Power supply method, device and equipment

By setting up a power management system for detecting the GPU status in the GPU server, powering the motherboard and GPU respectively, the waste of power resources caused by the difference in power requirements of GPU and other hardware devices is solved, and the GPU power supply efficiency is improved.

CN114123457BActive Publication Date: 2024-11-22XINHUASAN INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111452282.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-01
Publication Date
2024-11-22
Estimated Expiration
2041-12-01

AI Technical Summary

Technical Problem

In existing GPU servers, the use of boost devices reduces the utilization of power resources provided to GPUs due to the different rated voltages of power supply required by GPUs and other hardware devices.

Method used

The first power supply device and the second power supply device are arranged in the network device. By detecting whether the GPU is in place, powering the motherboard and the GPU respectively, avoiding the use of a boost device to directly output power resources that meet the GPU needs.

Benefits of technology

It improves the utilization rate of GPU power supply resources, avoids unnecessary power resource consumption, and enhances power supply efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114123457B_ABST
    Figure CN114123457B_ABST
Patent Text Reader

Abstract

The embodiments of the present application disclose a power supply method, device and equipment. The present application sets a first power supply device for outputting a first power resource and a second power supply device for outputting a second power resource in a network device, so that when it is necessary to power the hardware on the device, the first power supply device is used to power the mainboard, and when it is detected that the graphics processor GPU on the device is in place, the second power supply device is used to power the GPU, thereby forming two power supply circuits without overlap. In this embodiment, the second power supply device that powers the GPU can directly output a second power resource that meets the power supply requirements of the GPU, so there is no need to use a boost device to convert the power resources supplied to the GPU, thereby improving the utilization rate of the power resources supplied to the GPU.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of communications, and in particular to a power supply method, device and equipment. Background Art

[0002] As the performance of GPU (Graphics Processing Unit) improves, the voltage of power resources required to power GPU also increases. In existing GPU servers, GPU is powered by inputting power resources with a fixed voltage into the motherboard, which then inputs the power resources into the GPU. There are other hardware devices on the motherboard, such as BIOS chip, CPU chip, memory card, etc. The CPU also needs to input power resources to other hardware devices, and the rated voltage of the power required by other hardware devices is different from the rated voltage of the power required by the GPU.

[0003] For example, Figure 1 As shown in the figure, the voltage of the power resource generally input to the motherboard is 12V, but the rated voltage of the power required by the GPU is 48V or 54V, and the rated voltage of the power required by other hardware devices is 5V or 3.3V. In order to power the GPU, a boost device needs to be added between the motherboard and the GPU to increase the voltage of the power resource output by the motherboard from 12V to 48V or 54V through the boost device. The 48V or 54V unit resource obtained after the boost can be input to the GPU. However, when the above-mentioned boost device is in operation, it needs to consume the power resources supplied to the GPU, which will reduce the utilization rate of the power resources supplied to the GPU. Summary of the invention

[0004] The present application discloses a power supply method, device and equipment to improve the utilization rate of power resources supplied to a GPU.

[0005] According to a first aspect of an embodiment of the present application, a power supply method is provided, which is applied to a network device, wherein the network device is composed of at least a first power supply device, a second power supply device, a mainboard, and a power adapter board, wherein the first power supply device is connected to the mainboard, and the second power supply device is connected to the power adapter board, and the method includes:

[0006] Controlling the first power supply device to input the first power resource to the mainboard; wherein the first power supply device is used to output the first power resource, and the voltage of the first power resource is a first specified voltage;

[0007] Detect whether the graphics processor GPU on this device is in place. If so, control the second power supply device to input the second power resource to the power adapter board, and input the second power resource to the GPU through the power adapter board; wherein the second power supply device is used to output the second power resource, the voltage of the second power resource is a second specified voltage, the second specified voltage is the rated voltage of the power supply required by the GPU, and the first specified voltage and the second specified voltage have different values.

