Heat dissipation equipment management method and device, equipment and readable storage medium

Abstract

The invention provides a heat dissipation equipment management method, device and equipment and a readable storage medium, and the method comprises the steps: responding to an event that a co-processing system is started after a server is powered on, obtaining the power-off time information of the last power-off of the server and a heat dissipation strategy of the last power-off of the server, obtaining the current time information, and obtaining the current time information; obtaining a time difference parameter according to the current moment information and the power-off moment information; calling a preset attenuation formula, and obtaining a predicted heat dissipation parameter according to the time difference parameter and a heat dissipation strategy when the server is powered off last time; and according to the predicted heat dissipation parameters, controlling the heat dissipation equipment to execute heat dissipation. Through the technical scheme of the specification, the time difference parameter is calculated and the heat dissipation parameter is estimated to control the heat dissipation equipment in combination with the heat dissipation strategy of the last power-off and the current moment information at the initial stage of power-on of the server, so that the rotating speed of the fan is reduced, the noise and the energy consumption are reduced, and the stability and the efficiency of initial operation of the server are improved.

Description

TECHNICAL FIELD

[0001] The present specification relates to the technical field of communication, and in particular, to a heat dissipation device management method, device, equipment and readable storage medium. BACKGROUND

[0002] During the running process of the server, due to its high power consumption characteristics, the requirement for the heat dissipation system is extremely high. In order to ensure that the core components can work at a safe temperature, and avoid device failure or more serious board burning problems caused by overheating, the traditional heat dissipation mode mainly relies on the fan or liquid cooling + fan mode for control. Especially in the normal running state, the baseboard management controller (BMC) will adjust the pulse width modulation (PWM) of the fan according to the temperature sensor data of the inlet, outlet and core components monitored in real time, and combine the heat dissipation simulation results of each model to control the fan speed, so as to realize effective heat dissipation. However, in the power-on stage, the BMC has not been completely started and is not ready for heat dissipation management, at this time the complex programmable logic device (CPLD) usually takes over the heat dissipation control. Since the CPLD itself does not have temperature sensing monitoring and complex computing capability, it often adopts a more aggressive fan control strategy, such as setting a higher initial speed, to ensure that the heat dissipation demand in this stage can be met. Although this method can effectively prevent the equipment from overheating, it also brings significant noise problems and increases unnecessary energy consumption.

[0003] Although the BMC processor can be equipped with a coprocessor, which is responsible for specific transaction processing such as heat dissipation control, there are still several challenges in actual operation. First, for some components such as optical modules, intelligent network cards, etc., their temperature information cannot be obtained through the traditional I2C protocol, but need to rely on more complex MCTP Over PCIe, NCSI or MCTP Over I2C protocols, which brings additional burden to the real-time operating system (RTOS) on the coprocessor, because these protocols are not naturally supported by RTOS. Second, the hardware I2C topology, as a public resource, is not only used to obtain temperature sensing information, but also serves other information transmission tasks, which leads to the problem that there may be competitive access to I2C devices between the main system and the RTOS, affecting the stability and response speed of the system. Especially in the case of immediately re-powering after the server has completed the power-on and power-off operation, if some key components (such as GPU) are still in a high temperature state, but the coprocessor cannot timely and accurately obtain the relevant temperature sensing information, it may cause the risk of insufficient heat dissipation, thereby affecting the overall performance and reliability of the server. SUMMARY

[0004] Therefore, the present specification provides a heat dissipation device management method, device, equipment and readable storage medium to improve the problem that heat dissipation cannot be efficiently controlled in the above-mentioned specific scenarios.

[0005] The specific technical solution is as follows:

[0006] The present specification provides a heat dissipation device management method, applied to a coprocessing unit of a BMC device, the BMC device further comprising a main processing unit, a coprocessing system of a BMC system with faster startup speed than the main processing unit of the BMC device running on the coprocessing unit, the method comprising: in response to an event that the coprocessing system is started after the server is powered on, obtaining power-off time information of the last server power-off, a heat dissipation strategy at the time of the last server power-off, obtaining current time information, obtaining a time difference parameter according to the current time information and the power-off time information; calling a preset attenuation formula, obtaining an estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the time of the last server power-off; according to the estimated heat dissipation parameter, controlling the heat dissipation device to perform heat dissipation before the BMC system of the main processing unit of the BMC device is started.

[0007] As a technical solution, in response to the event that the coprocessing system is started after the server is powered on, the power-off time information of the last server power-off, the heat dissipation strategy at the time of the last server power-off, the current time information are obtained, and the time difference parameter is obtained according to the current time information and the power-off time information, comprising: obtaining the power-off time information of the last server power-off and the heat dissipation strategy at the time of the last server power-off in the shared cache, wherein the shared cache is a cache shared by the coprocessing unit and the main processing unit of the BMC device, and the power-off time information of the last server power-off and the heat dissipation strategy at the time of the last server power-off stored in the shared cache are updated by the main processing unit of the BMC device according to the running time and the heat dissipation strategy at the time of running; the current time information is obtained from the RTC clock and / or the NTP server.

