, including: Determine whether each working component has a temperature sensor; In response to the absence of a temperature sensor in the working component, a first adjustment coefficient is determined based on the current ambient temperature and the current effective configuration. Based on the first adjustment coefficient, the first heat dissipation strategy corresponding to each working component is adjusted, and a second heat dissipation strategy corresponding to each working component is determined. In response to the presence of a temperature sensor in the working component, the component temperature of the working component is obtained. Combined with the current effective configuration, a second adjustment coefficient is determined. Based on the second adjustment coefficient, the first heat dissipation strategy corresponding to each working component is adjusted, and a second heat dissipation strategy corresponding to each working component is determined.

[0066] Here, the current effective configuration is a dynamically updated data structure that records the actual operating status and physical state of all hardware under the current system state (e.g., S5). It not only records "what hardware is available," but more importantly, it records "who is the effective heat source" and "who is obstructing airflow." For example: {Effective heat source: [Network card A], Airflow obstruction: [Power-off GPU B], Ambient temperature: 25°C}.

[0067] Here, the temperature sensor is a thermistor integrated inside the component, used to monitor the temperature of its chip or critical areas in real time. Its readings can be read by the control element via management buses such as I2C / SMBus, and are the direct basis for the control element to perform precise temperature regulation.

[0068] The first adjustment factor is applied to components without temperature sensors, and the second adjustment factor is applied to components with temperature sensors.

[0069] The second heat dissipation strategy is the first heat dissipation strategy after contextual adjustments.

[0070] In this way, the flow is split based on whether a sensor is present or not, making the weighted calculation more targeted and feasible.

[0071] In some specific embodiments, in response to the absence of a temperature sensor in a working component, a first adjustment coefficient is determined based on the current ambient temperature and the current effective configuration. Based on the first adjustment coefficient, a first heat dissipation strategy corresponding to each working component is adjusted, and a second heat dissipation strategy corresponding to each working component is determined, including: Determine the number of working parts based on the current available configuration; The first adjustment coefficient is determined based on the number of working parts and the total number of pre-stored parts; Determine the basic heat dissipation requirements based on the current ambient temperature; Based on the first adjustment coefficient and the basic heat dissipation requirements, the first heat dissipation strategy corresponding to each working component is adjusted, and the second heat dissipation strategy corresponding to each working component is determined.

[0072] Here, the first adjustment factor, a numerical coefficient between 0 and 1 (usually denoted as α), quantifies the proportion of the decrease in heat dissipation demand due to the reduction in the number of currently active working components. The smaller the α value, the greater the reduction in required heat dissipation intensity relative to the full-load baseline.

[0073] Here, the parameters in the second heat dissipation strategy are the specific command values ​​that ultimately drive the heat dissipation actuator (usually a fan), such as the duty cycle of the fan speed (e.g., 30%) or the target speed value (e.g., 3000 RPM).

[0074] The basic cooling requirement is a value calculated based on the current ambient temperature by consulting a predefined "ambient temperature - basic cooling requirement" lookup table or through a simple linear model. For example, at an ambient temperature of 25°C, the basic requirement corresponds to a fan speed of 20%; at 30°C, it corresponds to 25%.

[0075] The first adjustment coefficient can be calculated by dividing the current number of working parts by the total number of pre-stored parts. The formula can incorporate adjustment factors or nonlinear mapping as needed.

[0076] The control parameters for the second heat dissipation strategy can be obtained by multiplying the first adjustment coefficient by the basic heat dissipation requirement.

[0077] In this way, even if the precise temperature of the components cannot be obtained, the ambient temperature requirement can be reasonably reduced based on the proportion of effective working components, thus achieving energy saving under the premise of safety.

[0078] In some specific implementations, the method further includes: Obtain the preset chassis airflow topology model; Based on the current effective configuration, determine the physical locations of the working components and the power-off components; Based on the preset chassis airflow topology model, the physical location of the working component and the physical location of the power-off component, the blocking component corresponding to the working component is determined, wherein the blocking component is at least one power-off component located in the same airflow as the working component.