[0008] Optionally, the network device further comprises a hardware switch and a voltage conversion device, the second power supply device is connected to the mainboard via the hardware switch, and the hardware switch is connected to the mainboard via the voltage conversion device;

[0009] The second power supply device is connected to the power adapter board via the hardware switch.

[0010] Optionally, controlling the second power supply device to input the second power supply resource to the power adapter board includes:

[0011] The second power supply device is controlled by the hardware switch to input the second power resource to the power adapter board.

[0012] Optionally, if it is detected that the GPU on the device is not in place, the method further includes:

[0013] The second power supply device is controlled by the hardware switch to input the second power supply resource into the voltage conversion device, the second specified voltage corresponding to the second power supply resource is converted into the first specified voltage by the voltage conversion device, and the converted second power supply resource output by the voltage conversion device is input into the mainboard.

[0014] Optionally, the step of detecting whether a graphics processor GPU on the device is in place includes:

[0015] When the network device is turned on, the flag bit pre-set for the GPU is accessed through the basic input and output system BIOS on the motherboard. If the value of the flag bit is a first value, it is determined that the GPU is in place; if the value of the flag bit is a second value, it is determined that the GPU is not in place.

[0016] Optionally, controlling the second power supply device to input the second power supply resource to the power adapter board includes:

[0017] When it is detected that the GPU is in place, the BIOS is triggered to issue a power resource control instruction to instruct the second power supply device to input the second power resource to the power adapter board.

[0018] According to a second aspect of an embodiment of the present application, a network device is provided, the network device comprising at least a first power supply device, a second power supply device, a mainboard and a power adapter board, the first power supply device is connected to the mainboard, and the second power supply device is connected to the power adapter board;

[0019] The first power supply device is used to output a first power resource, and the voltage of the first power resource is a first specified voltage;

[0020] The second power supply device is used to output a second power resource, wherein the voltage of the second power resource is a second specified voltage; wherein the second specified voltage is a rated voltage required to power the GPU, and the first specified voltage and the second specified voltage have different values;

[0021] The mainboard is used to receive the first power resource output by the first power supply device;

[0022] The power adapter board is used to input the second power resource output by the second power supply device to the GPU when it is detected that the graphics processor GPU on the device is in place.

[0023] Optionally, the network device further comprises a hardware switch and a voltage conversion device, the second power supply device is connected to the mainboard via the hardware switch, and the hardware switch is connected to the mainboard via the voltage conversion device;

[0024] The second power supply device is connected to the power adapter board via the hardware switch.

[0025] Optionally, the hardware switch is used to control the second power supply device to input the second power resource to the power adapter board when it is detected that the GPU on the device is in place, and to control the second power supply device to input the second power resource to the voltage conversion device when it is detected that the GPU on the device is not in place;

[0026] The voltage conversion device is used to convert the second specified voltage corresponding to the second power resource into a first specified voltage, and input the converted second power resource to the mainboard.

[0027] According to a third aspect of an embodiment of the present application, a power supply device is provided, wherein the network device is composed of at least a first power supply device, a second power supply device, a mainboard, and a power adapter board, wherein the first power supply device is connected to the mainboard, and the second power supply device is connected to the power adapter board, and the device includes:

[0028] A mainboard power supply unit, used for controlling the first power supply device to input the first power resource to the mainboard; wherein the first power supply device is used for outputting the first power resource, and the voltage of the first power resource is a first specified voltage;

[0029] The GPU power supply unit is used to control the second power supply device to input the second power supply resource to the power adapter board when detecting that the graphics processor GPU on this device is in place, and input the second power supply resource to the GPU through the power adapter board; wherein the second power supply device is used to output the second power supply resource, the voltage of the second power supply resource is a second specified voltage, the second specified voltage is the rated voltage of the power supply required by the GPU, and the first specified voltage and the second specified voltage have different values.