[0008] As a technical solution, the preset attenuation formula is called, and the estimated heat dissipation parameter is obtained according to the time difference parameter and the heat dissipation strategy at the time of the last server power-off, comprising: calling the preset attenuation formula, and obtaining the estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the time of the last server power-off, wherein the heat dissipation strategy contains that the heat dissipation device load parameter is negatively correlated with the time difference parameter, and the amount of decrease of the heat dissipation device load parameter decreases with the increase of the time difference parameter.

[0009] As a technical solution, the step of controlling the heat dissipation device to perform heat dissipation before the BMC system of the main processing unit of the BMC device is started is completed according to the estimated heat dissipation parameter, and further comprises: after the BMC system of the main processing unit of the BMC device is started, current temperature data fed back by each temperature sensor of the server is obtained from the shared cache, and a heat dissipation strategy is configured according to the current temperature data fed back by each temperature sensor; the shared cache is a cache shared by the coprocessing unit and the main processing unit of the BMC device, and each current temperature data stored in the shared cache is obtained by the main processing unit of the BMC device after communication with each temperature sensor and written.

[0010] As a technical solution, the step of controlling the heat dissipation device to perform heat dissipation before the BMC system of the main processing unit of the BMC device is started is completed according to the estimated heat dissipation parameter, and further comprises: through the I2C channel, temperature data fed back by the key temperature sensor is periodically obtained, and the estimated heat dissipation parameter is periodically adjusted according to the data fed back by the key temperature sensor.

[0011] The present specification also provides a heat dissipation device management apparatus, which is applied to a coprocessing unit of a BMC device, the BMC device further comprising a main processing unit, and a coprocessing system running on the coprocessing unit, the startup speed of the coprocessing system being faster than that of the BMC system of the main processing unit of the BMC device, the apparatus comprising: a first module configured to, in response to an event that the coprocessing system is started after the server is powered on, obtain power-off time information of the last server power-off, a heat dissipation strategy at the time of the last server power-off, obtain current time information, and obtain a time difference parameter according to the current time information and the power-off time information; a second module configured to call a preset attenuation formula, obtain an estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the time of the last server power-off; and a third module configured to control the heat dissipation device to perform heat dissipation according to the estimated heat dissipation parameter before the BMC system of the main processing unit of the BMC device is started.

[0012] As a technical solution, the step of controlling the heat dissipation device to perform heat dissipation before the BMC system of the main processing unit of the BMC device is started is completed according to the estimated heat dissipation parameter, and further comprises: through the I2C channel, temperature data fed back by the key temperature sensor is periodically obtained, and the estimated heat dissipation parameter is periodically adjusted according to the data fed back by the key temperature sensor.

[0013] As a technical solution, the preset attenuation formula is called to obtain the estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the last server power-off, wherein the heat dissipation strategy contains that the heat dissipation device load parameter is negatively correlated with the time difference parameter, and the amount of decrease of the heat dissipation device load parameter decreases with the increase of the time difference parameter.

[0014] As a technical solution, the third module is further configured to obtain current temperature data fed back by each temperature sensor of the server from the shared cache after the BMC system of the BMC device main processing unit is started, and configure a heat dissipation strategy according to the current temperature data fed back by each temperature sensor.

[0015] As a technical solution, after the step of controlling the heat dissipation device to perform heat dissipation according to the estimated heat dissipation parameter before the BMC system of the BMC device main processing unit is started, the method further comprises the following steps: periodically obtaining temperature data fed back by a key temperature sensor through an I2C channel, and periodically adjusting the estimated heat dissipation parameter according to the data fed back by the key temperature sensor.

[0016] The present specification also provides an electronic device comprising a processor and a readable storage medium, wherein the readable storage medium stores machine executable instructions capable of being executed by the processor, and the processor executes the machine executable instructions to implement the heat dissipation device management method.

[0017] The present specification also provides a readable storage medium, wherein the readable storage medium stores machine executable instructions, and the machine executable instructions cause the processor to implement the heat dissipation device management method when the machine executable instructions are called and executed by the processor.

[0018] The above technical solutions provided by the present specification at least have the following beneficial effects:

[0019] By utilizing the feature of the co-processing unit that can be quickly started at the initial stage of the server power-on, combining the time difference parameter calculated by the heat dissipation strategy at the last power-off and the current time information, and using the preset attenuation formula to obtain the estimated heat dissipation parameter to control the heat dissipation device, the method ensures effective heat dissipation before the BMC main system is completely started, reduces the fan speed, reduces the noise and energy consumption, and improves the stability and efficiency of the initial operation of the server. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present specification or the prior art, the drawings needed to be used in the description of the embodiments of the present specification or the prior art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present specification, and other drawings can also be obtained by those skilled in the art according to these drawings.

[0021] Figure 1 is a flow chart of a heat dissipation device management method in an embodiment of the present specification;

[0022] Figure 2 is a flow chart in an embodiment of the present specification;

[0023] Figure 3 is a structural diagram of a heat dissipation device management apparatus in an embodiment of the present specification;

[0024] Figure 4 is a hardware structural diagram of an electronic device in an embodiment of the present specification.

