Power supply distribution method, power supply distribution circuit, storage medium and electronic equipment

Abstract

The embodiment of the invention provides a power supply distribution method, a power supply distribution circuit, a storage medium and electronic equipment, and the power supply distribution method comprises the steps: carrying out the rectification conversion of at least two AC power supplies through at least two rectification units, so as to output a first DC power supply, each rectification unit comprises an automatic conversion unit and a power factor correction unit, and the number of the rectification units is consistent with that of the alternating current; receiving the first direct-current power supply through at least one step-down conversion unit, and adjusting the voltage corresponding to the first direct-current power supply to generate a second direct-current power supply; the second direct-current power supply is distributed to a plurality of power supply rails through a power supply redistribution module, and the power supply rails are used for supplying power to a target processor.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of circuit design, and more specifically, to a power distribution method and a power distribution circuit, a storage medium, and an electronic device. Background Art

[0002] As the market demand for artificial intelligence, cloud computing, and big data continues to increase, the server core processor CPU, accelerator graphics processing unit (GPU), tensor processing unit (TPU) and other application-specific integrated circuits (Application-Specific Integrated Circuit, referred to as ASIC) are constantly developing, the manufacturing of single chips is constantly upgraded, and the power consumption of single chips has risen sharply to the kW (kilowatt) level; the power consumption of a single node in the server system has also increased significantly with the continuous upgrading of the core processor; the power supply of a single node of the server is taken from the power grid to the AC-DC (alternating current-direct current converter) rectification, and from the DC output of a single power supply unit (Power Supply Unit, referred to as PSU) to the processor power voltage. Due to the space limitations of a single node, how to provide stable and efficient power supply to processors with increasingly high power consumption in a smaller space has become an important problem that needs to be urgently solved in the power supply field.

[0003] The related technologies solve the above problem in the following three ways: N+N redundant power supply architecture, N+1 redundant power supply architecture and multiple power supply architecture. However, the above methods all have the problem that the sum of the power supported by all power levels is much greater than the total power required by the processor. The large number of power levels makes it difficult to improve the power density of the server.

[0004] Therefore, the problem of low power density of the redundant power supply architecture caused by the integration of the rectifier unit and the buck unit in the N+N or N+1 redundant power supply architecture in the related art has not yet been effectively solved. Summary of the invention

[0005] The embodiments of the present application provide a power distribution method and a power distribution circuit, a storage medium and an electronic device, so as to at least solve the problem of low power density of the redundant power supply architecture caused by the integration of the rectifier unit and the step-down unit in the N+N or N+1 redundant power supply architecture in the related art.

[0006] According to an embodiment of the present application, a power distribution method is provided, which specifically includes: rectifying and converting at least two AC power supplies through at least two rectifying units to output a first DC power supply, wherein each rectifying unit includes: an automatic conversion unit and a power factor correction unit, and the number of rectifying units is consistent with the number of AC currents; receiving the first DC power supply through at least one step-down conversion unit, and adjusting the voltage corresponding to the first DC power supply to generate a second DC power supply; distributing the second DC power supply to multiple power rails through a power redistribution module, wherein the multiple power rails are used to power a target processor.

[0007] In an exemplary embodiment, at least two AC power supplies are rectified and converted by at least two rectifier units to output a first DC power supply, including: determining whether there is a failed AC power supply in the at least two AC power supplies; when it is determined that the first AC power supply in the at least two AC power supplies is a failed AC power supply, disconnecting the connection between the first AC power supply and the first rectifier unit by the automatic conversion unit, and maintaining the connection between other AC power supplies and other rectifier units by the automatic conversion unit, wherein the at least two AC power supplies include: the first AC power supply and the other AC power supplies, and the at least two rectifier units include: the first rectifier unit and the other rectifier units, and the first AC power supply and the first rectifier unit are in a series relationship; rectifying and converting the other AC power supplies by the other rectifier units to output the first DC power supply; when it is determined that the at least two AC power supplies are both valid AC power supplies, maintaining the connection between the at least two AC power supplies and the at least two rectifier units by the automatic conversion unit, and rectifying and converting the at least two AC power supplies by the at least two rectifier units to output the first DC power supply.

[0008] In an exemplary embodiment, at least two AC power sources are rectified and converted by at least two rectifying units to output a first DC power source, including: rectifying the at least two AC power sources into pulsating DC power by the at least two rectifying units; smoothing the pulsating DC power by a filter capacitor, and adjusting a DC voltage corresponding to the pulsating DC power by a voltage stabilizing circuit; and adjusting a waveform corresponding to the smoothed and stabilized pulsating DC power to generate the first DC power source.

[0009] In an exemplary embodiment, adjusting the voltage corresponding to the first DC power supply to generate a second DC power supply includes: receiving a target signal sent by the target processor, and determining a real-time power demand corresponding to the target processor through the target signal; and adjusting the current and voltage of the first DC power supply through a DC-DC converter in the at least one step-down conversion unit and the real-time power demand to generate the second DC power supply.

[0010] In an exemplary embodiment, distributing the second DC power supply to multiple power rails through a power redistribution module includes: detecting current demand information of each power rail through the power redistribution module; controlling the conduction state between each common power level and each power rail according to the current demand information and multiple switching switches in the power redistribution module to distribute the second DC power supply to the multiple power rails, wherein the at least one step-down conversion unit and the multiple power rails are connected through at least one power level, and the at least one power level includes: each dedicated power level corresponding to each power rail and a common power level corresponding to the multiple power rails.

[0011] In an exemplary embodiment, controlling the conduction state between each common power stage and each power rail according to the current demand information and a plurality of switching switches in the power redistribution module to distribute the second DC power to the plurality of power rails includes: acquiring power supply capability information of each dedicated power stage corresponding to each power rail; determining a first current required by each power rail according to the current demand information, and determining a second current allowed to be provided by each dedicated power stage for each power rail according to the power supply capability information; determining a magnitude relationship between the first current and the second current; and when the magnitude relationship indicates that the first current required by any power rail is greater than the second current corresponding to any dedicated power stage, closing a target switching switch among the plurality of switching switches to adjust the conduction state between the common power stage corresponding to any power rail and any power rail to a connected state, wherein any power rail is in series with any dedicated power stage, and the target switching switch is used to control the on / off state between any power rail and the common power stage corresponding to any power rail.

[0012] In an exemplary embodiment, after the second DC power supply is distributed to multiple power rails through the power redistribution module, the method further includes: monitoring the target state parameter corresponding to each buck conversion unit, and obtaining the preset state parameter corresponding to the at least one buck conversion unit; comparing each target state parameter with the preset state parameter respectively, and determining the working state corresponding to each buck conversion unit according to the comparison result; when the comparison result indicates that the preset state parameter contains each target state parameter, determining that the working state corresponding to the at least one buck conversion unit is a normal state; when the comparison result indicates that the preset state parameter does not contain the target state parameter corresponding to the first buck conversion unit of the at least one buck conversion unit, determining that the working state corresponding to the first buck conversion unit is an abnormal state; sending the working state corresponding to the first buck conversion unit to the target object, so that the target object removes the first buck conversion unit from the slot corresponding to the first buck conversion unit, and performs fault repair on the first buck conversion unit.