[0030] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0031] It can be seen from the above technical solutions that the solution provided by the present application is to set a first power supply device for outputting a first power resource and a second power supply device for outputting a second power resource in a network device, so that when it is necessary to power the hardware on the device, the first power supply device is used to power the mainboard, and when it is detected that the graphics processor GPU on the device is in place, the second power supply device is used to power the GPU, thereby forming two power supply circuits without overlap. In this embodiment, the second power supply device that powers the GPU can directly output a second power resource that meets the power supply requirements of the GPU, so there is no need to use a boost device to convert the power resources supplied to the GPU, thereby improving the utilization rate of the power resources supplied to the GPU.

[0032] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the specification and, together with the description, serve to explain the principles of the specification.

[0034] Figure 1 It is a hardware power supply structure diagram in a server provided by the related technology;

[0035] Figure 2 is a flow chart of a power supply method provided in an embodiment of the present application;

[0036] Figure 3 is an example diagram of a power supply structure provided in an embodiment of the present application;

[0037] Figure 4 is a flow chart of another power supply method provided in an embodiment of the present application;

[0038] Figure 5 is an example diagram of another power supply structure provided in an embodiment of the present application;

[0039] Figure 6 It is a schematic diagram of the structure of the power supply device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0040] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0041] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0042] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0043] In order to enable those skilled in the art to better understand the technical solutions provided by the embodiments of the present application and to make the above-mentioned purposes, features and advantages of the embodiments of the present application more obvious and understandable, the technical solutions in the embodiments of the present application are further described in detail below in conjunction with the accompanying drawings.

[0044] See also Figure 2 , Figure 2 A flowchart of a power supply method provided in an embodiment of the present application. As an embodiment, Figure 2 The process shown can be applied to network devices, such as GPU servers, etc. The power supply structure provided in this embodiment is as follows Figure 3As shown, the network device is composed of at least a first power supply device 31, a second power supply device 32, a mainboard 33 and a power adapter board 34, wherein the first power supply device 31 is connected to the mainboard 33. In this embodiment, the first power supply device 31 can supply power to other hardware devices 35 on the device through the voltage reduction device 37 on the mainboard 33, and the second power supply device 32 is connected to the power adapter board 34. The second power supply device 32 can supply power to the GPU 36 on the mainboard through the power adapter board 34 outside the mainboard. Among them, the voltage of the first power resource output by the first power supply device 31 is P1, the voltage of the second power resource output by the second power supply device 32 is P2, and the voltage of the power resource input to other hardware devices 35 after the first power resource is converted by the voltage reduction device 37 is P3. In specific applications, the value of P1 can be 12V, the value of P2 can be the rated voltage 48V / 54V of the power required by the GPU, and the value of P3 can be the rated voltage 5V / 3.3V of the power required by other hardware devices.

[0045] Exemplarily, the voltage of the first power resource may be 12V, and the voltage of the second power resource may be 48V or 54V, and the other hardware devices in this embodiment may be hardware devices other than the GPU device on the mainboard of the server, such as BIOS, CPU, memory, array card, network card and other devices. It should be noted that before providing the first power resource to other hardware devices, the first power resource needs to be further converted, such as further converting the first power resource through the voltage reduction device 37. The conversion process can refer to the relevant technology and will not be described here.

[0046] Optionally, in this embodiment, the first power supply device includes at least one first power supply unit PSU (Powersupply unit), and the second power supply device includes at least one second PSU. Optionally, if the first power supply device includes multiple first PSUs, the multiple first PSUs can be connected in series or in parallel. Similarly, if the second power supply device includes multiple second PSUs, the multiple second PSUs can also be connected in series or in parallel.

[0047] like Figure 2 As shown, the process of the power supply method provided in this embodiment may include the following steps:

[0048] Step 201, controlling a first power supply device to input a first power resource to a mainboard.