[0025] Reference signs: first module 21, second module 22, third module 23. DETAILED DESCRIPTION

[0026] The terms used in the embodiments of the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present specification. The singular forms "a", "an" and "the" used in the present specification and claims are intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to any or all possible combinations of one or more associated listed items.

[0027] It should be understood that although the terms first, second, third, etc. can be used in the embodiments of the present specification to describe various information, these information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information, without departing from the scope of the present specification. In addition, depending on the context, the word "if" used can be interpreted as "when" or "in response to determining" or "in response to ascertaining".

[0028] Therefore, the present specification provides a heat dissipation device management method, apparatus, device and readable storage medium to improve the above technical problems.

[0029] The specific technical solutions are described as follows.

[0030] In one embodiment, the present specification provides a heat dissipation device management method applied to a co-processing unit of a BMC device, the BMC device further comprising a main processing unit, and the co-processing unit runs a co-processing system of a BMC system with faster startup speed than the BMC system of the main processing unit of the BMC device, and the method comprises: in response to an event that the co-processing system is started after the server is powered on, obtaining power-off time information of the last server power-off, a heat dissipation strategy at the time of the last server power-off, obtaining current time information, obtaining a time difference parameter according to the current time information and the power-off time information; calling a preset attenuation formula, obtaining an estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the time of the last server power-off; and according to the estimated heat dissipation parameter, controlling the heat dissipation device to perform heat dissipation before the BMC system of the main processing unit of the BMC device is started.

[0031] Specifically, as Figure 1 , comprising the following steps:

[0032] Step S11, in response to an event that the co-processing system is started after the server is powered on, obtaining power-off time information of the last server power-off, a heat dissipation strategy at the time of the last server power-off.

[0033] Early heat dissipation control is realized through the co-processing system running on the co-processing unit. The co-processing system has faster startup speed than the BMC system of the main processing unit, and can quickly take over the heat dissipation task after the server is powered on. When obtaining the power-off time information and the heat dissipation strategy, the co-processing system can read these information from historical data stored in the BMC device. These data usually include temperature information of each key temperature sensing point before the server is powered off and corresponding fan speed and other heat dissipation strategy parameters.

[0034] Step S12, obtaining current time information, and obtaining a time difference parameter according to the current time information and the power-off time information.

[0035] The co-processing system obtains the current time information through a built-in clock module or an external RTC (real-time clock), so as to calculate the time difference parameter between the present power-on and the last power-off.

[0036] Step S13, calling a preset attenuation formula, obtaining an estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the time of the last server power-off.

[0037] The preset attenuation formula is a key to predict the current required heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the last power-off. The formula considers the influence of the time difference on the heat dissipation requirement, for example, if the time difference is short, it means that the internal temperature of the server after the last power-off may still be high, and therefore a relatively high heat dissipation parameter is required; on the contrary, if the time difference is long, the temperature may have been reduced, and a lower heat dissipation parameter can be used. In this way, the coprocessor system can dynamically adjust the speed of the heat dissipation device, avoiding the problem of fixed or overly aggressive fan speed in the traditional CPLD control mode, and reducing noise and power consumption.

[0038] In step S14, according to the estimated heat dissipation parameter, the heat dissipation device is controlled to perform heat dissipation before the BMC system of the main processing unit of the BMC device is started.

[0039] In the process of controlling the heat dissipation device to perform heat dissipation, the coprocessor system adjusts the fan speed comprehensively in combination with the historical temperature sensing data, the time difference parameter and the estimated heat dissipation parameter calculated by the preset attenuation formula.

[0040] In one embodiment, in response to the event that the coprocessor system is started after the server is powered on, the power-off time information of the last server power-off, the heat dissipation strategy at the last server power-off, the current time information are obtained, the time difference parameter is obtained according to the current time information and the power-off time information, including: obtaining the power-off time information of the last server power-off and the heat dissipation strategy at the last server power-off in the shared cache, wherein the shared cache is a cache shared by the coprocessor unit and the main processing unit of the BMC device, and the power-off time information of the last server power-off and the heat dissipation strategy at the last server power-off stored in the shared cache are updated by the main processing unit of the BMC device at runtime according to the runtime and the heat dissipation strategy at that time; obtaining the current time information from the RTC clock and / or NTP server.

[0041] The coprocessor system obtains the current time information as soon as possible after starting, which helps to more accurately determine the time difference between the current power-on and the last power-off, so as to more accurately refer to the last heat dissipation strategy. Time synchronization can be realized by an external RTC clock with a battery, so that even after AC power-off, the RTC can still work, ensuring that correct time information can be obtained after power-on. If there is no RTC clock, the temperature sensing change at the same position before and after AC can be referred to to roughly determine the power-off time interval.

[0042] In an embodiment, the preset attenuation formula is called to obtain the estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the last server power-off, wherein the heat dissipation strategy comprises that the heat dissipation device load parameter is negatively correlated with the time difference parameter, and the amount of decrease of the heat dissipation device load parameter decreases with the increase of the time difference parameter.

