Power supply method for network interface card, server, non-volatile readable storage medium, and electronic device

Abstract

Embodiments of the present application relate to the field of computers, and provide a power supply method for a network interface card, a server, a non-volatile readable storage medium, and an electronic device. The power supply method is applied to a server, and the server comprises: an electronic fuse, a power supply connector, and a network interface card. The power supply method for the network interface card comprises: determining a current working mode of the network interface card, wherein the working mode of the network interface card includes a network interface card mode and a data processing unit mode; when the current working mode is the network interface card mode, controlling an enable pin of the electronic fuse to be turned on when the server is in a running state; and when the current working mode is the data processing unit mode, controlling the enable pin of the electronic fuse to be turned on when the server is in a shutdown state.

Description

Network interface card power supply methods, servers, non-volatile readable storage media, and electronic devices.

[0001] Cross-references to related applications

[0002] This application claims priority to Chinese Patent Application No. 202411528275.2, filed on October 30, 2024, entitled "Power Supply Method for Network Card and Server, Storage Medium and Electronic Device", the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application relates to the field of computers, and in particular to a power supply method for a network interface card (NIC), as well as a server, a non-volatile readable storage medium, and an electronic device. Background Technology

[0004] With the widespread application of cloud computing and big data technologies in recent years, the network load of data centers has been increasing significantly. This has led to the gradual replacement of traditional standard network cards by smart network interface cards (NICs), which are now increasingly used in servers due to their higher performance, greater flexibility, and more powerful functions. Smart NICs typically operate in two modes: Network Interface Card (NIC) mode and Data Processing Unit (DPU) mode. The requirements for the server vary depending on the mode in which the smart NIC operates. In traditional server designs, switching the mode of a smart NIC requires updating the NIC's firmware (FW) software, followed by modifications and adjustments to the server to adapt it to the smart NIC. For example, the power supply port of the NIC might be changed from the core power supply (P12V_Core) port to the auxiliary power supply (P12V_AUX) port, thus expanding the working interface of the smart NIC. This series of tasks requires disassembling the server and modifying the hardware, adding or removing components. This results in a large workload when switching modes of the smart network card, making the switching efficiency of the smart network card low.

[0005] There is currently no effective solution to the problem that the server cannot adapt to different network card modes online during the network card switching process, resulting in low efficiency in network card switching. Summary of the Invention

[0006] This application provides a power supply method for a network interface card (NIC), a server, a non-volatile readable storage medium, and an electronic device, to at least solve the problem of low efficiency in NIC mode switching due to the server's inability to adapt to different NIC modes online during NIC mode switching.

[0007] According to a first aspect of the embodiments of this application, a power supply method for a network interface card (NIC) is provided, applied to a server. The server includes: an electronic fuse, a power connector, and the NIC; one end of the electronic fuse is electrically connected to a power supply unit, and the other end of the electronic fuse is electrically connected to the power connector. When the enable pin of the electronic fuse is turned on, the electronic fuse allows current to flow, and the power connector allows the NIC to be supplied with core power or auxiliary power. The method includes: determining the current operating mode of the NIC, wherein the operating mode of the NIC includes: a network interface card mode and a data processing unit mode; when the current operating mode is the network interface card mode, controlling the enable pin of the electronic fuse to turn on when the server is running; and when the current operating mode is the data processing unit mode, controlling the enable pin of the electronic fuse to turn on when the server is powered off.

[0008] In an exemplary embodiment, when the current operating mode is data processing unit mode, if the server is powered on, the enable pin of the electronic fuse is in the open state.

[0009] In an exemplary embodiment, the server further includes a controller and a programmable device; the method further includes: determining the current operating mode of the network interface card through the controller and sending the current operating mode to the programmable device; if the programmable device determines that the current operating mode is network interface card mode, controlling the enable pin of the electronic fuse to be turned on when the server is in a running state through the programmable device; if the programmable device determines that the current operating mode is data processing unit mode, controlling the enable pin of the electronic fuse to be turned on when the server is in a powered-off state through the programmable device.

[0010] In one exemplary embodiment, an integrated circuit interconnect bus is provided between the programmable device and the controller, and sending the current operating mode to the programmable device includes: sending the current operating mode to the programmable device via the integrated circuit interconnect bus.