[0049] As an embodiment, in order to prevent the mainboard and other hardware devices on the mainboard from being damaged due to a short circuit caused by damage to some hardware devices on the mainboard when the mainboard is directly connected to the normal power supply current after the network device is turned on, the mainboard will start the protection mechanism when the network device is turned on, and only connect the micro-current that can start the hardware logic CPLD (Complex Programmable Logic Device) pre-set in the mainboard for self-test, and input the micro-current to each hardware device on the mainboard. The hardware devices on the mainboard here can refer to the CPU chip, memory card, network card, etc. installed on the mainboard.

[0050] After the hardware logic CPLD is started, the hardware logic CPLD can start to send detection messages to various hardware devices on the mainboard. When receiving the response of the hardware device, it can be determined that the responding hardware device is in place and has not been damaged, and then the flag bit corresponding to the responding hardware device recorded in the register on the mainboard is updated. Among them, the flag bit corresponding to each hardware device recorded in the register is used to identify whether the hardware device is in place.

[0051] After the hardware logic CPID traverses and detects various hardware devices on the mainboard, if it is determined that the CPU chip, memory chip, BIOS chip and other hardware devices required to form a basic system on the mainboard are in place, the mainboard can be controlled to turn on the circuit that connects to the normal power supply current, that is, the first power supply device is controlled to input the first power supply resource into the mainboard.

[0052] In an embodiment of the present application, after receiving the first power resource, the mainboard can start to supply power to other hardware devices installed on the mainboard, such as the CPU, memory, network card, array card, etc., except for the GPU. The power supply process will also supply power to the CPU chip on the mainboard and the BIOS chip carrying the BIOS program, so that the BIOS on the mainboard can be started. After the BIOS starts running, the BIOS self-test program will be executed. The self-test program is used to detect whether the hardware devices on the mainboard (such as PCIE (peripheral component interconnect express, high-speed serial computer expansion bus standard) devices) are in place, and whether the hardware devices in place can work normally. The specific detection process of the BIOS firmware self-test can refer to the relevant technology and will not be described in detail here.

[0053] It should be noted that the first power supply device in the above embodiment is used to output a first power supply resource, and the voltage of the first power supply resource is a first specified voltage. The first specified voltage here can refer to the voltage of the power supply resource input to the mainboard in the related technology. For example, the first specified voltage can be 12V, and this application does not limit this.

[0054] Step 202, detect whether the graphics processor GPU on the device is in place, if so, control the second power supply device to input the second power resource to the power adapter board, and input the second power resource to the GPU through the power adapter board.

[0055] In an embodiment of the present application, the above-mentioned second power supply device is used to output a second power resource, the voltage of the second power resource is a second specified voltage, and the second specified voltage is the rated voltage of the power supply required by the GPU in this device, and in the embodiment of the present application, the values ​​of the first specified voltage and the second specified voltage are different.

[0056] As an embodiment, after the mainboard is powered through the above-mentioned step 201 and the BIOS self-test program is started, since the GPU in this embodiment is also a PCIE device, through the self-test of the BIOS firmware, it can be known whether the GPU on the device is in place and whether the GPU can work normally.

[0057] Optionally, in this embodiment, when the BIOS detects whether the GPU on the device is in place and whether it can work normally, since the presence of the GPU on the motherboard is detected based on the hardware logic CPID, and the flag bit corresponding to the GPU recorded in the register is updated, the flag bit corresponding to the GPU recorded in the register can be accessed by the BIOS in this embodiment to determine whether the GPU is in place. In this embodiment of the application, if the value of the flag bit accessed is the first value, it can be determined that the GPU corresponding to the flag bit is in place, and if the value of the flag bit is the second value, it can be determined that the GPU corresponding to the flag bit is not in place.

[0058] Exemplarily, the GPU flag is usually one bit. When the bit value is 1 (ie, the first value is 1), it indicates that the GPU in the server is in place. When the bit value is 0 (ie, the second value is 0), it indicates that the CPU in the server is not in place.

[0059] Further, as an embodiment, when the GPU in the device is detected to be in place, the BIOS can be triggered to issue a power resource control instruction through the newly added program in the BIOS in this embodiment, so as to instruct the GPU to turn on the circuit connected to the second power resource through the power resource control instruction, so that the second power supply device inputs the second power resource to the power adapter board. In this embodiment, the power resource control instruction can be implemented through the IPMI (Intelligent Platform Management Interface) protocol.

