Server and power consumption management method thereof, electronic equipment, medium and program product

Abstract

The invention discloses a server and a power consumption management method thereof, electronic equipment, a medium and a program product, and relates to the technical field of servers. The method comprises the following steps: determining a current power consumption configuration mode according to state information of each power supply unit collected in a server operation process and preset power supply configuration parameter information, if the current power consumption configuration mode is a first power consumption configuration mode, determining a target power consumption capping value based on a working current safety range and rated power consumption of each power supply unit, updating a power consumption limit value strategy according to the target power consumption capping value; and if the mode is the second power consumption configuration mode, determining a transition power consumption capping value according to the rated power consumption of each power supply unit for normal power supply and a preset power consumption safety adjustment factor, and updating a power consumption limiting value strategy according to the transition power consumption capping value. The method can solve the problem that the related technology cannot meet the requirements that the power supply does not fail and the power supply can exert the maximum performance at the same time, and can dynamically adjust the power consumption limit value of the server according to the power supply state.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of servers, and in particular to a server, a power consumption management method thereof, an electronic device, a medium and a program product. BACKGROUND

[0002] High availability of a server depends on normal work of a power supply unit such as a PSU (Power Supply Unit), and in related technologies, in order to avoid power failure, a fixed power consumption limit value is set according to a PSU state, and it is impossible to simultaneously satisfy that no power failure occurs and the power supply can exert maximum performance. SUMMARY

[0003] The present application provides a server power consumption and a management method thereof, an electronic device, a nonvolatile storage medium, and a computer program product, which can dynamically adjust a server power consumption limit value according to a power supply state, can not only effectively avoid server unexpected shutdown caused by power failure, but also can exert maximum power supply performance.

[0004] To solve the above technical problems, the present application provides the following technical solutions. The present application provides a server power consumption management method applied to a management controller, including: obtaining power supply configuration parameter information and a working current safety range of a power supply unit; switching a power consumption configuration mode according to the power supply configuration parameter information and state information of each power supply unit in a running process; in a first power consumption configuration mode, determining a target power consumption cap value according to rated power consumption of each power supply unit based on that an overcurrent early warning protection corresponding to the working current safety range is triggered and an overcurrent protection corresponding to the working current safety range is not triggered, and updating a power consumption limit value strategy according to the target power consumption cap value; in a second power consumption configuration mode, determining a transition power consumption cap value according to rated power consumption of each normally powered power supply unit and a preset power consumption safety adjustment factor, and updating the power consumption limit value strategy according to the transition power consumption cap value, the preset power consumption safety adjustment factor being less than 1.

[0005] The present application provides a server power consumption management method applied to a logic device, including: When receiving the power supply configuration parameter information, polling each power supply unit according to the power supply configuration parameter information to obtain each power supply unit state information; if the power supply units in the in-place state are less than or equal to the required power supply unit quantity, and at least one power supply unit state information includes an alarm signal, then sending notification information indicating that the power consumption configuration mode corresponding to the power supply state is the second power consumption configuration mode, to determine a transition power consumption cap value according to the rated power consumption of each normally powered power supply unit and a preset power consumption safety adjustment factor, and update the power consumption limit value strategy according to the transition power consumption cap value, and the preset power consumption safety adjustment factor is less than 1; wherein the power consumption configuration mode corresponds to the first power consumption configuration mode, based on the overcurrent early warning protection corresponding to the trigger working current safety range and does not trigger the overcurrent protection corresponding to the trigger working current safety range, determines the target power consumption cap value according to the rated power consumption of each power supply unit, and updates the power consumption limit value strategy according to the target power consumption cap value. Wherein, if each power supply unit state information matches the power supply configuration parameter information, based on the overcurrent early warning protection corresponding to the trigger working current safety range and does not trigger the overcurrent protection corresponding to the trigger working current safety range, determines the target power consumption cap value according to the rated power consumption of each power supply unit, and updates the power consumption limit value strategy according to the target power consumption cap value.

[0006] The application further provides an electronic device comprising a memory and a processor, wherein the processor is configured to execute a computer program stored in the memory to implement the steps of any of the above-mentioned server power consumption management methods.

[0007] The application further provides a non-volatile storage medium, wherein the non-volatile storage medium stores a computer program, and the computer program is configured to be executed by a processor to implement the steps of any of the above-mentioned server power consumption management methods.

[0008] The application further provides a computer program product, comprising a computer program / instruction, which is configured to be executed by a processor to implement the steps of any of the above-mentioned server power consumption management methods.

[0009] Finally, the application further provides a server comprising a logic device, a management controller and a processor; the logic device is connected to the management controller through a first bus, and is configured to execute a computer program to implement the steps of any of the above-mentioned server power consumption management methods; the logic device comprises at least a power supply state monitoring interface and a frequency reduction interaction interface, the power supply state monitoring interface is connected to each power supply unit, the frequency reduction interaction interface is connected to the processor, and the frequency reduction interaction interface is used to send a frequency reduction throttling trigger signal to the processor; the management controller is configured to execute a computer program to implement the steps of any of the above-mentioned server power consumption management methods; the management controller comprises a power consumption execution manager, the power consumption execution manager is connected to the processor, and the power consumption execution manager is configured to update the power consumption limit value strategy according to the target power consumption cap value or the transition power consumption cap value, and send the power consumption limit value strategy to the processor.

[0010] The server power consumption management method provided by the application has the advantages that the number of power supply units is configured to support dynamic configuration of the power consumption ceiling value and ensure the stability of server power supply; the power supply unit in the current state and the power supply unit state are compared with the power supply parameter configuration information to determine whether the server is in a normal state or an abnormal state, and then the power consumption configuration mode to be switched is determined; the power consumption limit value matched with the current power supply state is adjusted according to the power supply state, the server power consumption limit value is dynamically adjusted according to the power supply state, the server unexpected shutdown caused by power supply failure can be effectively avoided, the business continuity is ensured, the hot replacement of the power supply unit does not affect the business continuity, the high availability of the server is improved, the power supply performance is maximized, the server power consumption is reduced, the resources are maximized, and the edge scene deployed in the limited power supply environment is beneficial.

[0011] In addition, the application also provides an electronic device, a nonvolatile storage medium, a computer program product and a server for the server power consumption management method, which further makes the method more practical, and the electronic device, the nonvolatile storage medium, the computer program product and the server have corresponding advantages. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions of the application or the related art, the following will briefly introduce the drawings needed to be used in the embodiments or the related art description. Obviously, the drawings in the following description are only some embodiments of the application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0013] Figure 1 The hardware composition framework diagram applicable to the server power consumption management method provided by the application is shown in the figure. Figure 2 The flowchart of the server power consumption management method provided by the application is shown in the figure. Figure 3 The execution flowchart of the management controller in the power consumption management process provided by the application is shown in the figure. Figure 4 The power callback flowchart provided by the application is shown in the figure. Figure 5 The flowchart of another server power consumption management method provided by the application is shown in the figure. Figure 6 The structural framework diagram of the server power consumption management device provided by the application in an exemplary embodiment is shown in the figure. Figure 7 The structural framework diagram of the server power consumption management device provided by the application in another exemplary embodiment is shown in the figure. Figure 8A structural diagram of an exemplary embodiment of an electronic device provided by the present application is shown in FIG. 1. Figure 9 A structural framework diagram of a server provided by the present application in an exemplary embodiment is shown in FIG. 2. Figure 10 An internal device data interaction flowchart of a server provided by the present application is shown in FIG. 3. DETAILED DESCRIPTION

[0014] In order to enable those skilled in the art to better understand the technical solutions of the present application, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. In the specification and the above-mentioned drawings, the terms "first", "second", "third", "fourth", etc. are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. The term "exemplary" means "as an example, embodiment or illustration". Any embodiment described as "exemplary" herein is not necessarily interpreted as superior or better than other embodiments.

[0015] With the rapid development of cloud computing, artificial intelligence, big data analysis and high-performance computing, the scale and computing density of data centers are also increasing, which are applied in real-time financial transaction platforms, global e-commerce platforms, various enterprise and government information systems, life science research fields, autonomous driving fields and various business scenarios.