[0011] In one exemplary embodiment, the server further includes: a field-replaceable unit, which determines the current operating mode of the network interface card (NIC) through a controller, including: after the server is powered on, the controller completes initialization, and the controller reads the field-replaceable unit to determine whether the current operating mode of the NIC is network interface card mode or data processing unit mode.

[0012] In one exemplary embodiment, the method further includes: when the controller is about to disconnect the power supply to the network card, controlling the enable pin of the electronic fuse to turn off via the controller.

[0013] In an exemplary embodiment, the server further includes a fan; before determining the operating mode of the network card, the method further includes: obtaining a first preset rotation speed and controlling the fan to rotate at the first preset rotation speed; after determining the operating mode of the network card, the method further includes: determining a network card heat dissipation strategy corresponding to the operating mode of the network card to obtain a target network card heat dissipation strategy; determining the rotation speed of the fan according to the target network card heat dissipation strategy and controlling the fan to rotate at the rotation speed.

[0014] In an exemplary embodiment, the server further includes: a controller and a programmable device; wherein the programmable device is configured to control the fan to rotate at a first preset speed, and the controller is configured to control the fan to rotate at a rotational speed; the method further includes: sending a heartbeat signal to the programmable device at each preset time interval by the controller; after determining the working mode of the network card, obtaining a second preset speed by the programmable device without monitoring the heartbeat signal sent by the controller, and controlling the fan to rotate at the second preset speed.

[0015] In an exemplary embodiment, the server further includes: a controller; a network interface card (NIC) having a first network port, and a controller having a second network port, the first network port and the second network port being connected via a network cable; the method further includes: when the NIC is in data processing unit mode, obtaining data sent by the controller or sending data to the controller through the NIC based on the network communication connection between the NIC and the controller.

[0016] In an exemplary embodiment, the server further includes: a programmable device; a data line for transmitting a reset signal is provided between the network interface card (NIC) and the programmable device, and a data line for transmitting a reset signal is provided between the programmable device and the controller; the method further includes: performing fault diagnosis on the controller based on the operating data obtained by the NIC through the network communication connection between the NIC and the controller; and, if a controller fault is determined based on the operating data, sending a reset signal to the programmable device through the NIC and instructing the programmable device to forward the reset signal to the controller, wherein the reset signal is used to reset the controller.

[0017] In an exemplary embodiment, the server further includes: a controller and a programmable device; an integrated circuit interconnect bus and / or a universal asynchronous transceiver bus are provided between the network interface card and the programmable device, and an integrated circuit interconnect bus and / or a universal asynchronous transceiver bus are provided between the programmable device and the controller; the method further includes: sending data to the network interface card via the controller based on the integrated circuit interconnect bus and / or the universal asynchronous transceiver bus, or receiving data sent by the network interface card.

[0018] In an exemplary embodiment, the server further includes a controller; the method further includes: when the current operating mode of the network card is to be switched to a target operating mode, updating the firmware file in the network card to the firmware file corresponding to the target operating mode through the controller, wherein the operating mode of the network card includes the target operating mode; and controlling the enable pin of the electronic fuse to turn off through the controller.

[0019] In an exemplary embodiment, after updating the firmware file in the network card to the firmware file corresponding to the target working mode via the controller, the method further includes: disconnecting the power supply to the network card via the controller; and reconnecting the network card via the controller to switch the current working mode of the network card to the target working mode.

[0020] In an exemplary embodiment, the server further includes a programmable device and a field-replaceable unit; the method further includes: when the current operating mode of the network interface card (NIC) is to be switched to a target operating mode, updating the code in the field-replaceable unit to the code corresponding to the target operating mode via a controller; when the firmware file in the NIC and the code in the field-replaceable unit are updated, sending indication information to the programmable device via the controller, wherein the indication information is used to indicate that the operating mode of the NIC of the programmable device has been switched.

[0021] In an exemplary embodiment, the method further includes: after the firmware file in the network card and the code in the field replaceable unit have been updated, controlling the server to power on or restart while powered on, and switching the current operating mode of the network card to the target operating mode.

[0022] In an exemplary embodiment, after determining the current operating mode of the network interface card (NIC), the method further includes: determining a server resource allocation strategy corresponding to the current operating mode; and allocating internal resources of the server based on the server resource allocation strategy.