[0060] So far, completed Figure 2 The process shown.

[0061] pass Figure 2 As can be seen from the process shown, in this embodiment, a first power supply device for outputting a first power resource and a second power supply device for outputting a second power resource are set in the network device, so that when it is necessary to power the hardware on the device, the first power supply device is used to power the mainboard, and when it is detected that the graphics processor GPU on the device is in place, the second power supply device is used to power the GPU, thereby forming two power supply circuits without overlap. In this embodiment, the second power supply device that powers the GPU can directly output a second power resource that meets the power supply requirements of the GPU, so there is no need to use a boost device to convert the power resources supplied to the GPU, thereby improving the utilization rate of the power resources supplied to the GPU.

[0062] The above examples are only for ease of understanding and are not specifically limited to the embodiments of the present application.

[0063] Combine the following Figure 4 and Figure 5 Another embodiment of a power supply method provided by the present application is described. The power supply structure provided in this embodiment is as follows: Figure 5 As shown, the network device is composed of at least a first power supply device 51, a second power supply device 52, a hardware switch 53, a voltage conversion device 54, a mainboard 55 and a power adapter board 56. Among them, the first power supply device 51 is connected to the mainboard 55. In this embodiment, the first power supply device 51 can supply power to other hardware devices 58 on the device through the voltage reduction device 59 on the mainboard 55; the second power supply device 52 is connected to the voltage conversion device 54 through the hardware switch 53, and the second power supply device 52 can supply power to the GPU 57 on the mainboard through the hardware switch 53 outside the mainboard and the power adapter board 56; the second power supply device 52 is also connected to the mainboard 55 through the hardware switch 53 and the voltage conversion device 54, and the second power supply device 52 can supply power to the mainboard 55 on the device through the hardware switch 53 and the voltage conversion device 54, and then the mainboard 55 supplies power to other hardware devices 58 on the mainboard. Among them, the voltage of the first power resource output by the first power supply device 51 is P1, the voltage of the second power resource output by the second power supply device 52 is P2, and the voltage of the first power resource input to the power resource of other hardware devices 58 after conversion by the step-down device 59 is P3. In specific applications, the value of P1 can be 12V, the value of P2 can be the rated voltage of the power supply required by the GPU, 48V / 54V, and the value of P3 can be the rated voltage of the power supply required by other hardware devices, 5V / 3.3V.

[0064] like Figure 4 As shown, the process of the power supply method provided in this embodiment may include the following steps:

[0065] Step 401, controlling the first power supply device to input the first power resource to the mainboard.

[0066] In this embodiment, controlling the first power supply device to input the first power resource to the mainboard can refer to the above Figure 2 The description of the illustrated method embodiment will not be repeated here.

[0067] Step 402, detect whether the graphics processor GPU on the device is in place, if so, execute step 403, otherwise execute step 404.

[0068] Step 403: Control the second power supply device to input the second power resource to the power adapter board through the hardware switch.

[0069] In this embodiment, a hardware switch is added between the second power supply device and the power adapter board. When it is detected that the graphics processor GPU on the device is in place, the BIOS on the device can control the hardware switch to connect to the power adapter board and disconnect the connection with the power conversion device, thereby realizing the input of the above-mentioned second power resource to the power adapter board.

[0070] Step 404, controlling the second power supply device through the hardware switch to input the second power supply resource into the voltage conversion device, using the voltage conversion device to convert the second specified voltage corresponding to the second power supply resource into the first specified voltage, and inputting the converted second power supply resource output by the voltage conversion device into the mainboard.

[0071] In this embodiment, in order to make full use of the power resources in the second power supply device, when it is detected that the graphics processor GPU on the device is not in place, the BIOS on the device can control the hardware switch to connect with the power conversion device and disconnect the connection with the power adapter board, thereby realizing the input of the above-mentioned second power resource into the power conversion device, and then the voltage conversion device converts the second specified voltage corresponding to the second power resource into the first specified voltage.