[0079] Here, the chassis airflow topology model is a digital model established during the electronic device design phase, describing the airflow path inside the chassis. It defines: Airflow duct: the main path of airflow, usually from the front air intake to the rear exhaust duct, which may include multiple parallel branch airflow ducts; Relationship between component location and airflow duct: which airflow duct each hardware installation location (such as PCIe slots 1 and 2, hard drive backplane) is located on, and its relative upstream and downstream order in that airflow duct; Blocking potential: the degree to which different components, due to their physical size and shape, may obstruct airflow (e.g., a full-height, full-length GPU card is a strong obstruction, while a half-height network card is a weak obstruction).

[0080] Here, physical location refers to the specific installation coordinates of the component within the chassis.

[0081] Here, "blocking components" refers to power-off components that are located in the same air duct as a specific working component and are upstream of that working component.

[0082] In some specific implementations... In response to the presence of a temperature sensor on a working component, the component temperature is acquired. Based on the current effective configuration, a second adjustment coefficient is determined. According to the second adjustment coefficient, the first heat dissipation strategy corresponding to each working component is adjusted, and a second heat dissipation strategy corresponding to each working component is determined, including: Based on the component temperature of the working component, the initial heat dissipation requirements of the working component are determined through a global mapping table; Based on the preset chassis airflow topology model, the physical location of the working components and the physical location of the blocking components, a second adjustment coefficient for the heat dissipation efficiency of the working components by the blocking components is determined. Based on the second adjustment coefficient and the initial heat dissipation requirements, the first heat dissipation strategy corresponding to each working component is adjusted, and the second heat dissipation strategy corresponding to each working component is determined.

[0083] Here, the initial heat dissipation requirement is a theoretical heat dissipation requirement calculated solely based on the real-time temperature of the working components and their own thermal characteristics, without considering the influence of the current specific chassis airflow environment.

[0084] Here, the second adjustment coefficient is a quantified attenuation coefficient (usually denoted as β, where 0 < β ≤ 1), which is the environmental correction factor of this algorithm. It accurately characterizes the degree of negative impact of the obstructed component on the airflow efficiency towards a specific working component. The closer the β value is to 1, the smaller the obstruction effect; the smaller the β value, the more severe the airflow obstruction and the lower the airflow efficiency.

[0085] Among them, the control parameters of the second heat dissipation strategy can be the drive command values ​​that have been modified by the environment and will eventually be sent to the heat dissipation actuator (fan), such as the target speed or duty cycle.

[0086] Specifically, the chassis duct topology model, the location of the working component, and the location and type information of one or more upstream obstructing components are taken as input. Based on the input, an aerodynamic simplified model or empirical lookup algorithm is run, and a second adjustment coefficient is output. The model considers the physical properties of the obstructing components (size (full height / half height), thickness, surface roughness (affecting turbulence)), the superposition effect of obstruction (the cumulative effect of multiple upstream obstructing components), and distance and relative position (the distance between the obstructing component and the working component, whether it is directly opposite the center of the duct, etc.).

[0087] For example, the control parameters can be obtained by dividing the initial heat dissipation requirement by a second adjustment factor.

[0088] In this way, not only is the temperature of the component itself considered, but the airflow blockage effect caused by the power failure of the component on the same airflow channel is also quantitatively evaluated, so as to obtain a demand value that is more in line with the actual heat dissipation environment, avoid excessive heat dissipation due to unobstructed airflow channels, or insufficient heat dissipation due to blocked airflow channels, and achieve a balance between accuracy and safety.

[0089] In some specific embodiments, based on a preset air duct topology model, the physical location of the working component, and the physical location of the blocking component, a second adjustment coefficient for the heat dissipation efficiency of the working component by the blocking component is determined, including: Based on the preset duct topology model, the physical location of the working components and the physical location of the blocking components, determine the positional attenuation weight of each blocking component; Obtain the basic drag coefficient corresponding to the component type of each obstruction component; Based on the location attenuation weight and the basic drag coefficient, determine the single-point influence factor of each blocking component; By aggregating the influence factors of each blocking component, a comprehensive blocking factor is obtained; The second adjustment coefficient is determined based on the comprehensive blocking factor.