[0016] In order to meet the needs of various business scenarios, servers used to carry business need continuous availability and long-term stable operation. Continuous availability means providing almost "never interrupted" service for business applications. Server downtime, even if it is a short minute-level interruption, can cause serious consequences, such as: financial transaction platform missing instructions leading to user loss, online service large-scale paralysis affecting reputation and user trust, industrial control process interruption causing safety accidents or production loss, medical system downtime endangering patient safety. Long-term stable operation means that the server needs to run continuously for 7x24 hours, and the life cycle is as long as several years. This not only requires the server to work normally at startup, but also requires it to maintain high stability, performance consistency and reliability during work when experiencing component aging, environmental temperature fluctuations, load periodic changes, etc. Any performance degradation or unplanned downtime of the server will significantly increase the operation and maintenance cost and affect the service quality agreement.

[0017] The PSU is a server power supply component that converts AC mains power into DC power required by various computing components inside the server efficiently, stably and safely, and provides accurate voltage rails (such as 12V, 3V, etc.). The power electronic components (such as capacitors, magnetic components), control circuits and heat dissipation systems inside the PSU will have risks of capacitor aging, power device failure, fan stall and other failures during long-term operation under abnormal working conditions such as high temperature, high load and power grid fluctuations. PSU failure will affect the business continuity and long-term stable operation of the server. Therefore, the high availability of the server depends on the normal operation of the power supply. In order to avoid server downtime caused by power supply failure, related technologies set a fixed power consumption limit (i.e. power consumption cap) according to the PSU state to prevent actual power consumption from being too high to cause power supply failure. However, in order to ensure that the server power supply does not fail due to high power consumption in any scenario, the power consumption limit is usually conservative, which prevents the power supply from achieving its maximum performance. In addition, in some scenarios such as power supply aging, instantaneous power peak, phase imbalance, etc., the actual power consumption of the server does not reach the specified power consumption cap, which can also cause the server to power off. It can be seen that related technologies cannot simultaneously satisfy the conditions of no power supply failure and maximum power supply performance.

[0018] In view of this, the present application supports configuring the number of power supply units and the number of redundant power supply units required to maintain the normal running state of the server, that is, the total number of power supply units and the minimum number of power supply units required for normal operation can be configured. The management controller provides two working mode switching: stable state and abnormal state. According to the obtained power supply unit state information combined with the above power supply configuration parameter information, if the current normal power supply state meets the power supply state provided by the power supply configuration parameter information, it works in the stable state. At this time, the setting process of the power consumption limit is: based on triggering the overcurrent early warning protection corresponding to the working current safety range and not triggering the overcurrent protection corresponding to the working current safety range, determining the target power consumption cap according to the rated power consumption of each power supply unit. If it is not met, the power consumption cap is re-determined according to the rated power consumption of each normal power supply unit and the preset power consumption safety adjustment factor. The server power consumption limit can be dynamically adjusted according to the power supply state, which not only effectively avoids server downtime caused by power supply failure, but also maximizes the performance of the power supply.

[0019] In combination with the specific application environment architecture or specific hardware architecture on which the server power consumption management method depends, the specific application environment architecture or specific hardware architecture is described here. The following describes the specific application environment architecture or specific hardware architecture in combination with Figure 1 Some possible application scenarios related to the technical solutions of the present application are exemplarily introduced, which can include the following content: For each server of a cloud data center, or a server of the data center for carrying high-service-continuity-required services (such as financial transaction services), or each server of an artificial intelligence distributed training platform, or a server of an edge computing platform, a hardware architecture for performing power consumption management in the server is as follows: A visual page of a BMC (Baseboard Management Controller) is provided with a power capping opening option of a dynamic cap value configuration function, and a user-oriented management interface is defined. The BMC also has a power consumption execution manager connected to a processor. The power consumption execution manager updates a power consumption limit policy according to a target power consumption cap value or a transition power consumption cap value and sends the power consumption limit policy to the central processor. The BMC on the mainboard is connected to a CPLD (Complex Programmable Logic Device) through an I2C (Inter-Integrated Circuit) bus. The BMC sends a power consumption limit policy, power supply configuration information and throttling interrupt signals to the CPLD through the bus, and receives a throttling processing notification signal of the CPLD. The CPLD includes a power supply state monitoring interface, a frequency reduction interaction interface and a direct connection frequency reduction interface. Each PSU connects at least signal lines corresponding to PS_PRESENT# (a bit signal), PSU_ALERT_N (an alarm signal), VIN Good (a power supply voltage signal) and PSU_PWROK (a power supply unit voltage signal) to the power supply state monitoring interface. The CPLD connects signal lines drawn through the frequency reduction interaction interface to a PRC HOT (Processor Hot) input pin and a MEM HOT (Memory Hot) input pin of the central processor. When a frequency reduction throttling signal of the CPLD is received, the central processor can be triggered to perform a frequency reduction throttling operation. In addition, the CPLD is connected to B30 of a PCIE (Peripheral Component Interconnect Express) card with a power brake function and A70 of an OCP (Open Compute Project) 3.0 network interface card through a reserved line. When a frequency reduction throttling signal of the CPLD is received, a frequency reduction throttling operation can be triggered.

[0020] In the server running process, the user manually opens the dynamic capping value configuration function through the power capping opening option of the visual page, and transmits the power configuration parameter information through the user management interface, wherein the power configuration parameter information includes the number of demand power supply units and the number of redundant power supply units required to maintain the normal running state of the server. The BMC obtains the power configuration parameter information and the working current safety range of the power supply unit. The BMC periodically polls the in-place state and state word of each PSU. If the state information of each power supply unit in the running process matches the power configuration parameter information, the overcurrent warning protection corresponding to the working current safety range is triggered, and the overcurrent protection corresponding to the working current safety range is not triggered, the target power capping value is determined according to the rated power consumption of each power supply unit, and the power consumption limit value strategy is updated according to the target power capping value; if the state information of each power supply unit in the running process does not match the power configuration parameter information, the transition power capping value is determined according to the rated power consumption of each normal power supply unit and the preset power consumption safety adjustment factor, and the power consumption limit value strategy is updated according to the transition power capping value, and the preset power consumption safety adjustment factor is less than 1. The CPLD receives the power configuration parameter information of the BMC, and polls each power supply unit in real time according to the power configuration parameter information to obtain the state information of each power supply unit; if the number of power supply units in the in-place state is less than or equal to the number of demand power supply units, and at least one power supply unit state information includes an alarm signal, a frequency reduction throttling trigger signal and a throttling processing notification signal are generated, if the operation frequency to be reduced is a central processing unit, the frequency reduction throttling trigger signal is directly sent to the PRC HOT input pin and the MEM HOT input pin of the central processing unit through the frequency reduction interaction interface, if it is a peripheral device, it is sent to the corresponding peripheral device through the direct frequency reduction interface, and the throttling processing notification signal is sent to the BMC, so that the BMC determines that the state information of each power supply unit does not match the power configuration parameter information. The BMC polls the mainboard CPLD through I2C, and records the throttle (frequency reduction throttling operation) triggered after the dynamic capping configuration function is opened.

[0021] Through the aforementioned three-level collaborative architecture (Hardware-level Real-time Monitoring (CPLD) → Intelligent Decision-Making (BMC) → Policy Execution Power Consumption Manager), the response time of this invention is shortened to 100ns, supporting real-time dynamic power cap adjustment and full hot maintenance. In contrast, independent BMC management requires a response time >800ms, offers limited dynamic adjustment, and does not support hot maintenance. For the hardware protection circuit implementation, the response time is 10μs, and it does not support dynamic adjustment of the power cap or hot maintenance. For the software-defined power supply implementation, the response time is greater than 1s, supporting dynamic adjustment of the power cap and partial hot maintenance. Therefore, this invention can meet the 99.999% availability and millisecond-level fault response requirements of cloud data centers, achieving zero downtime maintenance and ensuring business continuity. It can also meet the burst power management requirements of high-performance devices such as image processors and accelerator cards in AI distributed training systems, achieving overcurrent protection, collaborative power control of heterogeneous computing units, and meeting the power supply requirements of edge computing platforms in constrained environments, enabling dynamic derating.

[0022] It should be noted that the above application scenarios are only shown to facilitate understanding of the ideas and principles of the present invention, and the embodiments of the present invention are not limited in any way. On the contrary, the embodiments of the present invention can be applied to any applicable scenario. After introducing the technical solution of the present invention, various non-limiting embodiments of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Please see first. Figure 2 , Figure 2 This is a flowchart illustrating a server power management method provided in this embodiment. This embodiment applies to the server's management controller, which may be, for example, a BMC and may include the following: S201: Obtain power configuration parameter information and the safe operating current range of the power supply unit, and switch the power consumption configuration mode according to the power configuration parameter information and the status information of each power supply unit during operation.