[0023] According to a second aspect of the embodiments of this application, a server is also provided, including: a management module, an electronic fuse, a power supply connector, and a network interface card (NIC); wherein one end of the electronic fuse is electrically connected to a power supply unit, and the other end of the electronic fuse is electrically connected to the power supply connector; when the enable pin of the electronic fuse is turned on, the electronic fuse allows current to flow, and the power supply connector allows core power or auxiliary power to be provided to the NIC; the management module is configured to determine the current operating mode of the NIC, wherein the operating mode of the NIC includes: a network interface card mode and a data processing unit mode; when the current operating mode is determined to be the network interface card mode, the management module controls the enable pin of the electronic fuse to be turned on when the server is in a running state; when the current operating mode is determined to be the data processing unit mode, the management module controls the enable pin of the electronic fuse to be turned on when the server is in a powered-off state.

[0024] According to a third aspect of the embodiments of this application, a non-volatile computer-readable storage medium is also provided, wherein a computer program is stored in the non-volatile computer-readable storage medium, and the computer program is configured to perform the steps in any of the above method embodiments when it is run.

[0025] According to a fourth aspect of the embodiments of this application, an electronic device is also provided, 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 perform the steps in any of the above method embodiments.

[0026] According to a fifth aspect of the embodiments of this application, a computer program product is also provided, the computer program product including a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0027] This application determines the current operating mode of the network interface card (NIC). When the NIC is in network interface card mode, the enable pin of the electronic fuse is turned on when the server is running. When the NIC is in data processing unit mode, the enable pin of the electronic fuse is turned on when the server is off. This allows the server to meet the power supply requirements of the NIC in different modes online, avoiding the need for disassembly and debugging of server hardware during NIC mode switching. This improves the efficiency of NIC mode switching and solves the problem of low efficiency caused by the server's inability to adapt to different NIC modes online during NIC mode switching. Attached Figure Description

[0028] The accompanying drawings, which are included to provide an understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0029] Figure 1 is a hardware structure block diagram of a server device according to an embodiment of the present application of a power supply method for a network card;

[0030] Figure 2 is a schematic diagram of an optional system architecture for implementing this application according to an embodiment of this application;

[0031] Figure 3 is a schematic diagram of the overall architecture of a power supply system for a network card according to an embodiment of this application;

[0032] Figure 4 is a schematic diagram of an optional power supply switching scheme according to an embodiment of this application;

[0033] Figure 5 is a schematic diagram of an optional fan speed control according to an embodiment of this application;

[0034] Figure 6 is a schematic diagram of signal transmission of an optional sideband signal control circuit according to an embodiment of this application;

[0035] Figure 7 is a structural block diagram of a server according to an embodiment of this application;

[0036] Figure 8 is a schematic diagram of the structure of an optional electronic device according to an embodiment of this application. Detailed Implementation

[0037] The embodiments of this application will be described in detail below with reference to the accompanying drawings and examples.

[0038] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0039] The network card power supply method embodiments provided in this application can be executed in server devices or similar computing devices. Taking a server device as an example, FIG1 is a hardware structure block diagram of a server device for a network card power supply method according to an embodiment of this application. As shown in FIG1, the server device may include one or more (only one is shown in FIG1) processors 102 (processor 102 may include, but is not limited to, microprocessors MCU or programmable logic devices FPGA, etc.) and a memory 104 configured to store data. The server device may also include a transmission device 106 configured for communication functions and an input / output device 108. Those skilled in the art will understand that the structure shown in FIG1 is only illustrative and does not limit the structure of the server device. For example, the server device may also include more or fewer components than shown in FIG1, or have a different configuration than shown in FIG1.

[0040] The memory 104 may be configured to store computer programs, such as application software programs and modules, like the computer program corresponding to the power supply method of the network card in this embodiment. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may include memory remotely located relative to the processor 102, and these remote memories can be connected to server devices via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0041] The transmission device 106 is configured to receive or transmit data via a network. Optional examples of the network may include a wireless network provided by the communication provider of the server device. In one example, the transmission device 106 may connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a radio frequency (RF) module configured to communicate wirelessly with the Internet.

[0042] To address the aforementioned problems in related technologies, this embodiment provides a power supply method for a network interface card (NIC), applied to a server. The server includes: an electronic fuse, a power connector, and a NIC. One end of the electronic fuse is electrically connected to a power supply unit, and the other end is electrically connected to the power connector. When the enable pin of the electronic fuse is turned on, the electronic fuse allows current to flow, and the power connector allows the NIC to receive core power or auxiliary power, as shown in Figure 2. This process includes the following steps S202 to S206:

[0043] In some embodiments, as shown in Figure 4, the power connector can provide both core power (P12V_Core) and auxiliary power (P12V_AUX) to the network card. When the network card is in different modes, the power connector in the server can provide the power required by different modes. Thus, when the network card switches modes, there is no need to debug the server hardware, improving the efficiency of network card mode switching.