[0013] According to another embodiment of the present application, a power distribution circuit is provided, comprising: at least two AC power supplies, at least two rectifier units respectively connected in series with each AC power supply, a buck conversion unit connected in series with the at least two rectifier units, and a power redistribution module connected in series with the buck conversion unit, wherein: the at least two rectifier units are used to rectify and convert the at least two AC power supplies to output a first DC power supply, wherein each rectifier unit comprises: an automatic conversion unit and a power factor correction unit, and the number of rectifier units is consistent with the number of AC currents; the at least one buck conversion unit is used to receive the first DC power supply and adjust the voltage corresponding to the first DC power supply to generate a second DC power supply; the power redistribution module is used to distribute the second DC power supply to multiple power rails, wherein the multiple power rails are used to power a target processor.

[0014] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps in the above method embodiment when running.

[0015] According to another embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in the above method embodiment.

[0016] According to another embodiment of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiment are implemented.

[0017] Through the power distribution method of the present application, at least two rectifier units are used to rectify and convert at least two AC power supplies that are consistent with the number of rectifier units to output a first DC power supply, wherein each rectifier unit includes an automatic conversion unit and a power factor correction unit; the first DC unit is received by at least one buck conversion unit, and the voltage corresponding to the first DC power supply is adjusted to generate a second DC power supply; the second DC power supply is distributed to multiple power rails for powering the target processor through a power redistribution module. That is, after the embodiment of the present application rectifies and converts the AC power supply into a first DC power supply through a rectifier unit including an automatic conversion unit and a power factor correction unit, the first DC power supply is converted into a second DC power supply through a buck conversion unit, and then the second DC power supply is distributed to multiple power rails through a power redistribution module. That is, the number of automatic conversion units in the embodiment of the present application does not need to increase with the number of buck conversion units, which reduces the number of power supply power elements, reduces the cost of power elements of the power supply unit, and increases the upper limit of the system power density improvement. It solves the problem of low power density of the redundant power supply architecture caused by the integration of the rectifier unit and the buck unit in the N+N or N+1 redundant power supply architecture in the related art. This improves the system power density. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a hardware structure block diagram of a computer terminal of a power distribution method according to an embodiment of the present application;

[0019] Figure 2 is a flow chart of a power distribution method according to an embodiment of the present application;

[0020] Figure 3 It is a schematic diagram of an N+N redundant power supply architecture of a server power supply unit in the related art;

[0021] Figure 4 It is a schematic diagram of an N+1 redundant power supply architecture of a server power supply unit in the related art;

[0022] Figure 5a It is a schematic diagram of multiple power supply of a processor in the related art;

[0023] Figure 5b It is a schematic diagram of the layout of existing key electronic components in the related technology;

[0024] Figure 6 This is a schematic diagram of power supply from the mains to the server node processor according to an optional embodiment of the present application;

[0025] Figure 7 is a schematic diagram of a power supply method of a power supply unit according to an optional embodiment of the present application;

[0026] Figure 8 This is a schematic diagram of power switching of a PRM power redistribution module according to an optional embodiment of the present application;

[0027] Fig. 9 It is a schematic diagram of the control logic relationship of the PRM power redistribution module according to an optional embodiment of the present application;

[0028] Fig.10 is a schematic diagram of the layout of key electronic components according to an optional embodiment of the present application;

[0029] Fig.11 is a schematic diagram of a power distribution circuit according to an embodiment of the present application. DETAILED DESCRIPTION

[0030] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0032] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1 1 is a hardware structure block diagram of a computer terminal of a power distribution method according to an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the above-mentioned computer terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0033] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for determining the interactive state in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0034] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0035] Figure 2 is a flow chart of a power distribution method according to an embodiment of the present application, which can be applied to Figure 1 In a computer terminal, such as Figure 2 As shown, the process includes the following steps:

[0036] Step S202, rectifying and converting at least two AC power supplies by at least two rectifying units to output a first DC power supply, wherein each rectifying unit includes: an automatic conversion unit and a power factor correction unit, and the number of rectifying units is consistent with the number of AC currents;

[0037] Each rectifier unit is composed of an automatic conversion switch and a power factor correction unit connected in series, and each automatic conversion unit is connected in series with an AC power source.

[0038] Step S204, receiving the first DC power supply through at least one step-down conversion unit, and adjusting a voltage corresponding to the first DC power supply to generate a second DC power supply;

[0039] Step S206: distribute the second DC power to a plurality of power rails through a power redistribution module, wherein the plurality of power rails are used to supply power to a target processor.

[0040] Through the power distribution method of the present application, at least two rectifier units are used to rectify and convert at least two AC power supplies that are consistent with the number of rectifier units to output a first DC power supply, wherein each rectifier unit includes an automatic conversion unit and a power factor correction unit; the first DC unit is received by at least one buck conversion unit, and the voltage corresponding to the first DC power supply is adjusted to generate a second DC power supply; the second DC power supply is distributed to multiple power rails for powering the target processor through a power redistribution module. That is, after the embodiment of the present application rectifies and converts the AC power supply into a first DC power supply through a rectifier unit including an automatic conversion unit and a power factor correction unit, the first DC power supply is converted into a second DC power supply through a buck conversion unit, and then the second DC power supply is distributed to multiple power rails through a power redistribution module. That is, the number of automatic conversion units in the embodiment of the present application does not need to increase with the number of buck conversion units, which reduces the number of power supply power elements, reduces the cost of power elements of the power supply unit, and increases the upper limit of the system power density improvement. It solves the problem of low power density of the redundant power supply architecture caused by the integration of the rectifier unit and the buck unit in the N+N or N+1 redundant power supply architecture in the related art. This improves the system power density.

[0041] Optionally, the above-mentioned step S202 of rectifying and converting at least two AC power supplies through at least two rectifier units to output a first DC power supply includes: determining whether there is a failed AC power supply among the at least two AC power supplies; when it is determined that the first AC power supply among the at least two AC power supplies is a failed AC power supply, disconnecting the connection between the first AC power supply and the first rectifier unit through the automatic conversion unit, and maintaining the connection between other AC power supplies and other rectifier units through the automatic conversion unit, wherein the at least two AC power supplies include: the first AC power supply and the other AC power supplies, and the at least two rectifier units include: the first rectifier unit and the other rectifier units, and the first AC power supply and the first rectifier unit are in series relationship; rectifying and converting the other AC power supplies through the other rectifier units to output the first DC power supply; when it is determined that the at least two AC power supplies are both valid AC power supplies, maintaining the connection between the at least two AC power supplies and the at least two rectifier units through the automatic conversion unit, and rectifying and converting the at least two AC power supplies through the at least two rectifier units to output the first DC power supply.