[0024] The power supply configuration parameter information includes a parameter of the number of power supply units, and the user can configure the parameter through a power supply unit number configuration function. The power supply configuration parameter information may, for example, include a required power supply unit number and a redundant power supply unit number required to maintain a normal operation state of the server. The management controller can provide an external management interface, such as redfish (interface name), and the user can configure the power supply parameters of the server through the management interface. The required power supply unit number refers to the number of power supply units required for the normal operation of the server, and the redundant power supply unit number is equivalent to a backup power supply unit. The power supply unit is a device that provides power for the server, such as a PSU. Correspondingly, the power supply configuration parameter information includes a PSU number N required for the normal operation of the server and a redundant PSU number M. The working current safety range refers to the safety range of the working input current of the power supply unit and the current values corresponding to various protection mechanisms, such as OPP (Over Power Protection, over power protection), OCP (Over Current Protection, over current protection), and OCW (Over-Current Warning, over current warning), and the corresponding response time. The working current safety range is obtained by testing in a power supply unit without redundant power supply configuration scenario. The response time includes an alarm delay time and a fault delay under 12V latching. An exemplary working safety range is shown in Table 1.

[0025] Table 1 Working current safety range table of power supply unit

[0026] The embodiment is for supporting dynamic power consumption cap configuration function. The management controller polls the in-place state and status of each power supply unit at a fixed frequency cycle. Of course, the management controller can also be triggered to perform the operation of obtaining the status information of each power supply unit. The status information of each power supply unit at least includes the in-place state and whether there is an alarm state. The status information of each power supply unit can be obtained by obtaining the status word of each power supply unit. The status word can include over-temperature, over-current, over-voltage, and other alarm information. The status information of each power supply unit can reflect the current power supply state of the server. The power supply configuration parameter information is used to reflect the server power supply state or the supply state that can ensure the stable and continuous operation of the server. The management controller obtains the status information of each power supply unit each time, compares the current status information of each power supply unit with the power supply configuration parameter information, and if the current status information of each power supply unit meets the power supply configuration parameter information, that is, the two are matched, it indicates that the current state is normal, and the power consumption configuration mode corresponding to the current power supply state is the first power consumption configuration mode. According to S202, the power consumption cap value matched with the current power supply state is determined. For ease of description, it can be defined as the target power consumption cap value. If the current status information of each power supply unit does not meet the power supply configuration parameter information, that is, the two are not matched, the power supply of the server may fail, that is, it is in an abnormal state, and the power consumption configuration mode corresponding to the current power supply state is the second power consumption configuration mode. S203 can be used to determine the power consumption cap value matched with the current power supply state, which can be defined as the transition power consumption cap value. That is, the current power consumption configuration mode is determined to be the first power consumption configuration mode or the second power consumption configuration mode by comparing the power supply configuration parameter information and the status information of each power supply unit in the running process. If the current power consumption configuration mode is the first power consumption configuration mode, and the power consumption configuration mode corresponding to the current power supply state is the second power consumption configuration mode, the second power consumption configuration mode is switched, and S203 is used to determine the power consumption cap value matched with the current power supply state to configure the power consumption limit value strategy. If the current power consumption configuration mode is the first power consumption configuration mode, and the power consumption configuration mode corresponding to the current power supply state is the first power consumption configuration mode, no switching is needed, and the power consumption limit value strategy does not need to be reconfigured.

[0027] S202: In the first power consumption configuration mode, based on triggering the over-current pre-warning protection corresponding to the working current safety range and not triggering the over-current protection corresponding to the working current safety range, determining the target power consumption cap value according to the rated power consumption of each power supply unit, and updating the power consumption limit value strategy according to the target power consumption cap value.

[0028] S203: In the second power consumption configuration mode, determining the transition power consumption cap value according to the rated power consumption of each normal power supply unit and a preset power consumption safety adjustment factor, and updating the power consumption limit value strategy according to the transition power consumption cap value.

[0029] The overcurrent pre-warning protection corresponding to the working current safety range and the overcurrent protection threshold corresponding to the working current safety range are 120% in Table 1, that is, when the input current is 1.2 times the rated current, the overcurrent pre-warning protection is triggered, and the minimum value of the overcurrent protection is also triggered. In order to ensure that the overcurrent protection is not triggered, a smaller safety value can be set, and the product of the two is adjusted to the total rated power consumption that can be provided, that is, the current maximum power consumption limit. The preset power consumption safety adjustment factor is a number less than 1, such as 0.9, 0.85, 0.88, or 0.8. The preset power consumption safety adjustment factor is used to adjust the total rated power consumption of the current normal power supply unit as the maximum power consumption value, which can ensure safe and reliable operation in the case of possible failure. The power consumption limit value strategy is a strategy for setting the power consumption limit value. Whenever a new power consumption cap value is generated, the power consumption cap value of the power consumption limit value strategy is updated. For example, in Table 1, the PSU rated power consumption is 1000w. If the current does not exceed 140%, the power consumption corresponds to 1.4, and it can last for about 11 seconds. If it exceeds 150%, it can only last for 5ms before power failure, which has a risk of failure. If the PSU has redundancy, the maximum 1.2*0.97 is used to limit the power consumption, which will not trigger the OCW and can maximize the performance. If the PSU fails and loses redundancy, the 0.8*0.97 control is used to prioritize safety, and the total system power consumption cannot exceed the PSU multiplied by 1.4 to ensure reliable control.

[0030] In the technical scheme provided in the embodiment, by configuring the number of power supply units, not only can the power consumption cap value be dynamically configured, but also the stability of the server power supply can be ensured. By comparing the current in-place power supply unit and the power supply unit state with the power supply parameter configuration information, it is determined whether the server is in a normal state or an abnormal state, and then the power consumption configuration mode to be switched is determined. The power consumption limit value matched with the current power supply state is adjusted according to the power supply situation in different states, the power consumption limit value of the server is dynamically adjusted according to the power supply state, the server unexpected shutdown caused by power failure can be effectively avoided, the business continuity can be ensured, the hot replacement of the power supply unit does not affect the business continuity, the high availability of the server is improved, the power supply performance is maximized, the server power consumption is reduced, the resources are maximized, and it is beneficial to deploy in edge scenarios in limited power supply environment.

[0031] Before performing the process of dynamically adjusting the power consumption cap value in the above embodiment, the dynamic cap value configuration function can be set. The embodiment can add an on option and an off option of the dynamic cap value configuration function (powercapping) to the visual page of the management controller. The user can manually turn on the dynamic cap value configuration function, and of course the default option can also be set. The default option is in the off state. When the dynamic cap value configuration function is turned off, the logic device can determine whether the frequency throttling operation needs to be performed according to the minimum redundancy number. Figure 3As shown, when the dynamic ceiling value configuration function is started, the power consumption limit strategy is set. First, it is necessary to determine whether the dynamic ceiling value configuration function is supported, and the determination process is as follows: The number of power supply units in place and the rated power consumption are obtained; if the maximum system power consumption of the server is less than or equal to the power supply demand power consumption under the overload protection corresponding to the trigger working current safety range, the power supply configuration parameter information is sent to the logic device, and the dynamic power consumption ceiling value configuration function is started. If the number of power supply units in place does not meet the power supply configuration parameter information, a redundancy loss alarm is performed; if the maximum system power consumption of the server is greater than the power supply demand power consumption, a prompt information is generated that the dynamic power consumption ceiling value configuration is not supported.

[0032] The maximum system power consumption of the server is the maximum power consumption required by the server, which can be estimated by the server model information and the asset information sent by the BIOS. For example, the power consumption hardware such as CPU / GPU / memory / hard disk, expansion card and the corresponding configuration information such as the rated power consumption of each power consumption hardware can be determined according to the server asset information. The rated power consumption of the corresponding power consumption hardware is determined from the configuration information of each power consumption hardware; the server basic power consumption, i.e. the loss of the mainboard, system fan, various cables and PCB, is determined according to the server model information (i.e. the size specification, such as 1U or 4U) and the cooling layout parameters (fan number); the maximum system power consumption is determined according to the power consumption hardware and the corresponding rated power consumption, the server basic power consumption and the margin coefficient, for example, the maximum system power consumption=(power consumption hardware 1x rated power consumption 1+…power consumption hardware nx rated power consumption n+server basic power consumption)x margin coefficient, wherein the margin coefficient is used to cover the peak value in the peak value, for example, it can be taken as 1.1~1.2.