[0090] Here, one working component corresponds to one or more blocking components.

[0091] Here, the position attenuation weight is a coefficient for a single obstructing component, used to quantify the strength of its influence on downstream working components based on its specific location in the duct. It considers not only straight-line distance (e.g., 1 slot vs. 3 slots apart), but more importantly, the airflow path relationship in the preset duct topology model. For example, upstream positions directly facing the duct have high weights, while positions located in bypasses or airflow dead zones have low weights.

[0092] Here, the base drag coefficient is an inherent property value pre-defined based on component category and physical form. It describes the degree to which the physical structure of a component (such as a "full-height, full-length, double-width GPU (Graphics Processing Unit)," a "half-height network card," or a "2.5-inch hard drive") inherently obstructs airflow, regardless of its location. This coefficient is typically determined during the thermal design phase of electronic devices through wind tunnel testing or CFD (Computational Fluid Dynamics Simulation) simulation and is pre-defined in the BMC's knowledge base.

[0093] Here, the single-point impact factor is the final impact value calculated for a single blocking component, taking into account its inherent wind resistance properties and the weight of its current location.

[0094] Here, the overall blockage factor is a single total value that aggregates the effects of all relevant blockage components (i.e., all power-off components located upstream of the working component on the same duct).

[0095] Specifically, based on the "preset air duct topology model," the spatial relationship between each obstructing component and the target working component is analyzed. A weighting function can be used, whose inputs include: airflow path distance, whether it is in the same direct air duct, and whether there are corners or obstructions in between. The output is a weight value between 0 and 1 (for example, the weight is 1.0 for being immediately upstream, and 0.3 for being two slots apart and having a deflection in the air duct).

[0096] Specifically, the single-point impact factor can be obtained by multiplying the basic drag coefficient by the location attenuation weight.

[0097] Specifically, the comprehensive impact factor can be obtained by aggregating single-point impact factors using a superposition model, a maximum value model, or a weighted average model.

[0098] Specifically, the second adjustment coefficient can be obtained by 1 - the comprehensive blocking factor.

[0099] In some specific implementations, the method further includes: In response to the detection of a restart, the identification information of each component is reread; Based on the preset global mapping table, the corresponding heat dissipation strategy for each component is determined.

[0100] In related technologies, if the BMC restarts but the OS does not after hot-swapping, the BMC may still use the old card's heat dissipation strategy to regulate the heat dissipation of the new card, which may lead to overheating or over-temperature issues.

[0101] Here, BMC restart can be performed when the electronic device is in S5 state or when the electronic device is in S0 state.

[0102] Here, the heat dissipation strategy can be either the first heat dissipation strategy or the second heat dissipation strategy.

[0103] Component information can be obtained through event triggering, active scanning, or direct reading.

[0104] Specifically, during each initialization, the BMC actively traverses all component slots and reads the actual PN (Partition Number) of each component via I2C / SMBus. The BMC then performs a mapping table lookup based on the read PN and directly executes the PN->VID / DID->dedicated policy thermal control path. This path does not depend on or wait for hardware information sent by the BIOS. While executing this independent control, the BMC compares the read PN with the corresponding slot PN reported by the BIOS during the current boot process. If they match, it indicates information synchronization. If they do not match, it indicates that the hardware configuration has changed but the BIOS information has not been updated (or the old BMC information is invalid). The BMC will record an audit log, noting the slot and PN differences, for maintenance personnel to verify, but the thermal control will still be based on the actually read PN.

[0105] In this way, the risk of policy backtracking caused by expired internal cache information of the controller is completely eliminated, ensuring that the heat dissipation strategy is strictly consistent with the actual physical hardware state under any abnormal conditions, which greatly improves the reliability and security of the system.