[0043] In an embodiment, after the step of controlling the heat dissipation device to perform heat dissipation according to the estimated heat dissipation parameter before the BMC system of the BMC device main processing unit is started, the method further comprises: obtaining current temperature data of each temperature sensor of the server from the shared cache after the BMC system of the BMC device main processing unit is started, and configuring a heat dissipation strategy according to the current temperature data of each temperature sensor; the shared cache is a cache shared by the coprocessing unit and the main processing unit of the BMC device, and each current temperature data stored in the shared cache is obtained by the main processing unit of the BMC device after communicating with each temperature sensor and then written.

[0044] After the BMC system of the main processing unit is started, the coprocessing system no longer directly obtains temperature sensing information from the actual device, but reads the temperature sensing information provided by the main processing unit from the shared memory. This design ensures that the main processing unit can fully take over the temperature sensing monitoring and heat dissipation control tasks after being started, and the coprocessing system can focus on other specific transaction processing, such as real-time heat dissipation speed adjustment.

[0045] In an embodiment, after the step of controlling the heat dissipation device to perform heat dissipation according to the estimated heat dissipation parameter before the BMC system of the BMC device main processing unit is started, the method further comprises: periodically obtaining temperature data of a key temperature sensor fed back through an I2C channel, and periodically adjusting the estimated heat dissipation parameter according to the data fed back by the key temperature sensor.

[0046] The coprocessing system only monitors information of key temperature sensing points, such as inlet air temperature sensing, outlet air temperature sensing, CPU temperature sensing, and PCI_Zone temperature sensing information. This is because when the power is first turned on, the business has not yet started completely, and it is neither realistic nor necessary to monitor all temperature sensing points. The coprocessing system adjusts the fan speed comprehensively according to the current temperature information of these key temperature sensing points, in combination with historical temperature sensing data, the time difference parameter, and the estimated heat dissipation parameter calculated by the preset attenuation formula. For example, if the inlet air temperature is high and the CPU temperature is normal, the coprocessing system may appropriately increase the fan speed to increase air flow and reduce the inlet air temperature; on the contrary, if the CPU temperature is high, the coprocessing system will focus on increasing the heat dissipation near the CPU.

[0047] In one embodiment, when the server completes the power-on operation, the co-processor unit starts and enters the working state, at this time the BMC main processing unit has not yet prepared to take over the overall monitoring and control task of the system. In order to ensure that the heat dissipation system can normally operate during this period, the co-processor unit needs to obtain some key information. Specifically, first, the specific time information at the last server power-off and the heat dissipation strategy adopted at that time are obtained. These information is usually stored in a non-volatile memory so that it can be quickly read when the server is restarted. At the same time, the co-processor unit also obtains the current time information from the external RTC (real-time clock). Based on the two sets of time data, the time difference parameter between this power-on and the last power-off is calculated, which is used to judge the temperature change since the last shutdown, thereby providing a basis for formulating a suitable heat dissipation strategy.

[0048] According to the preset attenuation formula, combined with the time difference parameter obtained in the foregoing and the heat dissipation strategy at the last server power-off, the co-processor unit calculates a predicted heat dissipation parameter. The attenuation formula mentioned here is used to simulate the trend of temperature change over time, and considering that the temperature of different components decreases at different speeds after power-off, this formula needs to be adjusted according to the specific model and application scenario. For example, in some cases, if the server is restarted in a short time, some key components inside it such as CPU, GPU, etc. may still maintain a high temperature, at which time relatively aggressive heat dissipation measures need to be taken; on the contrary, if it is started again after a long time, a more conservative heat dissipation strategy can be used according to the actual situation to save energy and reduce noise. In this way, the fan speed can be optimized as much as possible without affecting the heat dissipation effect, achieving the purpose of energy saving and emission reduction.

[0049] After the estimated heat dissipation parameter is determined, the co-processor unit controls the operation of the heat dissipation equipment according to the parameter. In this process, the co-processor unit can also focus on the temperature sensing points of several core areas, including the inlet, outlet, CPU, and PCI_Zone positions. By real-time monitoring of these key positions and combining historical temperature sensing data, the co-processor unit can dynamically adjust the speed of the fan to ensure that the temperature of the entire system is within a safe range. At this stage, since the BMC main processing unit has not yet fully prepared, all temperature sensing collection and analysis is completed independently by the co-processor unit. As the BMC main processing unit gradually starts and enters the working state, the co-processor unit no longer directly obtains temperature sensing data from the sensor, but instead reads the latest temperature sensing information provided by the BMC main processing unit from the shared memory. This seamless switching mechanism ensures the continuity and stability of the heat dissipation control process, avoiding any interruption or delay that may occur due to the transfer of control.