[0033] The determination process of the power supply demand power consumption can be: determining the total power consumption of the power supply units according to the number of demand power supply units and the rated power consumption of the power supply units; if the fault delay time of triggering the overload protection corresponding to the working current safety range is less than a first preset delay threshold, and the maximum fault delay time of triggering the overcurrent protection is greater than a second preset delay threshold, the second preset delay threshold being much greater than the first preset delay threshold, then determining the maximum current triggered by the overcurrent protection corresponding to the maximum fault delay time, and determining the power consumption critical adjustment factor according to the maximum current and the rated current; determining the power supply demand power consumption according to the total power consumption of the power supply units and the power consumption critical adjustment factor. As shown in Table 1, the response time of triggering the overload protection is 5 ms, which is too short and not easy to maintain, and the response time of triggering the first level overcurrent protection is 11.1 s. If the maximum power consumption state is just reached, the remaining PSU may trigger the overcurrent protection immediately due to excessive current, resulting in the failure of the logic device throttling. The alarm can inform the user of the risk in advance, so that the user can take precautions early. Therefore, the maximum current triggered by the overcurrent protection corresponding to the maximum fault delay time is 140%, and accordingly, if the maximum power consumption of the system is greater than 1.4 times the rated power consumption of N single power supply units, a log alarm is performed, for example, a visual page can display that the dynamic power capping cannot be supported (The current system configuration cannot be supported due to exceeding the system power cap), and points to 4b0000 to turn off the dynamic power capping. If the maximum power consumption of the current system is less than or equal to 1.4 times the rated power consumption of N single power supply units, it proves that the dynamic power consumption capping value configuration function is supported, and the power supply configuration parameter information is sent to the logic device. The logic device acquires the state of each power supply unit according to the received power supply configuration information. If the in-place power supply unit data is less than the number of demand power supply units, and there is still a power supply unit alarm, it indicates that the actual power consumption may exceed the power consumption capping value. The logic device will alarm and trigger the operation of reducing the operating frequency of the related power consumption device. The management controller will record the frequency reduction throttling operation in the log for subsequent power consumption management. After the logic device performs the frequency reduction throttling operation, the power consumption will be reduced, and the alarm can be removed, for example, the alarm pin of the power supply unit can be pulled high. The logic device will continue to perform the frequency reduction throttling operation until the management controller sends a throttling interrupt signal, thereby avoiding excessive loss of server performance.

[0034] The above embodiments do not limit how to determine the target power consumption value. Based on the above embodiments, the present application also provides an exemplary implementation, which can include the following content: Determine the power consumption providing information according to the power supply unit state information; determine the maximum pre-warning current triggering the over-current pre-warning according to the working current safety range, and determine the maximum power consumption adjustment factor according to the maximum pre-warning current and the rated current; determine the target power consumption ceiling value according to the power consumption providing information, the maximum power consumption adjustment factor and the first preset safety margin value.

[0035] The first preset safety margin value is to establish a safe, stable and reliable buffer zone between the hardware limit and the software control target, to cope with control delay, measurement error and transient peak, to ensure that the ceiling operation is smooth and effective, and will not trigger the protection mechanism of the hardware to cause system downtime. The value may be 0.97, 0.95 or 0.98, etc., which can be selected by a person skilled in the art according to the actual situation. The power consumption providing value refers to the total rated power consumption that the power supply unit can provide. For example, the target power supply unit in the normal working state among the power supply units in the in-place state can be determined according to the power supply unit state information. The power consumption providing value can be determined according to the rated power consumption value of each target power supply unit and the total number of target power supply units. Correspondingly, the target power consumption ceiling value = power consumption providing information × power consumption maximum adjustment factor × first preset safety margin. For example, the first preset safety margin is 0.97, and the target power consumption ceiling value = 1.2 × total rated power consumption that the power supply unit can provide × 0.97.

[0036] The above embodiments do not limit how to determine the transition power consumption ceiling value. Based on the above embodiments, the application also provides an exemplary implementation, which can include the following contents: For the scenario of supporting user-defined power consumption ceiling value configuration function through the user interface, the current power consumption providing value (i.e. new power consumption providing value) is determined according to the number of power supply units in the in-place state and the normal working state and the corresponding rated power. The preset power consumption safety adjustment factor is determined according to the working current safety range, and the power consumption ceiling adjustment value is determined according to the new power consumption providing value and the preset power consumption safety adjustment factor. When the user-defined power consumption adjustment request is received, if the user input power consumption ceiling custom value is less than or equal to the power consumption ceiling adjustment value, the power consumption ceiling custom value is taken as the new power consumption ceiling value; if the power consumption ceiling custom value is greater than the power consumption ceiling adjustment value, the power consumption ceiling adjustment value is taken as the transition power consumption ceiling value.

[0037] For the scenario of not supporting user-defined power consumption ceiling value configuration function or not receiving user-defined power consumption ceiling value, the new power consumption providing value can be determined according to the number of power supply units in the in-place state and the normal working state and the corresponding rated power. The power consumption safety adjustment factor is determined according to the working current safety range. The transition power consumption ceiling value is determined according to the new power consumption providing value, the preset power consumption safety adjustment factor and the second preset safety margin value.

[0038] The second preset safety margin value is to establish a safe, stable and reliable buffer between the hardware limit and the software control target, to cope with control delay, measurement error and transient peak, to ensure that the capping operation is smooth and effective, and will not trigger the protection mechanism of the hardware to cause system downtime. The value can be 0.97, 0.95 or 0.98, etc. Those skilled in the art can select according to the actual situation. The second preset safety margin value can be the same as or different from the first preset safety margin value, which does not affect the implementation of the present application. The new power consumption provided value is the sum of the current provided power consumption. For example, the power supply unit in the on state and normal working state is m (power supply unit 1, power supply unit 2, …, power supply unit m), and the new power consumption provided value = power supply unit 1 x rated power 1 + power supply unit 2 x rated power 2 + … + power supply unit m x rated power m. The power consumption capping adjustment value = power consumption safety adjustment factor x new power consumption provided value, and the transition power consumption capping value = new power consumption provided value x power consumption safety adjustment factor x second preset safety margin value. When the second preset safety margin is 0.97 and the power consumption safety adjustment factor is 0.8, the transition power consumption capping value = 0.8 x the sum of the rated power consumption provided by the power supply unit x 0.97.

[0039] As can be seen from the above, the embodiment supports the calculation of power consumption capping values in different scenarios, which not only dynamically adjusts the server power consumption limit according to the power supply state, but also maximizes the performance of the power supply.

[0040] Based on the above embodiment, when the logic device triggers the frequency reduction throttling operation, the embodiment also provides a power callback implementation method, as shown in Figure 4 The method can include the following contents: When the throttling processing notification signal is received, an alarm signal indicating that the power supply unit state information and the power supply configuration parameter information do not match is generated, and the periodic polling operation is stopped; at the same time, the state information of each power supply unit is continuously obtained until the power supply unit state information at the adjacent time satisfies the preset same similar condition.

[0041] The management controller acquires the power supply unit state information according to the preset state query frequency. When a throttling processing notification signal is received, the throttling processing notification signal indicates that the power consumption limit value strategy updated according to the target power consumption cap value takes effect, and at least one target power supply unit cannot normally supply power, an alarm operation is performed, and the current periodic polling of the power supply unit is stopped. The state of the power supply unit at the current time is immediately read multiple times. After the state of the power supply unit is stable, that is, the power supply unit state information at adjacent time meets the preset same similar condition, the new power consumption cap value, that is, the transition power consumption cap value, is recalculated by comprehensively stabilizing the acquired power supply unit state information. When the power consumption limit value strategy updated according to the transition power consumption cap value takes effect, a throttling interruption notification signal is sent. Of course, the interruption notification signal can also be sent after waiting for a few seconds, for example, 3 seconds, and the periodic polling operation is resumed. Whether the condition for removing the frequency reduction throttling operation is met is continuously monitored. If it is determined according to the latest power supply unit state information that the number of power supply units in a normal working state in the power supply units in place is greater than or equal to the required power supply unit number, an alarm removal signal is generated, the power consumption cap value of the server is restored to the target power consumption cap value, and the latest power consumption limit value strategy is updated according to the target power consumption cap value. For example, if the transition power consumption cap value is the total power consumption value provided by the current state normal PSU multiplied by 0.8 or a user preset value, when the periodic polling detects that the power supply unit has been replaced (that is, the number of current normal power supply units is greater than or equal to the power supply requirement number), the alarm operation is removed, and the power consumption strategy in which the total power consumption value provided by the current state normal power supply unit is multiplied by 0.8 is deleted. The power consumption strategy is set to the power consumption limit strategy for the whole machine power consumption cap setting, and the power consumption cap is continued to be regulated by the total power consumption value provided by the current state normal power supply unit*1.2*0.97.