[0044] Step S202: Determine the current operating mode of the network card, wherein the operating mode of the network card includes: network interface card mode and data processing unit mode;

[0045] It should be noted that when the network card is in network interface card mode, it is mainly responsible for ordinary network transmission tasks, and provides the server as a slave device of high-speed peripheral component interconnect express (PCIE). At this time, the server usually only needs to provide power supply, heat dissipation and basic in-band and out-of-band management functions in the running (S0) state.

[0046] When the network interface card (NIC) is in data processing unit mode, it needs to perform more complex network, storage, and computing tasks, running independently of the server's main processor (Central Processing Unit, or CPU). In this case, the NIC is an independent system with its own computing unit, implementing functions such as network, storage, and computing offloading, taking over services such as network virtualization and hardware resource pooling, operating independently of the server, and even managing the server as a host device. This frees up the server's general-purpose CPU resources, allowing the server to focus more on processing high-level computing tasks.

[0047] Step S204: When the current working mode is network interface card mode, the enable pin of the electronic fuse is turned on when the server is running.

[0048] In some embodiments, when the current operating mode is network interface card mode, the enable pin of the control electronic fuse is turned on when the server is in the running state (S0), and the server provides core power.

[0049] In some embodiments, when the current operating mode is network interface card mode, the enable pin of the control electronic fuse is turned off when the server is in a power-off state (S5).

[0050] Step S206: When the current working mode is data processing unit mode, the enable pin of the electronic fuse is turned on when the server is powered off.

[0051] In some embodiments, when the current operating mode is data processing unit mode, the enable pin of the control electronic fuse is turned on when the server is powered off, and the server provides auxiliary power.

[0052] It should be noted that when the current working mode is data processing unit mode, if the server is powered on, the enable pin of the electronic fuse is also in the open state.

[0053] It should be noted that Efuse is used as a controllable power switch, controlling whether power flows to the smart network card through its internal logic state.

[0054] The enable pin (EN) is a control pin on an electronic fuse. When the EN pin is set to a high level (usually indicating an "on" or "enabled" state), the electronic fuse allows current to flow, and the server can provide power to the network card. When the EN pin is set to a low level (usually indicating a "off" or "disabled" state), the electronic fuse cuts off the power, preventing current from flowing to the network card, and the server cannot provide power to the network card.

[0055] It should be noted that steps S204 and S206 are executed under different circumstances, and there is no specific order of execution.

[0056] The above steps determine the current operating mode of the network card. When the network card is in network interface card mode, the enable pin of the electronic fuse is turned on when the server is running. When the network card is in data processing unit mode, the enable pin of the electronic fuse is turned on when the server is off. This allows the server to meet the power supply requirements of the network card in different modes online, avoiding the need to disassemble and debug the server hardware during network card mode switching. This improves the efficiency of network card mode switching and solves the problem of low efficiency caused by the server's inability to adapt to different network card modes online during network card mode switching.

[0057] In one exemplary embodiment, the server further includes: a controller and a programmable device; the method further includes the steps S11 to S13:

[0058] In some embodiments, the controller is a Baseboard Manager Controller (BMC), and the programmable device is a Complex Programmable Logic Device (CPLD) and / or a Field Programmable Gate Array (FPGA).

[0059] Step S11: Determine the current operating mode of the network card through the controller and send the current operating mode to the programmable device;

[0060] In some embodiments, after the device (including the server) is powered on, the controller completes initialization, determines whether the current network card is in network interface card mode or data processing unit mode by reading the field replaceable unit (FRU) of the whole machine, and sends it to the programmable device through the inter-integrated circuit bus (I2C bus).

[0061] Step S12: When the programmable device determines that the current working mode is network interface card mode, the enable pin of the electronic fuse is turned on by controlling the programmable device to turn on when the server is running.

[0062] Step S13: When the programmable device determines that the current working mode is the data processing unit mode, the enable pin of the electronic fuse is turned on by controlling the programmable device to turn on when the server is in the power-off state.