[0106] In some specific implementations, the method further includes: In response to a hot-plug event detected via a component interface or system bus, read the first identifier of the new component; Compare the first identifier with the global mapping table; In response to the presence of a first identifier in the global mapping table, a new component is identified, and a corresponding first heat dissipation strategy is determined based on the global mapping table. If the first identifier does not exist in the global mapping table, a preset safe heat dissipation strategy is enabled for the new component; In response to the acquisition of the second and third identifiers of the new component, a mapping relationship is established between the first identifier and the second and third identifiers, and stored in the global mapping table.

[0107] In related technologies, when using the same heat dissipation control for the old component after hot-swapping, if the power consumption and temperature resistance of the old component are lower than those of the new component, the old component's heat dissipation strategy may not meet the new component's heat dissipation requirements, leading to overheating issues. If the power consumption of the old component is higher than that of the new component, there is an overheating problem.

[0108] Here, a hot-plug event refers to the physical insertion or removal of a hot-pluggable hardware module (such as a hard drive, network card, power supply, memory, etc.) without shutting down the system power. The system detects this action through a detection mechanism and generates a corresponding event notification to initiate configuration, uninstallation, resource reclamation, or fault handling processes.

[0109] The hot-plug event can occur when the electronic device is in a low-power or powered-off state (e.g., S5), or when the electronic device is in a normal operating state (e.g., S0).

[0110] Hot-plug events are detected via hardware pins (such as PRSNT for PCIe).

[0111] Specifically, after detecting a hot-plug event via a hardware pin (such as PCIe's PRSNT), the BMC immediately reads the new component's PN and executes a process to identify its identity via the PN. If the component is known, the policy switch can be completed quickly (typically within milliseconds); if the component is unknown, a security policy is immediately enabled, and the system waits for the next system reboot. When the BIOS can recognize the new component and send it to the BMC, the BMC updates its local mapping table and persists it. Subsequently, the BMC can identify the new component and match it with dedicated, precise thermal management based on the updated local mapping table.

[0112] For example, a safe heat dissipation strategy could be to use the default conservative heat dissipation strategy of such components, such as network cards, to regulate the alarm temperature threshold of 70°C.

[0113] In one embodiment, the electronic device is assumed to be a server, taking a 2U server as an example.

[0114] Specifically, the initial hardware configuration includes: Slot 1 (PCIe Slot 1): Installs a high-performance GPU computing card, part number PN is GPU-Adv-100, power consumption is 300W, and it has a built-in temperature sensor.

[0115] Slot 2 (PCIe Slot 2): Installs a smart OCP network card, part number PN is NIC-OCP-25G, power consumption is 50W, supports remote management, and has a built-in temperature sensor.

[0116] Chassis environment: Equipped with an ambient temperature sensor, the computer room temperature is kept constant at 25°C.

[0117] Pre-installed knowledge: The BMC firmware has a pre-installed component mapping table, which includes the PN, VID / DID information of GPU-Adv-100 and NIC-OCP-25G and their exclusive heat dissipation strategies (such as: GPU alarm threshold of 85°C, network card alarm threshold of 70°C, and corresponding optimized fan curves).

[0118] Specifically, the first phase: startup learning and benchmark establishment, includes: The server powers on and enters the S0 state. During the initial startup phase, the BMC actively scans the PCIe bus and other management interfaces. When it reads the PN of Slot 1 as GPU-Adv-100, it identifies it as a high-power GPU with a temperature sensor. When it reads the PN of Slot 2 as NIC-OCP-25G, it identifies it as a smart network interface card with a temperature sensor.

[0119] Snapshot generation: BMC integrates the scan results and generates a runtime configuration snapshot with the following content: {Slot1: Component PN=GPU-Adv-100, Type=GPU, Sensor present, Status=Operating; Slot2: Component PN=NIC-OCP-25G, Type=NIC, Sensor present, Status=Operating}.

[0120] Persistent storage: BMC writes this snapshot to its onboard Flash chip (non-volatile memory).

[0121] Initial cooling: Based on the configuration of the two high-power components, BMC loads the corresponding high-load cooling strategy, and the system fans run at a higher speed (e.g., a uniform speed set to 50% duty cycle) to ensure that the GPU and network card do not overheat under high load.