[0072] It should be noted that the above-mentioned control hardware switch is connected to the power conversion device or the power adapter board, and can be implemented through a power resource control instruction issued by the BIOS. The power resource control instruction can be an IPMI instruction that complies with the IPMI protocol.

[0073] So far, completed Figure 4 The process shown.

[0074] pass Figure 4As can be seen from the process shown, in this embodiment, a first power supply device for outputting a first power resource and a second power supply device for outputting a second power resource are set in the network device, so that when it is necessary to power the hardware on the device, the first power supply device is used to power the mainboard, and when it is detected that the graphics processor GPU on the device is in place, the second power supply device is used to power the GPU, thereby forming two power supply circuits without overlap. In this embodiment, the power supply to the GPU does not need to pass through the mainboard, and the second power supply device can directly output the second power resource that meets the power supply requirements of the GPU, so there is no need to use a boost device to convert the power resources supplied to the GPU, thereby improving the utilization rate of the power resources supplied to the GPU.

[0075] In addition, this embodiment adds a hardware switch and a power conversion device. When the GPU is not in place, the second power resource of the second power supply device is fully utilized to power the mainboard. Since the GPU on the network device is generally in place during actual application, the power resources consumed by the power conversion device are less than the power resources consumed by the boost device in the related technology. Therefore, this embodiment also improves the utilization rate of the power resources supplied to the GPU compared with the related technology.

[0076] The above describes the method provided in the embodiment of the present application. The following describes the device provided in the embodiment of the present application:

[0077] See also Figure 6 , Figure 6 A schematic diagram of a power supply method provided in an embodiment of the present application, the device is applied to a network device, the network device is composed of at least a first power supply device, a second power supply device, a mainboard and a power adapter board, the first power supply device is connected to the mainboard, the second power supply device is connected to the power adapter board, and the device includes:

[0078] The mainboard power supply unit 601 is used to control the first power supply device to input the first power resource to the mainboard; wherein the first power supply device is used to output the first power resource, and the voltage of the first power resource is a first specified voltage;

[0079] The GPU power supply unit 602 is used to control the second power supply device to input the second power supply resource to the power adapter board when it is detected that the graphics processor GPU on this device is in place, and input the second power supply resource to the GPU through the power adapter board; wherein the second power supply device is used to output the second power supply resource, the voltage of the second power supply resource is a second specified voltage, the second specified voltage is the rated voltage of the power supply required by the GPU, and the first specified voltage and the second specified voltage have different values.

[0080] Optionally, the network device further comprises a hardware switch and a voltage conversion device, the second power supply device is connected to the mainboard via the hardware switch, and the hardware switch is connected to the mainboard via the voltage conversion device;

[0081] The second power supply device is connected to the power adapter board and includes:

[0082] The second power supply device is connected to the power adapter board via the hardware switch.

[0083] Optionally, the GPU power supply unit 602 controls the second power supply device to input the second power resource to the power adapter board, including:

[0084] The second power supply device is controlled by the hardware switch to input the second power resource to the power adapter board.

[0085] Optionally, if it is detected that the GPU on the device is not in place, the method further includes:

[0086] The second power supply device is controlled by the hardware switch to input the second power supply resource into the voltage conversion device, the second specified voltage corresponding to the second power supply resource is converted into the first specified voltage by the voltage conversion device, and the converted second power supply resource output by the voltage conversion device is input into the mainboard.

[0087] Optionally, the device further includes: a GPU detection unit, used to detect whether a graphics processor GPU on the device is in place.

[0088] Optionally, the GPU detection unit is specifically used to access a flag bit pre-set for the GPU through the basic input and output system BIOS on the motherboard when the network device is turned on. If the value of the flag bit is a first value, it is determined that the GPU is in place; if the value of the flag bit is a second value, it is determined that the GPU is not in place.