[0042] If the management controller is restarted, the power consumption limit value strategy and the frequency reduction throttling data of the logic device stored before power failure are read during the restart process. According to the frequency reduction throttling data, the power consumption limit value strategy, and the latest power supply unit state information, the power consumption cap value of the power consumption limit value strategy is adjusted in the manner described in the above embodiments.

[0043] As can be seen from the above, the embodiment directly monitors the in-place state of the power supply unit and the alarm signal through the logic device, and directly triggers the component frequency reduction throttling through the hardware link, so that hardware-level fast response is realized, and the response time is ≤100 ns. The logic device jointly determines through the power supply configuration parameter information combined with the power supply unit state information. When the power supply fails, the frequency reduction throttling operation is started, and the current limiting signal is sent to the management controller. The management controller judges the power supply unit state after the system is stable, comprehensively considers the current power supply unit state and the user setting, makes a decision, regulates through the power consumption limit value strategy, and removes the frequency reduction throttling operation after the strategy takes effect, so that the performance loss of the server is avoided.

[0044] The application further provides another flowchart of a server power consumption management method, which is applied to any logical device of a server, for example, a CPLD, as shown in the figure, and can include the following contents: Figure 5 S501: When receiving the power supply configuration parameter information, polling each power supply unit according to the power supply configuration parameter information to obtain the state information of each power supply unit.

[0045] The power supply configuration parameter information includes the number of required power supply units and the number of redundant power supply units for maintaining the normal operation state of the server.

[0046] S502: If the number of power supply units in the on-site state is less than or equal to the number of required power supply units, and at least one power supply unit state information includes an alarm signal, the notification information indicating that the power consumption configuration mode corresponding to the power supply state is the second power consumption configuration mode is sent.

[0047] In this embodiment, after the logical device obtains the state information of each power supply unit at the current time through S501, it is determined according to the state information of each power supply unit that the number of power supply units in the on-site state is less than or equal to the number of required power supply units, and at least one power supply unit sends an alarm signal, which indicates that the server power supply may fail, that is, the power consumption configuration mode matching the current power supply state is the second power consumption configuration mode, and the notification information indicating that the power consumption configuration mode matching the current power supply state is the second power consumption configuration mode is sent to the management controller. After the management controller receives the notification information, if the current power consumption configuration mode is the first power consumption configuration mode, at this time, the management controller is based on the overcurrent pre-warning protection corresponding to the trigger working current safety range and does not trigger the overcurrent protection corresponding to the trigger working current safety range, determines the target power consumption cap value according to the rated power consumption of each power supply unit, and updates the power consumption limit value strategy according to the target power consumption cap value, and then switches to the second power consumption configuration mode. In this mode, the transition power consumption cap value is determined according to the rated power consumption of each normal power supply unit and the preset power consumption safety adjustment factor, and the power consumption limit value strategy is updated according to the transition power consumption cap value, and the preset power consumption safety adjustment factor is less than 1.

[0048] For example, the logical device can also generate a frequency reduction throttling trigger signal and send the frequency reduction throttling trigger signal directly to the power consumption hardware whose running frequency is to be reduced; generate a throttling processing notification signal and send the throttling processing notification signal; the throttling processing notification signal indicates that the state information of each power supply unit does not match the power supply configuration parameter information.

[0049] For example, the logical device obtains the state information of each power supply unit in real time, and if the throttling interruption notification signal is received during the process of obtaining the state information of each power supply unit, the obtaining of the state information of each power supply unit is stopped; according to the throttling interruption notification signal, the running frequency of the corresponding power consumption hardware is restored.​

[0050] Finally, in order for those skilled in the art to more clearly understand the technical solutions of the present application, the present application also provides an exemplary implementation. The present embodiment takes the management controller as BMC, the logic device as CPLD, and the power supply unit as PSU as an example. The process of dynamically regulating the power consumption of the server can include the following contents: A1: The BMC provides the power supply configuration parameter information of the server configured by the redfish interface: the number N of PSUs required for normal operation of the server and the number M of redundant PSUs.

[0051] A2: During the operation of the BMC, the number of PSUs in place and the rated power consumption are read in real time. If the number of PSUs is less than N+M, a redundant loss alarm is performed. After the server is powered on, the maximum power consumption of the current system is determined through the model information and asset information. If there is no PSU redundant loss alarm, it is determined whether the maximum power consumption of the current system is greater than 1.4 times the rated power consumption of a single PSU. If it is greater, A3 is performed. If it is not greater, A4 is performed.

[0052] A3: The BMC performs a log alarm.

[0053] A4: The BMC issues the power supply configuration parameter information to the CPLD.

[0054] A5: The BMC periodically polls the PSU state, calculates the total power consumption value that can be provided by the normal PSUs in the current state according to the state of each PSU, multiplies the value by 1.2*0.97 to obtain the target power consumption cap value, and sets the value to the power consumption limit value strategy of the power consumption execution manager.

[0055] In this step, if the PSU is normal, the risk of power consumption exceeding the limit can be avoided. However, if a PSU fails or is unexpectedly pulled out within the polling interval, the remaining PSUs may not be able to support the current power consumption, resulting in an overcurrent alarm. A serious overcurrent alarm will cause the PSU to shut down the output, thereby causing the system to shut down. At this time, the CPLD needs to cooperate to adjust the system power consumption in the shortest possible time, that is, the PROHOT pin connected to the processor through the CPLD triggers the processor to reduce the frequency to prevent the remaining PSUs from triggering an overcurrent.

[0056] A6: CPLD polls the PSU state according to the power configuration parameter information issued by BMC in real time, and if a throttle interruption notification sent by BMC is received, the first priority is given to processing the throttle interruption notification, and if there is no throttle interruption, the PSU state information is continuously acquired. The CPLD comprehensively judges whether the frequency reduction throttling needs to be triggered through the PSU in-place state (PRESENT#) and the ALERT# alarm signal: if the number of currently in-place PSUs is less than or equal to the number of PSUs set to be needed, and the ALERT# alarm of a PSU is triggered, the frequency reduction throttling operation is triggered, and a throttle processing notification signal is sent to BMC to inform BMC that the CPLD has triggered the frequency reduction throttling operation.

[0057] A7: BMC receives the throttle processing notification signal of CPLD, performs an alarm operation, stops the periodic polling of the PSU, immediately attempts to read the current PSU state multiple times, and after the PSU state is stable, comprehensively judges the results of the PSU state to determine a transition power consumption cap value by multiplying the total power consumption value that can be provided by the normal PSU by 0.8 by 0.97 or a user preset value, and sets the value to the power consumption execution manager again to perform the whole-machine power consumption capping setting. When the power consumption strategy takes effect, a command is sent to CPLD to release the current frequency reduction throttling operation, and the periodic polling of the PSU state is opened again.

[0058] The user preset value should not be greater than the total power consumption value that can be provided by the normal PSU multiplied by 0.8, and if it is greater, the total power consumption value that can be provided by the normal PSU multiplied by 0.8 is used.

[0059] A8: When the periodic polling of BMC detects that the PSU has been replaced, the alarm operation is released, the power consumption limiting strategy of the transition power consumption cap value setting is deleted, and the power consumption capping regulation is continued through the total power consumption value that can be provided by the normal PSU * 1.2 * 0.97.

[0060] A9: If BMC is restarted, the power consumption execution manager related regulation has been written to the CPU register, and the CPU continues to regulate. If the system is powered off, the CPLD frequency reduction throttling data and the power consumption limit strategy will be stored before power off, and after BMC is powered on, the real-time state is dynamically regulated.

[0061] Of course, the management controller BMC or CPLD can also collect the CPU and memory occupancy rate of the server system, train using any kind of large language model, and issue a PSU state early warning in advance before the business peak.