[0042] It is understandable that there may be a failed AC power source in at least two AC power sources, so it is necessary to first determine whether there is a failed AC power source in at least two AC power sources, and then output the first DC power in different ways according to different situations. Specifically:

[0043] Assume that there is a server whose power supply unit adopts a circuit with two AC power inputs. Each AC power is connected to its own rectifier unit through an automatic conversion unit. The rectifier unit is responsible for converting AC power into DC power to provide a stable power supply for the server.

[0044] When it is determined that the first AC power source among at least two AC power sources is a failed AC power source, and the other AC power sources are valid AC power sources, the automatic transfer switch connected to the first AC power source will disconnect the first AC power source from the first rectifier unit to prevent the faulty power source from affecting the rectification process. The other automatic transfer switches connected to the other AC power sources are notified to maintain the connection between the other AC power sources and the other rectifier units to ensure that the valid AC power source can continue to supply power to the server.

[0045] When it is determined that at least two AC power sources are valid AC power sources, the automatic conversion unit in at least two rectifier units will maintain the connection between all AC power sources and all rectifier units, allowing at least two AC power sources to be rectified simultaneously.

[0046] The design of at least two rectifier units can automatically switch to the effective power supply when the AC power fails, thus avoiding power interruption. At the same time, when both power supplies are effective, they can be used to supply power at the same time, improving the stability of power supply and system redundancy. It can effectively deal with power grid fluctuations or power failures to ensure the normal operation of the server.

[0047] Optionally, the above-mentioned step S202 of rectifying and converting at least two AC power supplies through at least two rectifying units to output a first DC power supply includes: rectifying the at least two AC power supplies into pulsating DC power through the at least two rectifying units; smoothing the pulsating DC power through a filter capacitor, and adjusting the DC voltage corresponding to the pulsating DC power through a voltage stabilizing circuit; adjusting the waveform corresponding to the pulsating DC power after smoothing and voltage stabilization to generate the first DC power supply.

[0048] It is understood that the process of rectification conversion includes:

[0049] Rectification to pulsating DC: When AC power is input to the power supply unit, it passes through at least two rectifier units (such as RectifierO and Rectifier1). The function of the rectifier unit is to convert AC power into DC power, but the rectified DC power is pulsating, that is, the voltage fluctuates over time. This is because the sinusoidal wave characteristics of AC power, after full-wave or half-wave rectification, although the negative half cycle is eliminated, the voltage of the positive half cycle still changes periodically over time. For example, when using a diode bridge rectifier, the AC voltage is converted into a series of forward pulse voltages to form pulsating DC power.

[0050] Filtering and smoothing: The pulsating DC power is then smoothed by the filter capacitor. The filter capacitor can store charge and release it when the voltage fluctuates, thereby smoothing the voltage waveform and reducing pulsation. In the power conversion process, the filter capacitor is usually placed between the rectifier and the subsequent voltage stabilization circuit to ensure the stability and quality of the DC voltage and provide a cleaner DC input for subsequent voltage stabilization and conversion.

[0051] Voltage regulation: The smoothed pulsating DC power also needs to be regulated by a voltage regulator circuit. The voltage regulator circuit ensures that the output voltage remains constant, even if the input voltage or load conditions change. This is usually achieved by using a linear regulator, a switching regulator, or other voltage regulation techniques. The purpose of the voltage regulator circuit is to provide a stable DC voltage that is not affected by input fluctuations to meet the strict power requirements of the components inside the server.

[0052] Waveform adjustment to generate the first DC power supply: After rectification, filtering and smoothing, and voltage regulation, the resulting DC voltage waveform needs to be further adjusted to ensure that it fully meets the requirements of the processor and other key components in the server. This may involve fine-tuning the voltage accuracy, stability, and response time characteristics to ensure that the output DC power supply can provide the required current and voltage and has good dynamic responsiveness to adapt to the changes in server demand under different load conditions.

[0053] Through the above steps, the server power supply unit can convert the unstable AC power into a stable, high-quality first DC power, providing reliable and efficient power support for various electronic components inside the server.

[0054] Optionally, the above-mentioned step S204 of adjusting the voltage corresponding to the first DC power supply to generate a second DC power supply includes: receiving a target signal sent by the target processor, and determining the real-time power requirement corresponding to the target processor through the target signal; adjusting the current and voltage of the first DC power supply through the DC-DC converter in the at least one step-down conversion unit and the real-time power requirement to generate the second DC power supply.

[0055] It is understandable that the output current and voltage can be adjusted by the step-down conversion unit to generate a second DC power supply suitable for the processor. Specifically: Receive the target signal sent by the target processor and determine the real-time power demand. In the power supply architecture of the server, the processor sends a target signal (usually a SW Control signal), which contains information about the current operating status and required power of the processor. For example, when the processor performs intensive calculations, it requires more power to support its operation. The target signal notifies the power supply unit that the real-time power demand of the processor has increased. This demand may be reflected in the form of current, that is, the processor needs to extract more current from the power rail.

[0056] The first DC power supply is adjusted by the buck conversion unit to generate the second DC power supply. Once the real-time power demand of the processor is determined, the buck conversion unit in the power supply unit will adjust the current and voltage of the first DC power supply (which can be DC_IN) according to this demand. The function of the DC-DC converter is to convert a fixed DC voltage into another variable DC voltage. This voltage and current adjustment is based on the feedback of the real-time power demand of the processor.

[0057] In this way, the server power supply unit can dynamically adjust the output to respond to the real-time power demand changes of the processor, ensuring the efficient and stable operation of the server system. The above technical solution can save energy and reduce power waste at low loads, and can also ensure that the processor can obtain sufficient power at high loads, thereby improving the overall performance and reliability of the server.

[0058] Optionally, the above-mentioned step S206 of distributing the second DC power supply to multiple power rails through a power redistribution module includes: detecting current demand information of each power rail through the power redistribution module; controlling the conduction state between each common power level and each power rail according to the current demand information and multiple switching switches in the power redistribution module, so as to distribute the second DC power supply to the multiple power rails, wherein the at least one step-down conversion unit and the multiple power rails are connected through at least one power level, and the at least one power level includes: each dedicated power level corresponding to each power rail and a common power level corresponding to the multiple power rails.

[0059] Among them, controlling the conduction state between each common power level and each power rail according to the current demand information and multiple switching switches in the power redistribution module to distribute the second DC power supply to the multiple power rails includes: obtaining the power supply capability information of each dedicated power level corresponding to each power rail; determining the first current required by each power rail according to the current demand information, and determining the second current allowed to be provided by each dedicated power level for each power rail according to the power supply capability information; determining the magnitude relationship between the first current and the second current; and when the magnitude relationship indicates that the first current required by any power rail is greater than the second current corresponding to any dedicated power level, closing the target switching switch among the multiple switching switches to adjust the conduction state between the common power level corresponding to any power rail and any power rail to a connected state, wherein any power rail is connected in series with any dedicated power level, and the target switching switch is used to control the on-off state of any power rail and the common power level corresponding to any power rail.