[0050] In one embodiment, a server is under maintenance operation at night and then restarted immediately. In this case, the coprocessor unit starts working at the moment when the server is just powered on. First, the last power-off time and the corresponding cooling strategy stored in the non-volatile memory are read. Then, the exact time is obtained by querying the external RTC, and the interval between the two operations is calculated. Assuming that this time is very short, it means that many components inside the server such as GPU are still in a high temperature state. Therefore, the coprocessor unit calculates a new cooling parameter according to the pre-set algorithm, combined with the time difference and the previous cooling strategy. Based on this parameter, the coprocessor unit instructs the fan to operate at an appropriate speed, ensuring that the system can maintain a safe working temperature range even if the BMC main processing unit has not been fully started. As the BMC main processing unit gradually goes online, the coprocessor starts receiving the latest temperature sensing data from the shared memory, thereby achieving smooth transition control of the cooling system.

[0051] In one embodiment, there is a linux system running in the BMC main core and an RTOS system running in the coprocessor in the BMC chip, which is collectively referred to as a heterogeneous dual system. Since the RTOS kernel is smaller than the linux system, the task processing is single, and the startup speed is faster than the BMC main core.

[0052] When powered on, the monitoring efficiency and actual situation need to be considered, and all temperature sensing does not need to be monitored, only the temperature sensing information of sensitive components or positions such as outlet temperature sensing, inlet temperature sensing, CPU temperature sensing, PCI_Zone temperature sensing information, etc.

[0053] During operation, the current temperature sensing information (the current BMC sensor history curve is a general function) and the latest fan speed regulation strategy are recorded in the database. This information serves as a reference for the last cooling strategy of the RTOS perception at the next server restart.

[0054] After the RTOS starts, the current time information can be obtained as soon as possible to more accurately determine the time difference between the current AC and the last running time, which determines the credibility of the last cooling strategy. The time synchronization scheme can use an external RTC clock with a battery or an NTP server. In order to simplify, the embodiment uses an RTC with a button battery to provide a clock synchronization source. The battery ensures that the RTC is still working after the AC is powered off, and the correct clock information can be obtained after power-on.

[0055] When AC power is on, the coprocessor first completes the initialization work, and before the main processor is ready, it first obtains the pre-AC temperature and the last cooling strategy and the pre-AC absolute time from the configuration information, and then obtains the current time from the external clock. Combined with the two parts of information, the interval time before and after the AC and the overall temperature before the AC can be known, which is used for the coprocessor cooling reference.

[0056] The coprocessor only obtains the key temperature information, and comprehensively cools and regulates in combination with the historical temperature and the time difference. At the same time, it monitors whether the main processor is ready in real time, and when the main processor is ready, the coprocessor no longer obtains the temperature information from the actual device, but obtains the temperature information from the shared memory.

[0057] Under the premise of controllable overall temperature, a relatively conservative strategy is maintained, and when the component temperature rises, a relatively aggressive way is adopted, and a segmented cooling strategy is adopted as a whole.

[0058] Although the temperature sensing point information monitored in the above state is relatively small, the business operation amount is small when the power is just on, and the cooling satisfies the basic demand. But in order to take into account the power-on and power-off in a short time, the OCP, GPU temperature which is not completely monitored may still be at a high temperature, so the cooling parameter adjustment is mainly aimed at the cooling strategy of the last power-on state and the time interval of AC power-on and power-off.

[0059] In one embodiment, after the server AC power is reconnected, the coprocessor unit embedded in the BMC chip completes the initialization first by virtue of the lightweight characteristics of its real-time operating system. This process usually takes hundreds of milliseconds, which is much faster than the startup time of the main processor operating system. The first operation of the coprocessor system after startup is to establish a time reference system: access the real-time clock chip integrated with a backup battery on the motherboard through a special bus interface, and accurately obtain the current absolute time stamp. The time stamp is accurate to the second level and uses the international standard time format. At the same time, the system initiates an access request to the persistent storage area of the BMC, and extracts the encrypted data package recorded when the server was last normally shut down. The data package contains three dimensions of historical information: first, the final effective cooling strategy parameter set, including core parameters such as basic cooling coefficient, weight configuration of each key temperature sensing area, and temperature alarm threshold; second, the temperature snapshot value of the key temperature sensing area, including the actual temperature reading of the inlet, outlet, CPU core and PCIe area at the moment of power-off; and finally, the corrected server power-off accurate time record. The three types of data together form a holographic image of the historical cooling state.

[0060] The coprocessing system calculates the difference between the current time stamp and the historical power-off time stamp to generate an accurate time interval in seconds. This time interval directly represents the physical cooling duration during server power-off, and its value size presents a negative correlation with the inheritance of the system thermal state - a shorter time interval implies that the key electronic components may not have been fully cooled, especially the high-power components not included in real-time monitoring may still maintain a critical high temperature; while a longer time interval means that the system has reached a thermal equilibrium state, and a more conservative cooling strategy can be enabled. This time difference parameter, as the core variable of dynamic adjustment, is directly input into the preset cooling strategy decay model.

[0061] The system calls a preset exponential decay algorithm model in the firmware layer, and its mathematical expression embodies a dynamic adjustment mechanism based on time difference. This model takes the base coefficient in the historical cooling strategy as the starting point, introduces the minimum safety coefficient preset for the model as the lower limit of cooling, and combines the decay rate constant determined by thermodynamic simulation to build a regulation curve that converges dynamically with the power-off duration. During specific operation, the system substitutes the time difference parameter into the exponential function to calculate the cooling coefficient applicable at present. This coefficient value decreases monotonically with the increase of time difference, maintaining a value close to the aggressive characteristics of the historical strategy at the initial stage of power-off, and gradually approaching the conservative lower limit value as the power-off time prolongs. This design not only guarantees the safety margin of cooling in the short power-off and restart scenario, but also maximizes the reduction of energy consumption and noise after long-term power-off.