[0122] Specifically, the second phase: state synchronization and S5 intelligent weighted control, includes: State transition: The server task is completed, and it enters the ACPI S5 soft shutdown state. Major components such as the CPU and memory are powered off, but the BMC, chassis environmental sensors, and smart network cards remain powered on.

[0123] State Synchronization: The BMC immediately triggers a hardware presence check and finds: Slot 1: The GPU card is physically present but powered off (not a valid heat source); Slot 2: The network card is physically present and in operation (valid heat source). The BMC compares this detection result with the stored "Runtime Configuration Snapshot" and finds that the configuration has not changed physically, but the component's operating state has changed. Based on this, the BMC generates an updated "Current Valid Configuration": {Valid Heat Source: [Slot2-NIC-OCP-25G]; Airflow Obstruction: [Slot1-GPU-Adv-100]; Ambient Temperature: 25°C}. The BMC then permanently stores this "Current Valid Configuration" to the Flash chip, overwriting the previous complete snapshot to ensure that the memory is the latest state.

[0124] Hybrid Decision Making and Dynamic Weighting: The network card (with sensors) is currently the only heat dissipation target, and the second weighting algorithm is applied. Initial Demand Acquisition: The BMC reads the current network card temperature as 60°C and calculates an initial heat dissipation demand based on its proprietary strategy (threshold 70°C), assuming this demand corresponds to a fan speed of 30% when the system is powered on. Environmental Weighting: Based on the "current effective configuration," the BMC queries the pre-set chassis thermal topology model and finds that the powered-off GPU card (large in size) is located upstream of the network card, significantly obstructing airflow. The calculation shows its attenuation coefficient β = 0.7 on the network card's heat dissipation efficiency. Dynamic Weighting Calculation: Applying the weighting formula, the final heat dissipation demand = initial demand × β = 30% × 0.7 = 21%.

[0125] Execution result: The BMC controls the corresponding system fan area to operate at approximately 21% of its maximum speed. Compared to the 30% speed at startup, this represents a significant "downweighting" adjustment. This speed ensures the network card's heat dissipation safety when airflow is obstructed, while avoiding the energy waste associated with a fixed high-speed mode (such as the traditional 40%). If the GPU slot is empty (unobstructed, β≈0.5), the speed can be further reduced to approximately 15%, resulting in even better energy efficiency.

[0126] Specifically, the third stage: hot-plugging and system self-healing in S5 state includes: when the server is in S5 state, the maintenance personnel hot-plug the NIC-OCP-25G network card in Slot2 and insert a brand new, unverified high-speed FPGA acceleration card; the BMC immediately detects the hot-plugging event through the PRSNT signal of the PCIe slot; attempts to read the PN of the new component, but identification fails because its model is not recorded in the preset mapping table; according to preset rules, the BMC immediately marks the card as an "unknown component" and enables the preset "worst-case" safe heat dissipation strategy for Slot2 (for example, forcing the fan speed corresponding to the slot to a higher 40% to ensure that any high-power components will not overheat); the BMC immediately updates the "current effective configuration" to: {Slot2: Unknown component (safe mode);...}, and immediately writes this "snapshot" to the Flash chip for permanent storage.

[0127] Even if the BMC unexpectedly restarts for any reason, the latest configuration it reads from Flash after restarting will be "Slot 2 has an unknown component," instead of the old "Slot 2 has a network card." Therefore, the BMC will continue to employ a safe cooling strategy, avoiding the risk of the FPGA card overheating and being damaged due to incorrect loading of the old network card's low-power cooling strategy caused by information backtracking. The system achieves true self-correction and continuous security protection.

[0128] In one embodiment, assuming the electronic device is a server... Figure 4 This is a flowchart illustrating an embodiment of this application, such as... Figure 4 As shown, the process in this application includes: powering on the server and entering the power-on learning and database creation phase; determining whether the system has entered the S5 soft shutdown state; if not, returning to the power-on learning and database creation phase; if so, entering the S5 intelligent control main loop.