[0060] It is understandable that the power redistribution module can dynamically adjust the connection between the power stage and the power rail to meet the real-time current requirements of different power rails of the processor. Specifically:

[0061] Detecting current demand information: The power redistribution module monitors the current output of each power stage through the signal lsense connected to the power stage, and obtains the voltage of each power rail on the processor side through the Vsense signal. This information helps the power redistribution module understand the current demand of each power rail in real time to decide how to redistribute the output of the common power stage.

[0062] Control the conduction state to distribute the second DC power supply: The power redistribution module determines whether the common power stage needs to be connected to a power rail by controlling multiple internal switches (e.g., Q1 to Q8) based on the above current demand information and the power supply capability information of each dedicated power stage (assuming constant current capability). The specific steps include:

[0063] Obtaining power capability information: The power redistribution module obtains the power capability of the dedicated power stage corresponding to each power rail (assuming 30A). This helps determine whether the dedicated power stage is sufficient to meet the current requirements of the power rail.

[0064] Determine the relationship between current demand and power supply capability: The power redistribution module determines the real-time current (first current) required by each power rail based on the Isense signal. If the first current required by a power rail exceeds the power supply capability (second current) of its dedicated power stage, the power redistribution module will take action.

[0065] Adjusting the conduction state: If the first current is greater than the second current, it means that the dedicated power stage cannot meet the real-time demand of the power rail. In this case, the power redistribution module closes the target switch (such as Q4, Q5 or Q6) to adjust the connection between the common power stage and the power rail to the connected state. This means that in addition to the dedicated power stage, the common power stage will also start to supply power to the power rail to meet its current demand that exceeds the power supply capability of the dedicated power stage.

[0066] The above technical solution ensures that the server power supply unit can be flexibly adjusted to cope with the changes in the current demand of different power rails of the processor, while improving power density and efficiency. By utilizing the dynamic allocation characteristics of the common power level, even if the output capacity of a single or multiple dedicated power levels reaches the limit, the power redistribution module can be quickly adjusted to maintain or improve the power supply capacity and maintain the high performance of the server.

[0067] Optionally, after the second DC power supply is distributed to multiple power rails through the power redistribution module in the above step S206, the method also includes: monitoring the target state parameters corresponding to each buck conversion unit, and obtaining the preset state parameters corresponding to the at least one buck conversion unit; comparing each target state parameter with the preset state parameter respectively, and determining the working state corresponding to each buck conversion unit according to the comparison result; when the comparison result indicates that the preset state parameter contains each target state parameter, determining that the working state corresponding to the at least one buck conversion unit is a normal state; when the comparison result indicates that the preset state parameter does not contain the target state parameter corresponding to the first buck conversion unit of the at least one buck conversion unit, determining that the working state corresponding to the first buck conversion unit is an abnormal state; sending the working state corresponding to the first buck conversion unit to the target object, so that the target object removes the first buck conversion unit from the slot corresponding to the first buck conversion unit, and repairs the fault of the first buck conversion unit.

[0068] It is understandable that the embodiment of the present application can also detect the working state of the buck conversion unit. Specifically, in the server power supply unit, the operating state of each buck conversion unit (for example, DC-DC0 to DC-DC5) is continuously monitored, including monitoring key parameters such as voltage, current, and temperature (i.e., target state parameters). For example, through the lsense signal, the controller can monitor the output current of each power level (i.e., buck conversion unit) and the output voltage obtained through the Vsense signal.

[0069] It is also necessary to preset a set of normal operation preset state parameters, which are set according to the design specifications and operating environment of the buck converter unit. The preset state parameters define the voltage, current and temperature range of each buck converter unit during normal operation.

[0070] Once the target state parameters are monitored, these parameters are compared with the preset state parameters to determine whether the working state of the buck converter unit is normal. If the target state parameters are within the range of the preset state parameters, it means that the buck converter unit is in a normal state. If the target state parameters exceed the range of the preset state parameters, for example, the current or voltage suddenly increases or decreases abnormally, or the temperature is too high, then the buck converter unit may be in an abnormal state.

[0071] When the comparison result indicates that the target state parameters of all buck conversion units are within the preset state parameter range, it is determined that the working states of all buck conversion units are normal, which means that the operation of the entire power supply unit is healthy and can continue to provide stable and reliable power support for the server.

[0072] When the comparison result indicates that the preset state parameters do not contain the target state parameters of a buck conversion unit (e.g., DC-DC2), it is determined that the buck conversion unit is in an abnormal state. In this case, the system will send the working state of the buck conversion unit to the target object (which may be the management system or maintenance personnel of the server). After receiving the abnormal state report, the target object can take measures to remove the abnormal buck conversion unit from the corresponding slot for troubleshooting and repair.

[0073] Through the above-mentioned monitoring, comparison and fault management mechanisms, the server power supply unit can promptly detect and isolate abnormal step-down conversion units to avoid the spread of faults, while maintaining power supply to other normal units to ensure stable and efficient operation of the server.

[0074] In order to better understand the process of the above-mentioned power distribution method, the implementation process of the above-mentioned power distribution method is described below in combination with an optional embodiment, but it is not used to limit the technical solution of the embodiment of the present application.

[0075] Among them, the optional embodiments of the present application may include the following terms:

[0076] PSU: Power Supply Unit, power supply unit;

[0077] PFC: Power Factor Correction, power factor correction unit;

[0078] PDU: Power Distribution Unit, power distribution unit;

[0079] PRM: Power Re-mapping Module, power redistribution module;

[0080] ATS: Automatic Transformer Switch, automatic transfer switch;

[0081] PSU_RCT: PSU Rectifier, PSU rectifier part;

[0082] PSU_CM: PSU Current Multiplier, PSU voltage reduction part;

[0083] Power Stage: power stage;

[0084] Power Rail: Power rail.

[0085] The following power supply methods exist in the related art:

[0086] 1) Figure 3 Schematic diagram of N+N redundant power supply architecture of server power supply unit in related technology, such as Figure 3 As shown, the power distribution unit A (PDU_A for short) and PDU_B are two 1+1 redundant 220V_AC (alternating current) mains in the data center to ensure reliability. To ensure the N+N redundancy of the server system power supply unit, the A-way 220V_AC network is connected to PSU0 composed of a power factor correction unit 0Power Factor Correction (PFC0 for short) and DC-DC0 (direct current-direct current 0), PSU1 composed of PFC1 and DC-DC1, and PSU2 composed of PFC2 and DC-DC2; the B-way 220VAC network is connected to PSU3 composed of PFC3 and DC-DC3, PSU4 composed of PFC4 and DC-DC4, and PSU5 composed of PFC5 and DC-DC5. A single PSU converts AC power into DC power through PFC unit rectification, and then steps down the voltage to the voltage used by the internal power equipment of the server through the DC-DC unit. PSU0-PSU5 output DC_OUT in parallel to supply power to each power-consuming device inside the server, realizing N+N redundant power supply; when one AC bus fails or the PSU fails, the server power supply unit can still provide sufficient power support for the system to ensure the reliability of system power supply.