[0089] Optionally, controlling the second power supply device to input the second power supply resource to the power adapter board includes:

[0090] When it is detected that the GPU is in place, the BIOS is triggered to issue a power resource control instruction to instruct the second power supply device to input the second power resource to the power adapter board.

[0091] Optionally, the first power supply device includes at least one first power supply unit PSU, and the second power supply device includes at least one second power supply unit PSU.

[0092] So far, completed Figure 6 A block diagram of an embodiment of the device shown.

[0093] Correspondingly, an embodiment of the present application further provides a server, the server comprising at least: a network device at least consisting of a first power supply device, a second power supply device, a mainboard and a power adapter board, the first power supply device is connected to the mainboard, and the second power supply device is connected to the power adapter board;

[0094] The first power supply device is used to output a first power resource, and the voltage of the first power resource is a first specified voltage;

[0095] The second power supply device is used to output a second power resource, wherein the voltage of the second power resource is a second specified voltage; wherein the second specified voltage is a rated voltage required to power the GPU, and the first specified voltage and the second specified voltage have different values;

[0096] The mainboard is used to receive the first power resource output by the first power supply device;

[0097] The power adapter board is used to input the second power resource output by the second power supply device to the GPU when it is detected that the graphics processor GPU on the device is in place.

[0098] Optionally, the network device further comprises a hardware switch and a voltage conversion device, the second power supply device is connected to the mainboard via the hardware switch, and the hardware switch is connected to the mainboard via the voltage conversion device;

[0099] The second power supply device is connected to the power adapter board via the hardware switch.

[0100] Optionally, the hardware switch is used to control the second power supply device to input the second power resource to the power adapter board when it is detected that the GPU on the device is in place, and to control the second power supply device to input the second power resource to the voltage conversion device when it is detected that the GPU on the device is not in place;

[0101] The voltage conversion device is used to convert the second specified voltage corresponding to the second power resource into a first specified voltage, and input the converted second power resource to the mainboard.

[0102] It should be noted that the functions of the various hardware in the server provided in the embodiment of the present application can refer to the description of the embodiment of the power supply method described above, and the connection relationship between the various hardware in the server can be as follows: Figure 3 or Figure 5 One of them, I will not go into details here.

[0103] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A power supply method, characterized in that: The method is applied to a network device, the network device at least comprising a first power supply device, a second power supply device, a mainboard and a power adapter board, the first power supply device being connected to the mainboard, the second power supply device being connected to the power adapter board, the method comprising: When the network device is turned on, the hardware logic CPLD pre-set in the mainboard sends a detection message to the graphics processor GPU on the device, and upon receiving a response from the GPU, updates a flag bit corresponding to the GPU recorded in a register on the mainboard; the flag bit is used to identify whether the GPU is in place; Controlling the first power supply device to input the first power resource to the mainboard; wherein the first power supply device is used to output the first power resource, and the voltage of the first power resource is a first specified voltage; Accessing a flag bit corresponding to the GPU recorded in the register through a basic input / output system BIOS on the motherboard, and if the value of the flag bit is a first value, determining that the GPU corresponding to the flag bit is in place; if the value of the flag bit is a second value, determining that the GPU corresponding to the flag bit is not in place; If it is determined that the GPU is in place, the BIOS is triggered to issue a power resource control instruction to instruct the second power supply device to input the second power resource into the power adapter board, and input the second power resource into the GPU through the power adapter board; wherein the second power supply device is used to output the second power resource, the voltage of the second power resource is a second specified voltage, the second specified voltage is the rated voltage of the power supply required by the GPU, and the first specified voltage and the second specified voltage have different values.

2. The method according to claim 1, characterized in that: The network device is further composed of a hardware switch and a voltage conversion device, wherein the second power supply device is connected to the mainboard via the hardware switch, and the hardware switch is connected to the mainboard via the voltage conversion device; The second power supply device is connected to the power adapter board via the hardware switch.