[0062] Every specific time interval, the coprocessing system collects preset temperature data of four key areas through a dedicated physical channel: it obtains the ambient temperature of the air inlet and outlet through the thermistor interface; it reads the CPU core temperature through the dedicated processor interface; it collects the thermal radiation data of the PCIe area through the isolated serial bus. These real-time data are combined with historical weight parameters to calculate the weighted comprehensive temperature index. The system adopts a segmented strategy to perform dynamic speed regulation: when the weighted temperature is below the preset safety threshold, it adopts a gentle speed regulation strategy based on the predicted cooling coefficient to convert the calculated value into a fan driving signal; when the weighted temperature is detected to exceed the safety threshold, it immediately switches to the enhanced cooling mode and applies additional gain based on the predicted coefficient.

[0063] The co-processing system continuously detects the status flag of the main processing unit, and triggers the temperature monitoring handover program when it identifies that the main operating system completes the kernel loading. During the handover, the co-processing system stops direct access to the physical sensor, and instead reads the full set of temperature sensing data maintained by the main system from the shared storage area. The main system uses its complete protocol stack capability to obtain complete temperature information including complex expansion devices through various industry standard management interfaces, and writes into the predetermined memory mapping area. The co-processing system then converts to the polling reading side of the area, and continues to maintain the heat dissipation control until the main system is fully ready. This design not only overcomes the defect of limited protocol support of the co-processing system, but also realizes the smooth upgrade of the monitoring data source.

[0064] In the emergency restart scenario, when the server is powered off for three minutes and then immediately powered on, the time difference parameter calculated by the system triggers the high inheritance strategy. Even if the real-time temperature sensing reading is in the safe range, the system still maintains the heat dissipation intensity close to the historical value, which effectively prevents the risk of unmonitored component thermal collapse. In the overnight restart scenario, the power-off interval of several hours prompts the system to enable the conservative strategy close to the lower limit value, at which time the fan only maintains the basic rotation speed, significantly improving the noise environment. The special scenario processing module also includes a fault emergency mechanism. When abnormal historical data or clock failure is detected, the system automatically switches to the preset safe strategy mode, and generates traceable alarm events in the system log, ensuring the safety of heat dissipation in extreme conditions.

[0065] The real-time clock chip guaranteed by the backup battery can maintain accurate timing for tens of hours, coping with long-term power-off scenarios such as logistics turnover; the dedicated bus channel allocated to the co-processing system avoids resource conflicts during the initialization process of the main system; the shared data area implemented by the dual-port memory provides a nanosecond-level delay transmission channel for temperature sensing data handover; the decay model calibration based on thermal simulation experiments ensures that the algorithm parameters match the actual thermodynamic characteristics. The whole system reduces the peak noise during the startup phase, reduces invalid fan power consumption, and prevents the risk of rapid restart in three value dimensions, significantly optimizing the user experience and energy efficiency performance of the server startup phase under the premise of ensuring device safety.

[0066] In one embodiment, as Figure 3The present specification also provides a heat dissipation device management apparatus applied to a co-processing unit of a BMC device, the BMC device further comprising a main processing unit, and the co-processing unit runs a co-processing system of a BMC system with faster starting speed than the main processing unit of the BMC device, and the apparatus comprises: a first module configured to, in response to an event that the co-processing system is started after the server is powered on, acquire power-off time information of the last server power-off, a heat dissipation strategy at the time of the last server power-off, acquire current time information, and acquire a time difference parameter according to the current time information and the power-off time information; a second module configured to call a preset attenuation formula, acquire an estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the time of the last server power-off; and a third module configured to control the heat dissipation device to perform heat dissipation before the BMC system of the main processing unit of the BMC device is started.

[0067] In an embodiment, the acquiring, in response to an event that the co-processing system is started after the server is powered on, the power-off time information of the last server power-off, the heat dissipation strategy at the time of the last server power-off, the current time information, and the time difference parameter according to the current time information and the power-off time information comprises: acquiring, in a shared cache, the power-off time information of the last server power-off and the heat dissipation strategy at the time of the last server power-off, wherein the shared cache is a cache shared by the co-processing unit and the main processing unit of the BMC device, and the power-off time information of the last server power-off and the heat dissipation strategy at the time of the last server power-off stored in the shared cache are updated by the main processing unit of the BMC device according to a running time and a heat dissipation strategy at the time of running; and acquiring the current time information from an RTC clock and / or an NTP server.

[0068] In an embodiment, the calling the preset attenuation formula, acquiring the estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the time of the last server power-off comprises: calling the preset attenuation formula, and acquiring the estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy at the time of the last server power-off, wherein the heat dissipation strategy comprises that a heat dissipation device load parameter is negatively correlated with the time difference parameter, and the heat dissipation device load parameter decreases at a decreasing rate as the time difference parameter increases.