[0087] However, to ensure the reliability of the power supply unit, high-power nodes use an N+N redundant power supply architecture and use a large number of PSUs. Due to space limitations, the power that can be supplied in a standard server node is relatively small. As the power consumption of a single server node continues to increase, in addition to increasing the power density of a single PSU, the only option is to increase the number of PSUs to provide higher power support for each node.

[0088] 2) Figure 4 Schematic diagram of N+1 redundant power supply architecture of server power supply unit in related technology, such as Figure 4 As shown in the figure, 220V AC_A and 220V AC_B are two 220AC mains power supplies for the data center. Each automatic transfer switch (ATS) PSU consists of an ATS automatic transfer switch unit, a PFC power factor correction unit, and a DC-DC step-down unit. The two 220V AC busbars A and B are connected to each ATS PSU in the system at the same time. When the A busbar works normally and the B busbar does not work, the A busbar supplies power to the system; when the B busbar works normally and the A busbar does not work, the B busbar supplies power to the system; when both busbars work normally, the system decides which busbar to use for power supply; when both busbars work normally and one of them suddenly fails, the other busbar can be switched to supply power to the system to provide sufficient power support for the system and ensure the reliability of the system power supply.

[0089] However, compared with the N+N power supply architecture of the server power supply unit, the newly added ATS automatic transfer switch unit increases the space requirements of a single ATS PSU, and the entire PSU is larger, making it difficult to increase the power density. At the same time, when the power of a single server node is high and the number of PSUs is large, the number of AC cables doubles, making it difficult to plug and manage cables, and a single PSU is heavy, causing certain difficulties in data center operation and maintenance.

[0090] 3) Figure 5a It is a schematic diagram of multiple power supply of the existing processor in the related technology. Figure 5b It is a schematic diagram of the layout of existing key electronic components in related technologies, such as Figure 5a As shown, DC_IN represents the system voltage after the AC power is rectified and stepped down by the server power supply unit.

[0091] The Power Stagexx power stage is a DC-DC step-down conversion unit. DC_IN converts the voltage to the voltage required by the processor through multiple power stages to power the processor. The processor power supply usually consists of multiple Power Rails. According to the voltage and current requirements of the processor for each power rail, each power rail is output by one power stage or multiple power stages in parallel.

[0092] like Figure 5bAs shown in the figure, the system power DC_System converts the voltage required by the processor through the power stage and inductor to power the processor. The power stage and the processor are on the same side of the motherboard.

[0093] In the above technical solution, usually one power level can support up to about 30A, and the processor power rail voltage is usually between 0.8V-1.8V; however, in order to ensure the stable operation of a 300W processor, when the current of each power rail is in a steady state or changes dynamically, the output voltage of the power rail remains within the required range of the processor, and a processor usually requires more than 20 power levels. The sum of the power that can be supported by all power levels is much greater than the total power required by the processor. The large number of power levels poses a certain degree of difficulty in improving the power density of the server. The power level is on the side of the processor, the power supply path is long, the loss is large, the efficiency is low, and it is difficult to improve the power density.

[0094] The optional embodiment of the present application aims to provide a server node power supply method and control strategy. While ensuring the working reliability of the server power supply unit, the optional embodiment of the present application solves the problems of a large number of PSUs under the N+N architecture, large chassis space occupied, and difficulty in improving power density; and solves the operation and maintenance difficulties of the PSU under the N+1 architecture, which is large in size and weight, and has many AC input cables and is difficult to manage. At the same time, the power supply method of the optional embodiment of the present application aims to solve the problem of a large number of processor power supply power levels, large space occupied, and difficulty in improving power density.

[0095] Figure 6 Schematic diagram of power supply from mains to server node processor according to an optional embodiment of the present application, such as Figure 6 As shown in the figure, 220V AC_A and 220V AC_B are two 220VAC mains supplies (i.e. at least two AC power supplies) for 1+1 redundancy in the data center. The two mains supplies are connected to two independent ATSs ( Figure 6 The ATS0 and ATS1 in the circuit are used to switch between two mains. Then, two PFC ( Figure 6 PFC0 and PFC1) are rectified and then output DC_IN (i.e., the first DC power supply), which is the DC-DC ( Figure 6 DC_IN passes through the DC-DC step-down conversion unit (i.e., step-down conversion unit) to output DC_System (DC system power supply, i.e., the second DC power supply) to power each sub-unit in the server node. Figure 6 Only the subunits that power the processor are shown. DC_System undergoes multiple power stage voltage conversions to generate the power rails required by the processor ( Figure 6The voltages of Power Stage1-1, Power Stage1-2, Power Stage1-3, Power StageC-1, Power StageC-2, Power StageC-3, PowerStageC-4, PowerStage2-1, Power Stage2-2, Power Stage2-3 in the power supply are connected to the power redistribution module ( Figure 6 The power re-mapping module in the power distribution module redistributes the output of each power stage and then connects it to each power rail (i.e. Figure 6 Power Rail1-4) in the processor (i.e. Figure 6 The processor in the system is used for power supply.

[0096] Figure 7 is a schematic diagram of a power supply method of a power supply unit according to an optional embodiment of the present application, such as Figure 7 As shown, two mains supplies are connected to PSU Rectifier0 (rectifier part 0 of power supply unit, i.e. rectifier unit, referred to as PSU_RCT) and PSU Rectifier1 respectively to supply power to the node. PSU Rectifier0 is composed of automatic transfer switch ATS0 (i.e. automatic transfer unit) and power factor correction unit PFC0; PSU Rectifier1 is composed of automatic transfer switch ATS1 and power factor correction unit PFC1;

[0097] The function of PSU Rectifier, or PSU_RCT, is to automatically switch between two AC inputs and rectify AC-DC to DC. Each PSU_RCT (PSU rectifier part) is a pluggable component and can be maintained separately when there is a problem with the function.

[0098] To ensure power supply reliability, two PSU_RCTs are connected in parallel in a server node as 1+1 redundancy. PSU Current Multiplier, or PSU_CM (PSU step-down part, or step-down conversion unit), contains a DC-DC step-down conversion unit. Each PSU_CM is pluggable as a component and can be maintained separately when a function problem occurs. Figure 7 As shown, PSU_CMO to PSU_CM5 are multiple DC-DC step-down conversion units connected in parallel, providing sufficient power support and reliability guarantee for the node; Figure 7 The number of DC-DC buck conversion units shown in the figure is for illustration only, and the actual number is selected according to the power of a single conversion unit and the power required by the system.

[0099] The DC-DC step-down conversion unit outputs the system power DC_System as the power stage input to power the back-end processor. Each power stage output is connected to the Power Remapping Module (PRM for short), i.e., the power redistribution module, to redistribute the power stage-power rail.