[0062] From the above, the embodiment realizes the server dynamic power consumption management through dynamic redundancy management and double security mechanism, the BMC provides two working mode switching of stable state and abnormal state: the BMC combines the N+M power supply configuration according to the in-place and state word of the PSU, if the current normal PSU meets the state of N+M, the working mode is stable state, at this time, the set power consumption limit value is 1.2*PSU can provide the rated power consumption sum*0.97, if there is a CPLD frequency reduction throttling signal trigger, or the abnormality of the PSU is monitored, the working mode is abnormal state, at this time, the set power consumption limit value is 0.8*PSU can provide the rated power consumption sum*0.97 or the user-defined value meeting the requirement. The CPLD combines the in-place state and alarm information of the PSU through the power supply configuration parameter information to jointly determine, starts the frequency reduction throttling operation when the power supply fails, and sends the current limiting signal to the BMC, the BMC judges the PSU state after the system is stable, comprehensively considers the current PSU state and user setting to make a decision, and controls through the power consumption limit value strategy, and releases the frequency reduction throttling operation after the strategy takes effect, so that the performance loss of the server is avoided. The in-place state and alarm signal of the power supply unit are directly monitored through the logic device, and the component frequency reduction throttling is directly triggered through the hardware link, so that the hardware level rapid response is realized, and the response time is less than or equal to 100ns.

[0063] It should be noted that there is no strict execution order between the steps in the present application, as long as the logical order is met, the steps can be executed simultaneously, or executed in a certain preset order, Figures 2-5 It is only an illustrative way, and does not mean that only such an execution order can be executed.

[0064] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiment can be realized by means of software and necessary general hardware platform, of course, it can also be realized by hardware, but in many cases, the former is a better embodiment.

[0065] The application also provides a corresponding device for the server power consumption management method, further making the method more practical. The device can be described from the perspective of functional modules and the perspective of hardware. The server power consumption management device provided by the application is introduced below. The device is used to implement the server power consumption management method provided by the application. In this embodiment, the server power consumption management device can include or be divided into one or more program modules, which are stored in a storage medium and executed by one or more processors to complete the server power consumption management method disclosed in embodiment one. The program module referred to in this embodiment refers to a series of computer program instruction segments that can complete a specific function. It is more suitable to describe the execution process of the server power consumption management device in the storage medium than the program itself. The following description will specifically introduce the functions of each program module in this embodiment. The server power consumption management device described below can be mutually corresponding to the server power consumption management method described above.

[0066] Based on the perspective of functional modules, see Figure 6 , Figure 6 The structural diagram of the server power consumption management device provided in this embodiment in a specific implementation manner is shown in the figure. The device is applied to a management controller, which can include: The data acquisition module 601 is configured to acquire power supply configuration parameter information and working current safety range of the power supply unit.

[0067] The dynamic regulation and control module 602 is configured to switch the power consumption configuration mode according to the power supply configuration parameter information and the state information of each power supply unit in the running process. In the first power consumption configuration mode, the overcurrent pre-warning protection corresponding to the working current safety range is triggered, and the overcurrent protection corresponding to the working current safety range is not triggered. The target power consumption cap value is determined according to the rated power consumption of each power supply unit, and the power consumption limit value strategy is updated according to the target power consumption cap value. In the second power consumption configuration mode, the transition power consumption cap value is determined according to the rated power consumption of each normally powered power supply unit and the preset power consumption safety adjustment factor, and the power consumption limit value strategy is updated according to the transition power consumption cap value. The preset power consumption safety adjustment factor is less than 1.

[0068] For example, in some embodiments of this embodiment, the above device can also include a function support identification module, which is configured to acquire the number of power supply units in place and the rated power consumption. If the maximum system power consumption of the server is less than or equal to the power supply demand power consumption under the trigger working current safety range corresponding to the overload protection, the power supply configuration parameter information is sent to the logic device, and the dynamic power consumption cap value configuration function is started.

[0069] As an exemplary implementation of the above embodiment, the function support identification module can be further configured to: determine the total power consumption of the power supply units according to the number of power supply units and the rated power consumption of the power supply units; determine the maximum current triggered by the overcurrent protection corresponding to the maximum fault delay time of the overcurrent protection if the fault delay time of the overload protection triggered by the working current safety range is less than a first preset delay threshold and the maximum fault delay time of the overcurrent protection is greater than a second preset delay threshold; and determine the power consumption critical adjustment factor according to the maximum current and the rated current; and determine the power supply demand power consumption according to the total power consumption of the power supply units and the power consumption critical adjustment factor.

[0070] As another exemplary implementation of the above embodiment, the function support identification module can be further configured to: determine each power consumption hardware and corresponding configuration information according to the server asset information, and determine the rated power consumption of the corresponding power consumption hardware from the configuration information of each power consumption hardware; determine the server basic power consumption according to the server model information and the heat dissipation layout parameters; and determine the maximum system power consumption according to each power consumption hardware and the corresponding rated power consumption, the server basic power consumption, and the margin coefficient.

[0071] As another exemplary implementation of the above embodiment, the function support identification module can be further configured to: perform redundancy loss alarm if the number of power supply units in place does not meet the power supply configuration parameter information; and generate prompt information that does not support dynamic power consumption ceiling value configuration if the maximum system power consumption of the server is greater than the power supply demand power consumption.

[0072] Exemplarily, in some other embodiments of the present embodiment, the dynamic regulation module 602 can be further configured to: determine the power consumption providing information according to the state information of each power supply unit; determine the maximum warning current triggering the overcurrent warning according to the working current safety range, and determine the maximum power consumption adjustment factor according to the maximum warning current and the rated current; and determine the target power consumption ceiling value according to the power consumption providing information, the maximum power consumption adjustment factor, and the first preset safety margin value.

[0073] As an exemplary implementation of the above embodiment, the dynamic regulation module 602 can be further configured to: determine the target power supply unit in the normal working state from each power supply unit in place according to the state information of each power supply unit; and determine the power consumption providing value according to the rated power consumption value of each target power supply unit and the total number of target power supply units.

[0074] For example, in some other embodiments of this embodiment, the dynamic control module 602 can also be used to: determine a new power consumption provision value based on the number of power supply units in the in-situ and normal working state and their corresponding rated power; determine a preset power consumption safety adjustment factor based on the safe range of the working current, and determine a power consumption cap adjustment value based on the new power consumption provision value and the preset power consumption safety adjustment factor; when a user-defined power consumption adjustment request is received, if the user-inputted power consumption cap custom value is less than or equal to the power consumption cap adjustment value, then the power consumption cap custom value is used as the new power consumption cap value; if the power consumption cap custom value is greater than the power consumption cap adjustment value, then the power consumption cap adjustment value is used as the transitional power consumption cap value.

[0075] For example, in some other embodiments of this embodiment, the dynamic control module 602 can also be used to: determine a new power consumption provision value based on the number of power supply units in the in-situ and normal working state and their corresponding rated power; determine a power consumption safety adjustment factor based on the safe range of the working current; and determine a transition power consumption cap value based on the new power consumption provision value, the preset power consumption safety adjustment factor, and the second preset safety margin value.

[0076] For example, in some other embodiments of this embodiment, the dynamic control module 602 can also be used to: acquire the status information of each power supply unit according to a preset status query frequency; when a throttling notification signal is received, generate an alarm signal indicating that the status information of each power supply unit does not match the power configuration parameter information, and stop the periodic polling operation; the throttling notification signal indicates that after the power limit strategy updated according to the target power consumption cap value takes effect, at least one target power supply unit cannot supply power normally; at the same time, continuously acquire the status information of each power supply unit until the status information of the power supply units at adjacent times meets the same preset similar conditions.

[0077] As an exemplary implementation of the above embodiments, the dynamic control module 602 can be further used to: send a throttling interruption notification signal and resume periodic polling operation after the power limit policy updated according to the transition power cap value takes effect; if it is determined according to the latest power supply unit status information that the number of power supply units in normal working condition among the in-situ power supply units is greater than or equal to the number of power supply units required, then generate an alarm cancellation signal, restore the server's power cap value to the target power cap value, and update the latest power limit policy according to the target power cap value.

[0078] From the perspective of functional modules, see Figure 7 , Figure 7 This is a structural diagram of the server power management device provided in this embodiment under another specific implementation. The device is applied to a logic device and may include: The polling module 701 is configured to poll each power supply unit according to the power supply configuration parameter information to obtain the state information of each power supply unit when the power supply configuration parameter information is received.