[0069] In an embodiment, the third module is further configured to, after the BMC system of the main processing unit of the BMC device is started, acquire current temperature data of each temperature sensor of the server from the shared cache, and configure the heat dissipation strategy according to the current temperature data of each temperature sensor; the shared cache is a cache shared by the co-processing unit and the main processing unit of the BMC device, and each current temperature data stored in the shared cache is acquired by the main processing unit of the BMC device after communication with each temperature sensor and written into the shared cache.

[0070] In an embodiment, the step of controlling the heat dissipation device to perform heat dissipation according to the estimated heat dissipation parameter before the BMC system of the main processing unit of the BMC device is started up is completed, further comprises: periodically obtaining temperature data fed back by the key temperature sensor through the I2C channel, and periodically adjusting the estimated heat dissipation parameter according to the data fed back by the key temperature sensor.

[0071] The device embodiments are the same as or similar to the corresponding method embodiments, and are not described here again.

[0072] In an embodiment, the present specification provides an electronic device, comprising a processor and a readable storage medium, the readable storage medium stores machine executable instructions capable of being executed by the processor, and the processor executes the machine executable instructions to implement the foregoing heat dissipation device management method. From the hardware level, the hardware architecture diagram can be seen from Figure 4 .

[0073] In an embodiment, the present specification provides a readable storage medium, the readable storage medium stores machine executable instructions, and the machine executable instructions, when invoked and executed by a processor, cause the processor to implement the foregoing heat dissipation device management method.

[0074] Here, the readable storage medium can be any electronic, magnetic, optical or other physical storage apparatus, and can contain or store information such as executable instructions, data, etc. For example, the readable storage medium can be: RAM (Random Access Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as hard drive), solid state disk, any type of storage disk (such as optical disk, dvd, etc.), or similar storage medium, or a combination thereof.

[0075] The system, device, module or unit illustrated in the above embodiments can be specifically implemented by a computer chip or entity, or by a product with certain functions. A typical implementation device is a computer, and the specific form of the computer can be a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email transceiver device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0076] For the convenience of description, the above device is described as various units respectively described in functions. Of course, the functions of each unit can be implemented in the same or more software and / or hardware when implementing the present specification.

[0077] Those skilled in the art will appreciate that embodiments of the present description can be devised for a method, a system, or a computer program product. Accordingly, the present description can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present description can take the form of a computer program product on one or more computer readable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code thereon for use by or in connection with an instruction execution system. For the purposes of this description, a computer usable or computer readable storage medium can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

[0078] The present description is described in reference to the flow diagrams and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present description. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing system or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified by one or more blocks Figure 1 one or more functions specified by one or more blocks.

[0079] These computer program instructions can also be stored in a computer readable storage medium that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable storage medium produce an article of manufacture including instructions which implement the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified by one or more blocks Figure 1 one or more functions specified by one or more blocks.

[0080] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified by one or more blocks Figure 1 one or more functions specified by one or more blocks.

[0081] Those skilled in the art will appreciate that embodiments of the present specification can be devised for a method, a system, or a computer program product. Accordingly, the present specification can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present specification can take the form of a computer program product on one or more computer readable storage media (which can include both volatile and non-volatile storage media, and which can also include computer storage media having additional features such as rewritable, random-access, memory, read-only, etc.). Moreover, the present specification can take the form of a computer program product on one or more computer readable storage media implementing program instructions for execution on one or more data processing apparatuses (e.g., any suitable computing device or other data processing apparatuses). Accordingly, the present specification should not be limited to only the above described embodiments.

[0082] The above description is merely illustrative of the embodiments of the present specification and is not intended to limit the present specification. Alternatives and variations of the present specification can be carried out by those skilled in the art. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present specification should be included in the scope of the claims of the present specification.

Claims

1. A heat dissipation device management method, characterized in that: A co-processing unit applied to a BMC device, the BMC device further including a main processing unit, wherein a co-processing system of a BMC system having a startup speed faster than that of the main processing unit of the BMC device runs on the co-processing unit, the method comprising: In response to an event indicating that the co-processing system has completed startup after the server is powered on, obtaining power-off time information of the last server power-off and the heat dissipation strategy when the server was powered off, obtaining current time information, and obtaining a time difference parameter based on the current time information and the power-off time information; Calling a preset attenuation formula to obtain an estimated heat dissipation parameter based on the time difference parameter and the heat dissipation strategy when the server was last powered off; According to the estimated heat dissipation parameters, the heat dissipation device is controlled to perform heat dissipation before the BMC system of the main processing unit of the BMC device is completely started.

2. The method according to claim 1, characterized in that The method responds to the event that the co-processing system is started after the server is powered on, obtains the power-off time information of the last server power-off, the heat dissipation strategy when the server was powered off, obtains the current time information, and obtains the time difference parameter based on the current time information and the power-off time information, including: Obtaining, from a shared cache, information about the last server power-off and a cooling policy applied during the last server power-off. The shared cache is a cache to which the co-processing unit and the main processing unit of the BMC device have shared access rights. The information about the last server power-off and the cooling policy applied during the last server power-off stored in the shared cache are updated by the main processing unit of the BMC device during operation based on the current operating time and cooling policy. Get the current time from the RTC clock and / or NTP server.