[0129] Specifically, boot learning and database building include: BMC actively scanning the entire system hardware; identifying component identities, locations, and sensor capabilities; generating runtime configuration snapshots; and storing the snapshots in non-volatile memory.

[0130] Specifically, the S5 intelligent control main loop includes: state perception synchronization; hybrid intelligent decision engine; execution of precise strategies after dynamic weight reduction; continuous monitoring of system status to determine whether a hot-plug event has been detected; if so, return a hot-plug response subprocess.

[0131] Specifically, the hot-swap response sub-process includes: attempting to read and identify the new component PN; determining whether identification was successful; if so, matching the dedicated policy; if not, enabling the safety fallback policy; updating the snapshot and storing it permanently.

[0132] It should be understood that, although Figures 1-4The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figures 1-4 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0133] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method.

[0134] Embodiments of this application also provide a heat dissipation control device, comprising: a first processing module 501, configured to read the identification information of each component within an electronic device, determine and execute a first heat dissipation strategy corresponding to each component according to a preset global mapping table; a second processing module 502, configured to monitor the operating status of the electronic device, and determine the presence and power supply status of each component in response to detecting that the electronic device has entered a low-power state; a third processing module 503, configured to determine the current effective configuration of the electronic device based on the presence and power supply status of each component, wherein the current effective configuration includes working components and power-off components; a fourth processing module 504, configured to determine the adjustment coefficient of each working component based on the current effective configuration, adjust the first heat dissipation strategy corresponding to each working component according to the adjustment coefficient, and determine a second heat dissipation strategy corresponding to each working component; and a fifth processing module 505, configured to perform heat dissipation control on each working component according to the second heat dissipation strategy and continuously monitor the status of each working component.

[0135] In a preferred embodiment of this application, the device further includes a construction module, which is specifically used to: obtain a first identifier, a second identifier, and a third identifier of each component; establish an association database with the first identifier as an index and associated with the second identifier and the third identifier; extract relevant mapping relationships from the association database according to the model of the electronic device, and generate a global mapping table corresponding to the model; and burn the firmware containing the global mapping table into the electronic device.

[0136] In a preferred embodiment of this application, the fourth processing module 504 is specifically configured to: determine whether each working component has a temperature sensor; in response to the absence of a temperature sensor in a working component, determine a first adjustment coefficient based on the current ambient temperature and the current effective configuration, adjust the first heat dissipation strategy corresponding to each working component according to the first adjustment coefficient, and determine a second heat dissipation strategy corresponding to each working component; in response to the presence of a temperature sensor in a working component, acquire the component temperature of the working component, determine a second adjustment coefficient in conjunction with the current effective configuration, adjust the first heat dissipation strategy corresponding to each working component according to the second adjustment coefficient, and determine a second heat dissipation strategy corresponding to each working component.

[0137] In a preferred embodiment of this application, the fourth processing module 504 is further configured to: determine the number of working components based on the current effective configuration; determine a first adjustment coefficient based on the number of working components and the total number of pre-stored components; determine the basic heat dissipation requirements based on the current ambient temperature; and adjust the first heat dissipation strategy corresponding to each working component based on the first adjustment coefficient and the basic heat dissipation requirements to determine the second heat dissipation strategy corresponding to each working component.

[0138] In a preferred embodiment of this application, the fourth processing module 504 is further configured to: obtain a preset chassis air duct topology model; determine the physical location of the working component and the physical location of the power-off component based on the current effective configuration; and determine the blocking component corresponding to the working component based on the preset chassis air duct topology model, the physical location of the working component and the physical location of the power-off component, wherein the blocking component is at least one power-off component located in the same air duct as the working component.

[0139] In a preferred embodiment of this application, the fourth processing module 504 is further configured to: determine the initial heat dissipation requirements of the working component based on the component temperature of the working component through a global mapping table; determine the second adjustment coefficient of the blocking component on the heat dissipation efficiency of the working component based on a preset chassis airflow topology model, the physical location of the working component, and the physical location of the blocking component; and adjust the first heat dissipation strategy corresponding to each working component based on the second adjustment coefficient and the initial heat dissipation requirements to determine the second heat dissipation strategy corresponding to each working component.