[0079] The abnormality identification module 702 is configured to send notification information indicating that the power consumption configuration mode corresponding to the power supply state is the second power consumption configuration mode if the number of power supply units in the on state is less than or equal to the number of required power supply units and the at least one power supply unit state information includes an alarm signal, determine a transition power consumption cap value according to the rated power consumption of each normally powered power supply unit and a preset power consumption safety adjustment factor, and update the power consumption limit value strategy according to the transition power consumption cap value, the preset power consumption safety adjustment factor being less than 1; wherein the power consumption configuration mode corresponds to the first power consumption configuration mode, the overcurrent warning protection corresponding to the triggered working current safety range is triggered and the overcurrent protection corresponding to the triggered working current safety range is not triggered, a target power consumption cap value is determined according to the rated power consumption of each power supply unit, and the power consumption limit value strategy is updated according to the target power consumption cap value.

[0080] For example, in some embodiments of the present embodiment, the abnormality identification module 702 can be further configured to generate a frequency reduction throttling trigger signal and send the frequency reduction throttling trigger signal directly to the power consumption hardware whose operating frequency is to be reduced, generate a throttling processing notification signal and send the throttling processing notification signal, and the throttling processing notification signal indicates that the power supply unit state information does not match the power supply configuration parameter information.

[0081] For example, in some other embodiments of the present embodiment, the device can further include an interruption processing module configured to stop obtaining the state information of each power supply unit when the throttling interruption notification signal is received during the process of obtaining the state information of each power supply unit, and restore the operating frequency of the corresponding power consumption hardware according to the throttling interruption notification signal.

[0082] The features of the embodiments of the server power consumption management device can be referred to the related descriptions of the embodiments of the server power consumption management method, which will not be described herein.

[0083] The server power consumption management device mentioned above is described from the perspective of functional modules, and further, the present application provides an electronic device which is described from the perspective of hardware. Figure 8 The structure of the electronic device provided by the embodiments of the present application in one embodiment is shown in the structure diagram. The electronic device includes a memory 801 and a processor 802, the memory 801 stores a computer program, and the processor 802 is configured to run the computer program to perform the steps in any of the above server power consumption management method embodiments.

[0084] The embodiment of the present application further provides a computer readable storage medium, and the computer readable storage medium stores a computer program.

[0085] In an example embodiment, the computer readable storage medium can include, but is not limited to, a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media capable of storing a computer program.

[0086] The embodiment of the present application further provides a computer program product, and the computer program product includes a computer program.

[0087] The embodiment of the present application further provides another computer program product, and the computer program product includes a non-volatile computer readable storage medium, i.e., a non-volatile storage medium, and the non-volatile computer readable storage medium stores a computer program.

[0088] Finally, the present application further provides a server, please refer to Figure 9 , the server can include a logic device 901, a management controller 902 and a target processor 903, and data interaction between the devices is as follows Figure 10The target processor 903 of the embodiment can be a central processor or other processor capable of directly reducing the operating frequency. The logic device 901 is connected to the management controller 902 through a first bus such as I2C or SMbus (System Management Bus), and the management controller 902 sends the power consumption limit policy, power supply configuration information and throttle interrupt signal to the logic device 901 through the bus, and receives the throttle processing notification signal from the logic device 901. The management controller 902 is used to implement the steps of the server power consumption management method as described in any one of the above embodiments, which are executed by the logic device 901. The logic device 901 includes at least a power supply state monitoring interface and a frequency reduction interaction interface, and is connected to each power supply unit through the power supply state monitoring interface, and is connected to the target processor 903 through the frequency reduction interaction interface, and sends the frequency reduction throttle trigger signal to the target processor 903 through the frequency reduction interaction interface. Each power supply unit is connected to the power supply state monitoring interface through at least the signal lines corresponding to PS_PRESENT# (bit signal), PSU_ALERT_N (alert signal), VIN Good (power supply voltage signal) and PSU_PWROK (power supply unit voltage signal). The logic device 901 is connected to the PRC HOT input pin and the MEM HOT input pin of the target processor 903 such as a central processor through the signal lines of the frequency reduction interaction interface. When the frequency reduction throttle signal of the logic device 901 is received, the target processor 903 can be triggered to perform the frequency reduction throttle operation. After the logic device detects that the power consumption cap value is set, if at least one target power supply unit cannot supply power normally (faulty or unplugged), the logic device sends the operating frequency reduction information to the target processor 903 through the frequency reduction interaction interface. The management controller 902 is used to implement the steps of the server power consumption management method as described in any one of the above embodiments, which are executed by the management controller 902. The management controller 902 includes a power consumption execution manager, which is connected to the target processor 903. The power consumption execution manager updates the power consumption limit policy according to the target power consumption cap value or the transition power consumption cap value, and sends it to the target processor 903.

[0089] Further, the logic device 901 further comprises at least one direct connection frequency reduction interface, and the direct connection frequency reduction interface is connected with the power consumption hardware, and the direct connection frequency reduction interface is used for sending the frequency reduction throttling trigger signal to the corresponding power consumption hardware. For example, the logic device 901 is connected to the B30 of the PCIE (Peripheral Component Interconnect Express, high-speed expansion bus) card with the power brake function and the A70 of the OCP (Open Compute Project, network interface card) 3.0 through the reserved line, and when the frequency reduction throttling signal of the CPLD is received, the frequency reduction throttling operation can be triggered. Further, the management controller 902 comprises a management interface and a visual page; the visual page adds a dynamic cap value configuration function on / off option, when a parameter configuration request carrying authorized user information is received through the management interface, the number of demand power supply units and the number of redundant power supply units required to maintain the normal running state of the server are obtained from the parameter configuration request; the power supply configuration parameter information is generated according to the number of demand power supply units and the number of redundant power supply units.

[0090] The server and the power consumption management method thereof, the electronic device, the non-volatile storage medium and the computer program product are described in detail above. Each embodiment in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same or similar parts of each embodiment can be referred to each other. The units and algorithm steps of each example described by each disclosed embodiment are executed by electronic hardware or computer software, which depends on the specific application and design constraints of the technical solution. The professional technicians can use different methods to implement the described functions for each specific application, and such implementation should not be considered beyond the scope of the present application. Without departing from the principles of the present application, the present application can be improved and modified, and these improvements and modifications also fall within the scope of the present application.

Claims

1. A method for managing power consumption of a server, the method comprising: The application is applied to a management controller, comprising: obtaining power supply configuration parameter information and working current safety range of a power supply unit; switching power consumption configuration mode according to the power supply configuration parameter information and state information of each power supply unit in operation; in the first power consumption configuration mode, based on triggering overcurrent early warning protection corresponding to the working current safety range and not triggering overcurrent protection corresponding to the working current safety range, determining a target power consumption ceiling value according to rated power consumption of each power supply unit, and updating power consumption limit value strategy according to the target power consumption ceiling value; in the second power consumption configuration mode, determining a transition power consumption ceiling value according to rated power consumption of each normally powered power supply unit and a preset power consumption safety adjustment factor, and updating power consumption limit value strategy according to the transition power consumption ceiling value, wherein the preset power consumption safety adjustment factor is less than 1.

2. The method of claim 1, wherein, After obtaining the power supply configuration parameter information and the working current safety range of the power supply unit, before determining the target power consumption ceiling value according to the rated power consumption of each power supply unit, further comprising: obtaining the number of power supply units in place and the rated power consumption; if the maximum system power consumption of the server is less than or equal to the power supply demand power consumption under triggering overload protection corresponding to the working current safety range, the power supply configuration parameter information is sent to a logic device, and a dynamic power consumption ceiling value configuration function is started.

3. The method of claim 2, wherein, The power supply configuration parameter information comprises the number of demand power supply units and the number of redundant power supply units required to maintain the normal running state of the server, and after obtaining the number of power supply units in place and the rated power consumption, further comprising: determining total power consumption of the power supply unit according to the number of demand power supply units and the rated power consumption of the power supply unit; if the fault delay time of triggering overload protection corresponding to the working current safety range is less than a first preset delay threshold, and the maximum fault delay time of triggering overcurrent protection is greater than a second preset delay threshold, the maximum current of triggering overcurrent protection corresponding to the maximum fault delay time is determined, and a power consumption critical adjustment factor is determined according to the maximum current and the rated current; determining power supply demand power consumption according to the total power consumption of the power supply unit and the power consumption critical adjustment factor.