3. The method according to claim 2, characterized in that If it is detected that the GPU on the device is not in place, the method further includes: The second power supply device is controlled by the hardware switch to input the second power supply resource into the voltage conversion device, the second specified voltage corresponding to the second power supply resource is converted into the first specified voltage by the voltage conversion device, and the converted second power supply resource output by the voltage conversion device is input into the mainboard.

4. A network device, characterized in that: The network device is composed of at least a first power supply device, a second power supply device, a mainboard and a power adapter board, wherein the first power supply device is connected to the mainboard, and the second power supply device is connected to the power adapter board; The first power supply device is used to output a first power resource, and the voltage of the first power resource is a first specified voltage; The second power supply device is used to output a second power resource, the voltage of the second power resource is a second specified voltage; wherein the second specified voltage is a rated voltage required to power a graphics processor GPU, and the first specified voltage and the second specified voltage have different values; The mainboard is used to receive the first power resource output by the first power supply device; The power adapter board is used to input the second power resource output by the second power supply device to the GPU when it is detected that the graphics processor GPU on the device is in place; wherein, the second power resource is input to the power adapter board by the second power supply device after receiving the power resource control instruction issued by the basic input and output system BIOS on the mainboard; the detecting that the graphics processor GPU on the device is in place includes: when the network device is turned on, the hardware logic CPLD pre-set in the mainboard sends a detection message to the graphics processor GPU on the device, and when receiving the response of the GPU, updates the flag bit corresponding to the GPU recorded in the register on the mainboard; the flag bit is used to identify whether the GPU is in place; access the flag bit corresponding to the GPU recorded in the register through the basic input and output system BIOS on the mainboard, if the value of the flag bit is a first value, it is determined that the GPU corresponding to the flag bit is in place, and if the value of the flag bit is a second value, it is determined that the GPU corresponding to the flag bit is not in place.

5. The device according to claim 4, characterized in that The network device is further composed of a hardware switch and a voltage conversion device, wherein the second power supply device is connected to the mainboard via the hardware switch, and the hardware switch is connected to the mainboard via the voltage conversion device; The second power supply device is connected to the power adapter board via the hardware switch.

6. The device according to claim 5, characterized in that The hardware switch is used to control the second power supply device to input the second power supply resource to the voltage conversion device when it is detected that the GPU on the device is not in place; The voltage conversion device is used to convert the second specified voltage corresponding to the second power resource into a first specified voltage, and input the converted second power resource to the mainboard.

7. A power supply device, characterized in that: The device is applied to a network device, which is composed of at least a first power supply device, a second power supply device, a mainboard and a power adapter board, wherein the first power supply device is connected to the mainboard, the second power supply device is connected to the power adapter board, and the device comprises: A mainboard power supply unit, used for, when the network device is turned on, the hardware logic CPLD pre-set in the mainboard sends a detection message to the graphics processor GPU on the device, and upon receiving a response from the GPU, updates a flag bit corresponding to the GPU recorded in a register on the mainboard; the flag bit is used to identify whether the GPU is in place; controls the first power supply device to input the first power resource to the mainboard; wherein the first power supply device is used to output the first power resource, and the voltage of the first power resource is a first specified voltage; The GPU power supply unit is used to access the flag bit corresponding to the GPU recorded in the register through the basic input and output system BIOS on the mainboard. If the value of the flag bit is a first value, it is determined that the GPU corresponding to the flag bit is in place; if the value of the flag bit is a second value, it is determined that the GPU corresponding to the flag bit is not in place; when it is detected that the graphics processor GPU on this device is in place, the basic input and output system BIOS on the mainboard is triggered to issue a power resource control instruction to instruct the second power supply device to input the second power resource to the power adapter board, and input the second power resource to the GPU through the power adapter board; wherein the second power supply device is used to output the second power resource, the voltage of the second power resource is a second specified voltage, the second specified voltage is the rated voltage of the power supply required by the GPU, and the first specified voltage and the second specified voltage have different values.