[0063] In some embodiments, when the programmable device determines that the current operating mode of the network interface card is network interface card mode, it will turn on the enable pin of the control electronic fuse when the server is running; when it determines that the current operating mode of the network interface card is data processing unit mode, it will turn on the enable pin of the control electronic fuse when the server is powered off.

[0064] Steps S11 to S13 above, by designing a power supply circuit that supplies power according to the different mode requirements of the network interface card (NIC), ensure that the server maintains a stable power supply even when the server is powered off in data processing unit mode. This effectively supports the independent operation and high-efficiency services of the NIC in data processing unit mode, such as network virtualization and hardware resource pooling, while avoiding the impact on the normal function of the NIC during server restarts or maintenance. This automated mode recognition and power switching mechanism reduces manual intervention, simplifies server management and the mode switching process of the smart NIC, and improves the reliability and flexibility of the system.

[0065] In one exemplary embodiment, the method further includes the step of: when the controller is about to disconnect the power to the network card, controlling the enable pin of the electronic fuse to turn off via the controller.

[0066] It should be noted that by controlling the enable pin of the electronic fuse to turn off when the network card's power is to be disconnected, the controller can effectively protect the network card from power abnormalities or system failures. For example, during a server restart, disconnecting the power can prevent data loss or hardware damage, ensuring the security of the smart network card. Furthermore, when the network card is undergoing software firmware upgrades or mode switching, controlling the enable pin of the electronic fuse to turn off and then back on ensures that the new firmware file or new operating mode is correctly loaded and initialized, avoiding upgrade failures or unsuccessful mode switching due to power instability, thus enhancing the overall reliability and stability of the system.

[0067] In one exemplary embodiment, the server further includes a fan; before determining the operating mode of the network card, the method further includes the steps of: obtaining a first preset speed and controlling the fan to rotate at the first preset speed;

[0068] In some embodiments, as shown in FIG5, before determining the working mode of the network card, the programmable device enables the fan by default and controls the fan to rotate at 30% of the maximum speed (i.e., the first preset speed).

[0069] It should be noted that after the alternating current (AC) is supplied, the device enters the working state and each device component begins initialization. Since the programmable device is hardware, it completes initialization before the controller. Therefore, before the controller determines the working mode of the network card, in order to ensure that the network card will not overheat and crash due to working in the data processing unit mode, the programmable device enables the fan by default, thereby ensuring the safety and stability of the system.

[0070] In an exemplary embodiment, after determining the operating mode of the network card, the method further includes the following steps: determining a network card heat dissipation strategy corresponding to the operating mode of the network card to obtain a target network card heat dissipation strategy; determining the fan rotation speed according to the target network card heat dissipation strategy, and controlling the fan to rotate at the rotation speed.

[0071] In some embodiments, based on the network interface card's (NIC) operating mode, the controller obtains the corresponding target NIC heat dissipation strategy. This target NIC heat dissipation strategy guides the operation of the fan and the adjustment of other heat dissipation mechanisms to ensure that the temperature of the NIC and the entire server remains within a safe range. The target heat dissipation strategy includes, but is not limited to: limiting the minimum and / or maximum fan speed, pre-setting the server system temperature, and prioritizing the heat dissipation strategy.

[0072] By following the steps above, a balance between heat dissipation efficiency and energy consumption can be ensured.

[0073] In an exemplary embodiment, the programmable device is configured to control the fan to rotate at a first preset speed, and the controller is configured to control the fan to rotate at a rotational speed; the method further includes the following steps S21-S22:

[0074] Step S21: The controller sends a heartbeat signal to the programmable device at each preset time interval;

[0075] In some embodiments, as shown in Figure 5, after the controller completes initialization, it immediately sends a heartbeat signal to the programmable device. Subsequently, it sends a heartbeat signal to the programmable device at each preset time interval. After sending the first heartbeat signal, the controller starts to work normally, reads the information of the field replaceable unit through the integrated circuit interconnect bus, determines the current network card's working mode, and then calls the corresponding network card heat dissipation algorithm to control the fan to rotate at the rotation speed required by the corresponding network card heat dissipation algorithm, and continuously monitors the network card's temperature, thereby controlling the fan to dissipate heat from the machine according to the heat dissipation algorithm.

[0076] Step S22: After determining the working mode of the network card, the second preset speed is obtained through the programmable device without monitoring the heartbeat signal sent by the controller, and the fan is controlled to rotate at the second preset speed.