[0100] Figure 8 Schematic diagram of power switching of a PRM power redistribution module according to an optional embodiment of the present application, such as Figure 8 As shown, PowerStage1x is a dedicated power stage for Power Rail 1, PowerStage2x is a dedicated power stage for Power Rail 2, and similarly, PowerStage3x and PowerStage4x are dedicated power stages for Power Rail 3 and Power Rail 4. PowerStageCx is a common power stage, which means it can be connected to multiple power rails, but does not supply power to multiple power rails at the same time. Figure 8 Q1 to Q8 are switching switches. The power redistribution module can control the common power stage to be connected to a certain power rail by controlling the on or off of the switches according to the power demand of each power rail of the processor, so as to realize the redistribution of the power stage output.

[0101] Fig. 9 is a schematic diagram of the control logic relationship of the PRM power redistribution module according to an optional embodiment of the present application, such as Fig. 9 As shown, Controller represents a multi-power controller. The controller monitors the current of each power stage through the signal Isense connected to the power stage, and obtains the power stage output current parameters in real time by reading the Isense voltage; controls the upper and lower MOS tube switches in the power stage through the PWM signal, and controls the MOS tube switching frequency and output voltage in the power stage by controlling the PWM frequency and duty cycle; and controls the power stage to work or not through the En signal. The controller also obtains the processor-side power rail voltage through the signal Vsense connected to the processor to achieve real-time voltage control and compensation. At the same time, the processor informs the PRM of the current requirements of each power rail through the SW Control signal connected to the PRM power redistribution module. The PRM controls which switches should be turned on and which switches should be turned off according to the current requirements of each power rail. The PRM is also interconnected with the Controller through the Vcontrol signal. When the common power stage switches to different power supplies, it informs the controller of the voltage amplitude that the common power stage should output and adjusts it in real time.

[0102] Fig.10 is a schematic diagram of the layout of key electronic components according to an optional embodiment of the present application, such as Fig.10 As shown, Fig.10This is a cross-sectional view of the server motherboard. From top to bottom, the top is the heat sink 1; below the heat sink 1 is the core processor, i.e., CPU, GPU and other chips; Controller represents multiple power controllers; Motherboard represents the board where the processor is located; the bottom of the motherboard is connected to the PCB1 board through BGA ball planting; PCB1 has a PRM module, resistors, capacitors and connector 2; below PCB1 is PCB2, PCB2 has inductors, resistors, capacitors and connector 1; below PCB2 is PCB3, PCB3 has power chips, i.e., power stages, resistors and capacitors; below PCB3 is the heat sink 2.

[0103] Regarding the power flow, the motherboard draws power DC_System from the power supply unit, connects to PCB1 through the BGA pad, connects to PCB2 through connector 2, and connects to PCB3 through connector 1. The power stage output on PCB3 is connected to the inductor, PCB3 is connected to the inductor through welding, and connected to PCB2 through the inductor output. That is, it is converted into the required voltage for each power rail, and then each inductor output is connected to PCB1 through connector 2. After the power stage output is redistributed through the PRM module, it is connected to the motherboard through the BGA pad on the back of the motherboard to power the processor.

[0104] Regarding the direction of logic signals, Isense, PWM, and EN are connected to PCB1 through BGA pads, connected to PCB2 through connector 2, and connected to each power stage on PCB3 through connector 1; the Vsense signal is fed back from the processor end to the controller end through the motherboard; the Vcontrol signal is connected from PCB1 to the controller end through the BGA pad; and the SW control is connected from PCB1 to the processor end through the BGA pad.

[0105] Regarding the processing method, PCB3 is processed first, and the power stage and resistors and capacitors are soldered by reflow soldering. After the components on one side of the power stage are fixed with glue, the inductor and connector 1 on the back of PCB3 are soldered. After the inductor and connectors are soldered, they are fixed with glue again. After the components on PCB3 are soldered, the resistors and capacitors on PCB2 are soldered. After the soldering is completed, PCB2 is soldered and PCB3 is connected to PCB2 by reflow soldering, and then glue is used to fix it. PCB1 is soldered with PRM modules, resistors, capacitors, and connector 2 by reflow soldering. After the soldering is completed, it is fixed with glue. After fixing, PCB1 is soldered to PCB2 and PCB3 together. Finally, PCB1, PCB2, and PCB3 are soldered to the back of the mainboard as a whole through BGA and connected to the mainboard.

[0106] Regarding the heat dissipation method, according to the power supply method of the optional embodiment of the present application, the main heat-generating parts are the core processor and the power stage part, and the heat is removed by cold plate liquid cooling or air cooling through the heat dissipation device 1 below the processor and the heat dissipation device 2 below the power stage.

[0107] Table 1 is a schematic diagram of the maximum power supply of the system under different architectures and failure modes. For the server power supply unit, as shown in Table 1, assuming that a PSU (power supply unit) can provide a maximum power of X, and a standard server cabinet can accommodate 6 PSUs, the PSU RCT+PSU CM power supply architecture is compared with the existing power supply N+1 redundant and N+N redundant power supply architectures. In the case of one AC power failure, one PSU failure, or one AC power and one PSU, the maximum power that the system can supply is better than the N+N redundant architecture, which is the same as the N+1 redundant architecture.

[0108] While ensuring that the reliability of power supply to the data center servers remains roughly the same, the number of rectifier units has been changed from one per device, which must be installed or unplugged at the same time for maintenance, to two independent units. In high-power scenarios, the number of ATS (automatic transfer switch) units does not have to increase with the number of DC-DC units, which reduces the number of power supply components, reduces the cost of power supply unit power components, and increases the upper limit of system power density. As a step-down conversion unit, the PSU_CM unit can increase the number of parallel connections according to the node power and can be flexibly matched.

[0109] Table 1

[0110]

[0111] At the same time, comparing the number of input power lines, the number of input power lines in the N+N redundant architecture is 6; the number of input power lines in the N+1 redundant architecture is 12; the number of input power lines in the PSU RCT+PSU CM architecture is 4. Compared with the N+N and N+1 redundant architectures, the optional embodiment of the present application greatly reduces the number of input AC power lines, greatly reduces the difficulty of cabinet power line plugging and wiring, and provides convenience for node installation and later maintenance.

[0112] Compared with the situation in which a single PSU in an N+1 redundant architecture is large in size and weight, and a single PSU is difficult to put on the shelf, the single PSU RCT or PSU CM monomer after being split in the optional embodiment of the present application is small in size and light in weight, which is convenient for operation and maintenance personnel to operate, and provides convenience for the later operation and maintenance of the data center.

[0113] In terms of processor power supply, compared with the side power supply in the related technology, the PRM module newly added in the optional embodiment of the present application can switch the common power level between different power rails. While ensuring that the processor power supply remains unchanged, the total number of power levels, power chips, and inductors used to power all power rails of a single processor are reduced, and the power supply part occupies less space in the entire node, which is convenient for improving the power density of the server node and reducing the cost of motherboard components.