[0140] In a preferred embodiment of this application, the fourth processing module 504 is further configured to: determine the position attenuation weight of each blocking component based on the preset duct topology model, the physical position of the working component, and the physical position of the blocking component; obtain the basic drag coefficient corresponding to the component type to which each blocking component belongs; determine the single-point influence factor of each blocking component based on the position attenuation weight and the basic drag coefficient; aggregate the influence factors of each blocking component to obtain a comprehensive blocking factor; and determine the second adjustment coefficient based on the comprehensive blocking factor.

[0141] In a preferred embodiment of this application, the device further includes a sixth processing module, which is specifically used for: rereading the identification information of each component in response to detecting a restart; and determining the heat dissipation strategy corresponding to each component according to a preset global mapping table.

[0142] In a preferred embodiment of this application, the device further includes a seventh processing module, which is specifically configured to: read the first identifier of the new component in response to detecting a hot-plug event through a component interface or system bus; compare the first identifier with a global mapping table; identify the new component in response to the presence of the first identifier in the global mapping table, and determine a corresponding first heat dissipation strategy for the new component based on the global mapping table; enable a preset safe heat dissipation strategy for the new component in response to the absence of the first identifier in the global mapping table; and establish a mapping relationship between the first identifier and the second and third identifiers when the second and third identifiers of the new component are obtained, and store them in the global mapping table.

[0143] For a description of the features in the embodiment corresponding to the heat dissipation control device, please refer to the relevant description of the embodiment corresponding to the heat dissipation control method, which will not be repeated here.

[0144] Embodiments of this application also provide 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 perform the steps in any of the above embodiments of the heat dissipation control method.

[0145] The electronic device can be the target device, and its internal structure diagram can be as follows: Figure 6 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and the database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores thermal control data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a thermal control method.

[0146] Embodiments of this application also provide a computer-readable storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above-described embodiments of the heat dissipation control method when it is run.

[0147] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0148] Embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above-described heat dissipation control method embodiments.

[0149] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in any of the above-described heat dissipation control method embodiments.

[0150] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can 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.

[0151] The above provides a detailed description of a heat dissipation control method, electronic device, storage medium, and computer program product provided by this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only intended to help understand the method and core ideas of this application. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A heat dissipation control method, characterized in that, The method includes: Read the identification information of each component in the electronic device, determine the first heat dissipation strategy corresponding to each component according to the preset global mapping table, and execute it; Monitor the operating status of the electronic device, and in response to detecting that the electronic device has entered a low-power state, determine the presence and power supply status of each component; Based on the in-situ status of each component and the power supply status, the current effective configuration of the electronic device is determined, wherein the current effective configuration includes working components and power-off components; Based on the current effective configuration, the adjustment coefficient of each working component is determined, and the first heat dissipation strategy corresponding to each working component is adjusted according to the adjustment coefficient, and the second heat dissipation strategy corresponding to each working component is determined. According to the second heat dissipation strategy, heat dissipation control is performed on each working component, and the status of each working component is continuously monitored.

2. The heat dissipation control method according to claim 1, characterized in that, The method further includes: Obtain the first, second, and third identifiers of each component; Establish a database of associations with the first identifier as an index, and the second and third identifiers associated with it; Based on the model of the electronic device, relevant mapping relationships are extracted from the association database to generate a global mapping table corresponding to the model; The firmware containing the global mapping table is burned into the electronic device.

3. The heat dissipation control method according to claim 1, characterized in that, The process of determining adjustment coefficients for each working component based on the current effective configuration, adjusting the first heat dissipation strategy corresponding to each working component according to the adjustment coefficients, and determining the second heat dissipation strategy corresponding to each working component includes: Determine whether each of the aforementioned working components has a temperature sensor; In response to the absence of a temperature sensor in the working component, a first adjustment coefficient is determined based on the current ambient temperature and the current effective configuration. Based on the first adjustment coefficient, the first heat dissipation strategy corresponding to each working component is adjusted, and a second heat dissipation strategy corresponding to each working component is determined. In response to the presence of a temperature sensor in the working component, the component temperature of the working component is acquired. Based on the current effective configuration, a second adjustment coefficient is determined. According to the second adjustment coefficient, the first heat dissipation strategy corresponding to each working component is adjusted, and a second heat dissipation strategy corresponding to each working component is determined.