3. The method according to claim 1, characterized in that The calling of a preset attenuation formula to obtain an estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy when the server was last powered off includes: A preset attenuation formula is called to obtain an estimated heat dissipation parameter based on the time difference parameter and the heat dissipation strategy when the server was last powered off, wherein the heat dissipation strategy includes a negative correlation between the heat dissipation device load parameter and the time difference parameter, and as the time difference parameter increases, the amount by which the heat dissipation device load parameter decreases decreases.

4. The method according to claim 1, wherein After the step of controlling the heat dissipation device to perform heat dissipation based on the estimated heat dissipation parameters and before the BMC system of the main processing unit of the BMC device is completely started, the method further includes: After the BMC system of the main processing unit of the BMC device is started, the current temperature data fed back by each temperature sensor of the server is obtained from the shared cache, and the cooling strategy is configured according to the current temperature data fed back by each temperature sensor; The shared cache is a cache to which the co-processing unit and the main processing unit of the BMC device share access rights. The current temperature data stored in the shared cache is acquired and written after the main processing unit of the BMC device communicates with each temperature sensor.

5. The method according to claim 1, wherein After the step of controlling the heat dissipation device to perform heat dissipation based on the estimated heat dissipation parameters and before the BMC system of the main processing unit of the BMC device is completely started, the method further includes: The temperature data fed back by the key temperature sensor is periodically obtained through the I2C channel, and the heat dissipation parameters are periodically adjusted and estimated based on the data fed back by the key temperature sensor.

6. A heat dissipation equipment management device, characterized in that: A co-processing unit applied to a BMC device, the BMC device also including a main processing unit, the co-processing unit running a co-processing system of a BMC system having a startup speed faster than that of the main processing unit of the BMC device, the device comprising: The first module is configured to, in response to an event indicating that the co-processing system has completed startup after the server is powered on, obtain power-off time information of the last server power-off, a heat dissipation strategy used when the server was powered off, obtain current time information, and obtain a time difference parameter based on the current time information and the power-off time information; The second module is used to call a preset attenuation formula to obtain an estimated heat dissipation parameter based on the time difference parameter and the heat dissipation strategy when the server was last powered off; The third module is configured to control the heat dissipation device to perform heat dissipation according to the estimated heat dissipation parameters before the BMC system of the main processing unit of the BMC device is started.

7. The device according to claim 6, characterized in that The method responds to the event that the co-processing system is started after the server is powered on, obtains the power-off time information of the last server power-off, the heat dissipation strategy when the server was powered off, obtains the current time information, and obtains the time difference parameter based on the current time information and the power-off time information, including: Obtaining, from a shared cache, information about the last server power-off and a cooling policy applied during the last server power-off. The shared cache is a cache to which the co-processing unit and the main processing unit of the BMC device have shared access rights. The information about the last server power-off and the cooling policy applied during the last server power-off stored in the shared cache are updated by the main processing unit of the BMC device during operation based on the current operating time and cooling policy. Get the current time from the RTC clock and / or NTP server.

8. The device according to claim 6, characterized in that The calling of a preset attenuation formula to obtain an estimated heat dissipation parameter according to the time difference parameter and the heat dissipation strategy when the server was last powered off includes: A preset attenuation formula is called to obtain an estimated heat dissipation parameter based on the time difference parameter and the heat dissipation strategy when the server was last powered off, wherein the heat dissipation strategy includes a negative correlation between the heat dissipation device load parameter and the time difference parameter, and as the time difference parameter increases, the amount by which the heat dissipation device load parameter decreases decreases.

9. The device according to claim 6, characterized in that The third module is also used to obtain the current temperature data fed back by each temperature sensor of the server from the shared cache after the BMC system of the main processing unit of the BMC device is started, and configure the heat dissipation strategy according to the current temperature data fed back by each temperature sensor; The shared cache is a cache to which the co-processing unit and the main processing unit of the BMC device share access rights. The current temperature data stored in the shared cache is acquired and written after the main processing unit of the BMC device communicates with each temperature sensor.

10. The device according to claim 6, characterized in that After the step of controlling the heat dissipation device to perform heat dissipation based on the estimated heat dissipation parameters and before the BMC system of the main processing unit of the BMC device is completely started, the method further includes: The temperature data fed back by the key temperature sensor is periodically obtained through the I2C channel, and the heat dissipation parameters are periodically adjusted and estimated based on the data fed back by the key temperature sensor.

11. An electronic device, characterized in that: include: A processor and a readable storage medium, wherein the readable storage medium stores machine-executable instructions that can be executed by the processor, and the processor executes the machine-executable instructions to implement the method according to any one of claims 1 to 5.

12. A readable storage medium, characterized in that: The readable storage medium stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions prompt the processor to implement the method according to any one of claims 1 to 5.

Images