[0114] Compared with side power supply, the optional embodiment of the present application places the power stage, inductor, and PRM module on the back of the processor, which greatly reduces the length of the processor power supply path, reduces the impact of parasitic parameters on the processor power supply, and improves the dynamic response performance of multiple power supplies. At the same time, the shorter power supply path reduces the loss of processor power supply on the PCB, reduces heat generation, improves efficiency, and brings benefits to node heat dissipation.

[0115] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0116] In this embodiment, a power distribution device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0117] Fig.11 is a schematic diagram of a power distribution circuit according to an embodiment of the present application, such as Fig.11 As shown, the circuit includes: at least two AC power supplies 112, at least two rectifier units 114 connected in series with each AC power supply, a buck conversion unit 116 connected in series with the at least two rectifier units, and a power redistribution module 118 connected in series with the buck conversion unit, wherein:

[0118] The at least two rectifying units 114 are used to rectify and convert the at least two AC power supplies 112 to output a first DC power supply, wherein each rectifying unit includes: an automatic conversion unit and a power factor correction unit, and the number of rectifying units is consistent with the number of AC currents;

[0119] The at least one step-down conversion unit 116 is configured to receive the first DC power supply and adjust a voltage corresponding to the first DC power supply to generate a second DC power supply;

[0120] The power redistribution module 118 is used to distribute the second DC power to a plurality of power rails, wherein the plurality of power rails are used to supply power to a target processor.

[0121] Through the power distribution circuit of the present application, at least two rectifier units are used to rectify and convert at least two AC power supplies that are consistent with the number of rectifier units to output a first DC power supply, wherein each rectifier unit includes an automatic conversion unit and a power factor correction unit; the first DC unit is received through at least one buck conversion unit, and the voltage corresponding to the first DC power supply is adjusted to generate a second DC power supply; the second DC power supply is distributed to multiple power rails for powering the target processor through a power redistribution module. That is, after the embodiment of the present application rectifies and converts the AC power supply into a first DC power supply through a rectifier unit including an automatic conversion unit and a power factor correction unit, the first DC power supply is converted into a second DC power supply through a buck conversion unit, and then the second DC power supply is distributed to multiple power rails through a power redistribution module. That is, the number of automatic conversion units in the embodiment of the present application does not need to increase with the number of buck conversion units, which reduces the number of power supply power elements, reduces the cost of power elements of the power supply unit, and increases the upper limit of the system power density improvement. It solves the problem of low power density of the redundant power supply architecture caused by the integration of the rectifier unit and the buck unit in the N+N or N+1 redundant power supply architecture in the related art. This improves the system power density.

[0122] In an exemplary embodiment, the at least two rectifier units 114 are also used to determine whether there is a failed AC power supply in the at least two AC power supplies; when it is determined that the first AC power supply in the at least two AC power supplies is a failed AC power supply, the connection between the first AC power supply and the first rectifier unit is disconnected by the automatic conversion unit, and the connection between other AC power supplies and other rectifier units is maintained by the automatic conversion unit, wherein the at least two AC power supplies include: the first AC power supply and the other AC power supplies, and the at least two rectifier units include: the first rectifier unit and the other rectifier units, and the first AC power supply and the first rectifier unit are in a series relationship; the other AC power supplies are rectified and converted by the other rectifier units to output the first DC power supply; when it is determined that the at least two AC power supplies are both valid AC power supplies, the connection between the at least two AC power supplies and the at least two rectifier units is maintained by the automatic conversion unit, and the at least two AC power supplies are rectified and converted by the at least two rectifier units to output the first DC power supply.

[0123] In an exemplary embodiment, the at least two rectifier units 114 are also used to rectify the at least two AC power supplies into pulsating DC power through the at least two rectifier units; smooth the pulsating DC power through a filter capacitor, and adjust the DC voltage corresponding to the pulsating DC power through a voltage stabilizing circuit; adjust the waveform corresponding to the pulsating DC power after smoothing and voltage stabilization to generate the first DC power supply.

[0124] In an exemplary embodiment, the at least one buck conversion unit is further used to receive a target signal sent by the target processor, and determine the real-time power requirement corresponding to the target processor through the target signal; and adjust the current and voltage of the first DC power supply through the DC-DC converter in the at least one buck conversion unit and the real-time power requirement to generate the second DC power supply.

[0125] In an exemplary embodiment, the power redistribution module 118 is further used to detect current demand information of each power rail through the power redistribution module; control the conduction state between each common power level and each power rail according to the current demand information and multiple switching switches in the power redistribution module to distribute the second DC power supply to the multiple power rails, wherein the at least one step-down conversion unit and the multiple power rails are connected through at least one power level, and the at least one power level includes: each dedicated power level corresponding to each power rail and a common power level corresponding to the multiple power rails.

[0126] In an exemplary embodiment, the power redistribution module 118 is further used to obtain power supply capability information of each dedicated power level corresponding to each power rail; determine a first current required by each power rail according to the current demand information, and determine a second current allowed to be provided by each dedicated power level for each power rail according to the power supply capability information; determine a magnitude relationship between the first current and the second current; and when the magnitude relationship indicates that the first current required by any power rail is greater than the second current corresponding to any dedicated power level, close a target switching switch among the multiple switching switches to adjust the conduction state between the common power level corresponding to any power rail and any power rail to a connected state, wherein any power rail is connected in series with any dedicated power level, and the target switching switch is used to control the on-off state of any power rail and the common power level corresponding to any power rail.

[0127] In an exemplary embodiment, the distribution circuit of the power supply further includes: a fault detection unit, which is used to monitor the target state parameters corresponding to each buck conversion unit and obtain the preset state parameters corresponding to the at least one buck conversion unit; compare each target state parameter with the preset state parameter respectively, and determine the working state corresponding to each buck conversion unit according to the comparison result; when the comparison result indicates that the preset state parameter contains each target state parameter, determine that the working state corresponding to the at least one buck conversion unit is a normal state; when the comparison result indicates that the preset state parameter does not contain the target state parameter corresponding to the first buck conversion unit of the at least one buck conversion unit, determine that the working state corresponding to the first buck conversion unit is an abnormal state; send the working state corresponding to the first buck conversion unit to the target object, so that the target object removes the first buck conversion unit from the slot corresponding to the first buck conversion unit, and repairs the fault of the first buck conversion unit.

[0128] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps in the above method embodiment when running.

[0129] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0130] Optionally, in this embodiment, the storage medium may be configured to store program codes for executing the following steps:

[0131] S1, rectifying and converting at least two AC power supplies through at least two rectifying units to output a first DC power supply, wherein each rectifying unit includes: an automatic conversion unit and a power factor correction unit, and the number of rectifying units is consistent with the number of AC currents;

[0132] S2, receiving the first DC power supply through at least one step-down conversion unit, and adjusting a voltage corresponding to the first DC power supply to generate a second DC power supply;

[0133] S3, distributing the second DC power supply to a plurality of power rails through a power redistribution module, wherein the plurality of power rails are used to supply power to a target processor.