4. The heat dissipation control method according to claim 3, characterized in that, In response to the absence of a temperature sensor in the working component, a first adjustment coefficient is determined based on the current ambient temperature and the current effective configuration. Based on the first adjustment coefficient, a first heat dissipation strategy corresponding to each working component is adjusted, and a second heat dissipation strategy corresponding to each working component is determined, including: The number of the working components is determined based on the currently effective configuration; The first adjustment coefficient is determined based on the number of working components and the total number of pre-stored components; Based on the current ambient temperature, determine the basic heat dissipation requirements; Based on the first adjustment coefficient and the basic heat dissipation requirements, the first heat dissipation strategy corresponding to each working component is adjusted to determine the second heat dissipation strategy corresponding to each working component.

5. The heat dissipation control method according to claim 3, characterized in that, The method further includes: Obtain the preset chassis airflow topology model; Based on the current effective configuration, determine the physical location of the working component and the physical location of the power-off component; Based on the preset chassis airflow topology model, the physical location of the working component and the physical location of the power-off component, the blocking component corresponding to the working component is determined, wherein the blocking component is at least one power-off component located in the same airflow as the working component.

6. The heat dissipation control method according to claim 5, characterized in that, In response to the presence of a temperature sensor in the working component, the component temperature of the working component is acquired. Combined with the current effective configuration, a second adjustment coefficient is determined. Based on the second adjustment coefficient, the first heat dissipation strategy corresponding to each working component is adjusted, and a second heat dissipation strategy corresponding to each working component is determined, including: Based on the component temperature of the working component, the initial heat dissipation requirements of the working component are determined through the global mapping table; Based on the preset chassis airflow topology model, the physical location of the working component, and the physical location of the blocking component, a second adjustment coefficient for the heat dissipation efficiency of the working component by the blocking component is determined; Based on the second adjustment coefficient and the initial heat dissipation requirement, the first heat dissipation strategy corresponding to each working component is adjusted to determine the second heat dissipation strategy corresponding to each working component.

7. The heat dissipation control method according to claim 6, characterized in that, The step of determining a second adjustment coefficient for the heat dissipation efficiency of the working component by the blocking component based on the preset air duct topology model, the physical location of the working component, and the physical location of the blocking component includes: Based on the preset duct topology model, the physical location of the working component, and the physical location of the blocking component, determine the positional attenuation weight of each blocking component; Obtain the basic drag coefficient corresponding to the component type of each blocking component; Based on the location attenuation weight and the basic drag coefficient, the single-point influence factor of each blocking component is determined; By aggregating the influence factors of each blocking component, a comprehensive blocking factor is obtained; The second adjustment coefficient is determined based on the comprehensive blocking factor.

8. The heat dissipation control method according to claim 1, characterized in that, The method further includes: In response to detecting a restart, the identification information of each component is reread; Based on the preset global mapping table, the heat dissipation strategy corresponding to each component is determined.

9. The heat dissipation control method according to claim 1, characterized in that, The method further includes: In response to a hot-plug event detected via a component interface or system bus, read the first identifier of the new component; Compare the first identifier with the global mapping table; In response to the presence of the first identifier in the global mapping table, the new component is identified, and a corresponding first heat dissipation strategy is determined for the new component based on the global mapping table; In response to the absence of the first identifier in the global mapping table, a preset safe heat dissipation strategy is enabled for the new component; In response to obtaining the second and third identifiers of the new component, a mapping relationship is established between the first identifier and the second and third identifiers, and stored in the global mapping table.

10. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the heat dissipation control method as described in any one of claims 1 to 9 when executing the computer program.

Images