4. The method of claim 2, wherein, After obtaining the number of power supply units in place and the rated power consumption, further comprising: determining each power consumption hardware and corresponding configuration information according to server asset information, and determining the rated power consumption of the corresponding power consumption hardware from each power consumption hardware configuration information; determining the server basic power consumption according to the server model information and the heat dissipation layout parameters; determining the maximum system power consumption according to each power consumption hardware and corresponding rated power consumption, server basic power consumption and margin coefficient.

5. The method of claim 2, wherein, After obtaining the number of power supply units in place and the rated power consumption, further comprising: if the number of power supply units in place does not meet the power supply configuration parameter information, a redundant loss alarm is performed; if the maximum system power consumption of the server is greater than the power supply demand power consumption, prompt information not supporting dynamic power consumption ceiling value configuration is generated.

6. The method of claim 1 to 5, wherein, Based on triggering overcurrent early warning protection corresponding to the working current safety range and not triggering overcurrent protection corresponding to the working current safety range, determining a target power consumption ceiling value according to the rated power consumption of each power supply unit, comprising: determining power consumption providing information according to the state information of each power supply unit; determining a maximum pre-warning current triggering an over-current pre-warning according to the working current safety range, and determining a maximum power consumption adjustment factor according to the maximum pre-warning current and the rated current; determining a target power consumption ceiling value according to the power consumption providing information, the maximum power consumption adjustment factor and a first preset safety margin value.

7. The method of claim 6, wherein, The power consumption providing information is determined according to the state information of each power supply unit, including: determining target power supply units in a normal working state from the power supply units in the in-place state according to the state information of each power supply unit; determining a power consumption providing value according to the rated power consumption values of the target power supply units and the total number of the target power supply units.

8. The method of claim 1 to 5, wherein, determining a transition power consumption ceiling value according to the rated power consumption of each normally powered power supply unit and a preset power consumption safety adjustment factor, including: determining a new power consumption providing value according to the number of power supply units in the in-place state and the normal working state and the corresponding rated power; determining a preset power consumption safety adjustment factor according to the working current safety range, and determining a power consumption ceiling adjustment value according to the new power consumption providing value and the preset power consumption safety adjustment factor; when receiving a user-defined power consumption adjustment request, if the user-input power consumption ceiling custom value is less than or equal to the power consumption ceiling adjustment value, the power consumption ceiling custom value is taken as a new power consumption ceiling value; if the power consumption ceiling custom value is greater than the power consumption ceiling adjustment value, the power consumption ceiling adjustment value is taken as a transition power consumption ceiling value.

9. The method of claim 1 to 5, wherein, determining a transition power consumption ceiling value according to the rated power consumption of each normally powered power supply unit and a preset power consumption safety adjustment factor, including: determining a new power consumption providing value according to the number of power supply units in the in-place state and the normal working state and the corresponding rated power; determining a preset power consumption safety adjustment factor according to the working current safety range; determining a transition power consumption ceiling value according to the new power consumption providing value, the preset power consumption safety adjustment factor and a second preset safety margin value.

10. The method of claim 1 to 5, wherein, The power supply configuration parameter information includes the number of demand power supply units and the number of redundant power supply units required to maintain the normal running state of the server, and the state information of each power supply unit is obtained according to a preset state query frequency, before determining the transition power consumption ceiling value according to the rated power consumption of each normally powered power supply unit and a preset power consumption safety adjustment factor, including: when receiving a throttling processing notification signal, an alarm signal indicating that the power supply configuration parameter information does not match the state information of each power supply unit is generated, and the periodic polling operation is stopped; the throttling processing notification signal indicates that after the power consumption limit value strategy updated according to the target power consumption ceiling value takes effect, at least one target power supply unit cannot be normally powered; at the same time, the state information of each power supply unit is continuously obtained until the state information of the power supply unit at the adjacent time meets the preset same similar condition.

11. The method of claim 10, wherein, after updating the power consumption limit value strategy according to the transition power consumption ceiling value, further including: when the power consumption limit value strategy updated according to the transition power consumption ceiling value takes effect, a throttling interruption notification signal is sent, and the periodic polling operation is resumed; If it is determined according to the latest power supply unit state information that the number of power supply units in normal working state among the in-place power supply units is greater than or equal to the required power supply unit number, an alarm cancellation signal is generated, and the power consumption cap value of the server is restored to the target power consumption cap value, and the latest power consumption limit value strategy is updated according to the target power consumption cap value.

12. A method for managing power consumption of a server, the method comprising: The application is applied to a logic device, comprising: When receiving power supply configuration parameter information, polling each power supply unit according to the power supply configuration parameter information to obtain power supply unit state information; If the number of power supply units in in-place state is less than or equal to the required power supply unit number, and at least one power supply unit state information includes an alarm signal, notification information indicating that the power consumption configuration mode corresponding to the power supply state is the second power consumption configuration mode is sent, a transition power consumption cap value is determined according to the rated power consumption of each normally powered power supply unit and a preset power consumption safety adjustment factor, and the power consumption limit value strategy is updated according to the transition power consumption cap value, wherein the preset power consumption safety adjustment factor is less than 1; If the number of power supply units in in-place state is less than or equal to the required power supply unit number, and at least one power supply unit state information includes an alarm signal, notification information indicating that the power consumption configuration mode corresponding to the power supply state is the second power consumption configuration mode is sent, a transition power consumption cap value is determined according to the rated power consumption of each normally powered power supply unit and a preset power consumption safety adjustment factor, and the power consumption limit value strategy is updated according to the transition power consumption cap value, wherein the preset power consumption safety adjustment factor is less than 1; 13. The method of claim 12, wherein, If the number of power supply units in in-place state is less than or equal to the required power supply unit number, and at least one power supply unit state information includes an alarm signal, notification information indicating that the power consumption configuration mode corresponding to the power supply state is the second power consumption configuration mode is sent, a transition power consumption cap value is determined according to the rated power consumption of each normally powered power supply unit and a preset power consumption safety adjustment factor, and the power consumption limit value strategy is updated according to the transition power consumption cap value, wherein the preset power consumption safety adjustment factor is less than 1; Generating a frequency reduction throttling trigger signal and sending the frequency reduction throttling trigger signal directly to the power consumption hardware to be reduced in operating frequency; Generating a throttling processing notification signal and sending the throttling processing notification signal; the throttling processing notification signal indicates that the power supply unit state information does not match the power supply configuration parameter information.

14. The method of claim 12, wherein, Obtaining power supply unit state information, comprising: When receiving a throttling interruption notification signal during the process of obtaining power supply unit state information, stop obtaining power supply unit state information; According to the throttling interruption notification signal, the operating frequency of the corresponding power consumption hardware is restored.

15. An electronic device, comprising: Comprising: A memory for storing a computer program; A processor for executing the computer program to implement the steps of the server power consumption management method according to any one of claims 1 to 14.

16. A non-volatile storage medium, comprising: The computer program is stored on the non-volatile storage medium, and the computer program is executed by the processor to implement the steps of the server power consumption management method according to any one of claims 1 to 14.

17. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instruction is executed by the processor to implement the steps of the server power consumption management method according to any one of claims 1 to 14.

18. A server, characterized by Comprising a logic device, a management controller and a target processor; The logic device is connected to the management controller through a first bus, and is configured to implement the steps of the server power consumption management method according to any one of claims 12 to 14 when executing a computer program; the logic device comprises at least a power supply state monitoring interface and a frequency reduction interaction interface, the power supply state monitoring interface is connected to each power supply unit, the frequency reduction interaction interface is connected to the target processor, and the frequency reduction throttling trigger signal is sent to the target processor through the frequency reduction interaction interface; The management controller is configured to implement the steps of the server power consumption management method according to any one of claims 1 to 11 when executing a computer program; The management controller comprises a power consumption execution manager, the power consumption execution manager is connected to the target processor, the power consumption execution manager updates the power consumption limit policy according to the target power consumption cap value or the transition power consumption cap value, and sends the power consumption limit policy to the target processor.

19. The server of claim 18, wherein, The logic device further comprises at least one direct connection frequency reduction interface, the direct connection frequency reduction interface is connected to the power consumption hardware, and the frequency reduction throttling trigger signal is sent to the corresponding power consumption hardware through the direct connection frequency reduction interface.

20. The server of claim 18, wherein, The management controller further comprises a management interface; When a parameter configuration request carrying authorized user information is received through the management interface, the number of required power supply units and the number of redundant power supply units required to maintain the normal operation state of the server are obtained from the parameter configuration request; The power supply configuration parameter information is generated according to the number of required power supply units and the number of redundant power supply units.

Images