[0134] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in the above method embodiment.

[0135] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0136] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0137] S1, rectifying and converting at least two AC power supplies through at least two rectifying units to output a first DC power supply, wherein each rectifying unit includes: an automatic conversion unit and a power factor correction unit, and the number of rectifying units is consistent with the number of AC currents;

[0138] S2, receiving the first DC power supply through at least one step-down conversion unit, and adjusting a voltage corresponding to the first DC power supply to generate a second DC power supply;

[0139] S3, distributing the second DC power supply to a plurality of power rails through a power redistribution module, wherein the plurality of power rails are used to supply power to a target processor.

[0140] An embodiment of the present application further provides a computer program product, including a computer program, which is implemented to perform the steps in the above method embodiment when executed by a processor.

[0141] Optionally, in this embodiment, the computer program product may be executed by a processor to perform the following steps:

[0142] S1, rectifying and converting at least two AC power supplies through at least two rectifying units to output a first DC power supply, wherein each rectifying unit includes: an automatic conversion unit and a power factor correction unit, and the number of rectifying units is consistent with the number of AC currents;

[0143] S2, receiving the first DC power supply through at least one step-down conversion unit, and adjusting a voltage corresponding to the first DC power supply to generate a second DC power supply;

[0144] S3, distributing the second DC power supply to a plurality of power rails through a power redistribution module, wherein the plurality of power rails are used to supply power to a target processor.

[0145] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0146] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0147] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for distributing power, characterized in that: include: At least two AC power supplies are rectified and converted by at least two rectifying units to output a first DC power supply, wherein each rectifying unit includes: an automatic conversion unit and a power factor correction unit, and the number of rectifying units is consistent with the number of AC currents; receiving the first DC power supply through at least one step-down conversion unit, and adjusting a voltage corresponding to the first DC power supply to generate a second DC power supply; The second DC power supply is distributed to a plurality of power rails through a power redistribution module, wherein the plurality of power rails are used to supply power to a target processor.

2. The method according to claim 1, characterized in that At least two rectifier units are used to rectify and convert at least two AC power supplies to output a first DC power supply, including: Determining whether there is a failed AC power source in the at least two AC power sources; In the case where it is determined that the first AC power supply of the at least two AC power supplies is a failed AC power supply, the connection between the first AC power supply and the first rectifier unit is disconnected by the automatic conversion unit, and the connection between other AC power supplies and other rectifier units is maintained by the automatic conversion unit, wherein the at least two AC power supplies include: the first AC power supply and the other AC power supplies, the at least two rectifier units include: the first rectifier unit and the other rectifier unit, and the first AC power supply and the first rectifier unit are in series relationship; Rectifying and converting the other AC power source through the other rectifying unit to output the first DC power source; When it is determined that the at least two AC power sources are both valid AC power sources, the connection between the at least two AC power sources and the at least two rectifier units is maintained by the automatic conversion unit, and the at least two AC power sources are rectified and converted by the at least two rectifier units to output the first DC power source.

3. The method according to claim 1, characterized in that At least two rectifier units are used to rectify and convert at least two AC power supplies to output a first DC power supply, including: Rectifying the at least two AC power supplies into pulsating DC power by the at least two rectifying units; The pulsating direct current is smoothed by a filter capacitor, and the direct current voltage corresponding to the pulsating direct current is adjusted by a voltage stabilizing circuit; The waveform corresponding to the pulsating direct current after smoothing and voltage stabilization is adjusted to generate the first direct current power supply.

4. The method according to claim 1, characterized in that: Adjusting a voltage corresponding to the first DC power supply to generate a second DC power supply includes: receiving a target signal sent by the target processor, and determining a real-time power demand corresponding to the target processor through the target signal; The current and voltage of the first DC power source are adjusted by a DC-DC converter in the at least one step-down conversion unit and the real-time power demand to generate the second DC power source.

5. The method according to claim 1, characterized in that The second DC power supply is distributed to a plurality of power rails through a power redistribution module, including: Detecting current demand information of each power rail through the power redistribution module; The conduction state between each common power stage and each power rail is controlled according to the current demand information and a plurality of switching switches in the power redistribution module to distribute the second DC power supply to the plurality of power rails, wherein the at least one step-down conversion unit is connected to the plurality of power rails via at least one power stage, and the at least one power stage comprises: each dedicated power stage corresponding to each power rail and a common power stage corresponding to the plurality of power rails.

6. The method according to claim 5, characterized in that Controlling the conduction state between each common power stage and each power rail according to the current demand information and a plurality of switches in the power redistribution module to distribute the second DC power to the plurality of power rails, including: Obtaining power supply capability information of each dedicated power level corresponding to each power rail; Determine a first current required by each power rail according to the current demand information, and determine a second current allowed to be provided by each dedicated power stage for each power rail according to the power supply capability information; determining a magnitude relationship between the first current and the second current; In a case where the magnitude relationship indicates that a first current required by any power rail is greater than a second current corresponding to any dedicated power level, a target switching switch among the multiple switching switches is closed to adjust the conduction state between the common power level corresponding to any power rail and the any power rail to a connected state, wherein the any power rail and the any dedicated power level are connected in series, and the target switching switch is used to control the on-off state of the any power rail and the common power level corresponding to the any power rail.

7. The method according to claim 1, characterized in that After distributing the second DC power to a plurality of power rails by the power redistribution module, the method further includes: Monitoring the target state parameter corresponding to each buck conversion unit, and obtaining the preset state parameter corresponding to the at least one buck conversion unit; Comparing each target state parameter with the preset state parameter respectively, and determining the working state corresponding to each buck conversion unit according to the comparison result; When the comparison result indicates that the preset state parameter includes each of the target state parameters, determining that the working states corresponding to the at least one buck conversion unit are all normal states; When the comparison result indicates that the preset state parameter does not include the target state parameter corresponding to the first buck conversion unit in the at least one buck conversion unit, determining that the working state corresponding to the first buck conversion unit is an abnormal state; The working state corresponding to the first buck conversion unit is sent to a target object, so that the target object removes the first buck conversion unit from a slot corresponding to the first buck conversion unit and performs fault repair on the first buck conversion unit.

8. A power distribution circuit, It is characterized in that include: At least two AC power supplies, at least two rectifier units connected in series with each AC power supply, a buck converter unit connected in series with the at least two rectifier units, and a power redistribution module connected in series with the buck converter unit, wherein: The at least two rectifying units are used to rectify and convert the at least two AC power supplies to output a first DC power supply, wherein each rectifying unit includes: an automatic conversion unit and a power factor correction unit, and the number of rectifying units is consistent with the number of AC currents; The at least one step-down conversion unit is configured to receive the first DC power supply and adjust a voltage corresponding to the first DC power supply to generate a second DC power supply; The power redistribution module is used to distribute the second DC power supply to a plurality of power rails, wherein the plurality of power rails are used to supply power to a target processor.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method according to any one of claims 1 to 7 when executed by a processor.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

Images