[0077] In some embodiments, during operation, the programmable device constantly monitors the heartbeat signal of the controller. When the programmable device fails to monitor the heartbeat signal sent by the controller, it indicates that the controller has malfunctioned (hung up). The programmable device will acquire a second preset speed (e.g., 80% of the maximum fan speed) to prevent the system from overheating and crashing until the heartbeat signal sent by the controller is monitored again.

[0078] It should be noted that, through the above steps, when the controller malfunctions or malfunctions and fails to send a heartbeat signal, the programmable device will automatically take over the thermal control, controlling the fan to dissipate heat according to the preset second speed. This mechanism ensures that even in the event of controller failure, the system can maintain sufficient heat dissipation capacity, preventing equipment damage or system crashes due to overheating, significantly enhancing the stability and reliability of the server. Furthermore, by setting different preset speeds, heat dissipation can be adjusted according to the network card's operating mode and the current system's cooling requirements, optimizing fan speed control, avoiding unnecessary power consumption and noise, while ensuring sufficient cooling effect, achieving a balance between heat dissipation efficiency and energy utilization efficiency.

[0079] In an exemplary embodiment, the network interface card (NIC) has a first network port, and the controller has a second network port, with the first network port and the second network port connected via a network cable. The method further includes the following steps: when the NIC is in data processing unit mode, obtaining data sent by the controller or sending data to the controller through the NIC based on the network communication connection between the NIC and the controller.

[0080] In some embodiments, as shown in Figure 6, the RJ45 management network of the network card is connected to the management network port of the controller via a network cable. When the network card is in data processing unit mode, it is used for the network card to manage the server.

[0081] In some embodiments, as shown in Figure 6, when the network interface card (NIC) is in data processing unit mode, it can function as an independent system, receiving or sending data through a network connection between its first network port and the controller's second network port. This allows data exchange between the NIC and the controller to be achieved without additional hardware interfaces or dedicated lines. This enables the NIC to acquire or send system management information even when the server is powered off or not fully booted, thereby receiving remote debugging commands, performing log capture, performance monitoring, and other tasks, enhancing the server's remote management and monitoring capabilities. Furthermore, by exchanging data via network communication, maintenance personnel can remotely monitor and manage the server without physical access. This is particularly important in data center environments, significantly reducing on-site operations by maintenance personnel and improving operational efficiency and response speed.

[0082] In an exemplary embodiment, a data line for transmitting a reset signal is provided between the network card and the programmable device, and a data line for transmitting a reset signal is provided between the programmable device and the controller; the method further includes the following steps S31-S32:

[0083] Step S31: Using the network card, and with the network communication connection between the network card and the controller, obtain the operating data sent by the controller, and perform fault diagnosis on the controller based on the operating data;

[0084] Step S32: If a controller fault is determined based on the operating data, a reset signal is sent to the programmable device via the network card, and the programmable device is instructed to forward the reset signal to the controller. The reset signal is used to reset the controller.

[0085] In some embodiments, as shown in FIG6, the network card sends a reset signal (Reset) to the programmable device via the data line, and finally to the controller, so that when the network card is in the data processing unit mode, the network card can act as the master device to reset the controller, thereby resetting the controller.

[0086] It should be noted that by following the above steps, when the server controller malfunctions or needs to be restarted, the network card can directly reset the controller via the Reset signal without physical power failure or manual intervention, thereby speeding up fault recovery and improving system reliability.

[0087] In an exemplary embodiment, the server further includes: a controller and a programmable device; an integrated circuit interconnect bus and / or a universal asynchronous transceiver bus are provided between the network card and the programmable device, and an integrated circuit interconnect bus and / or a universal asynchronous transceiver bus are provided between the programmable device and the controller; the method further includes the steps of: sending data to the network card through the controller based on the integrated circuit interconnect bus and / or the universal asynchronous transceiver bus, or receiving data sent by the network card.

[0088] In some embodiments, the controller sends the obtained operating mode of the network card to the programmable device via the integrated circuit interconnect bus.

[0089] In some embodiments, as shown in FIG6, the network card is first connected to the programmable device via an integrated circuit interconnect bus, and then connected to the controller via the integrated circuit interconnect bus.

[0090] In some embodiments, as shown in Figure 6, when the network card is in data processing unit mode, it requires independent debugging and log capture capabilities. This is achieved by connecting a Universal Asynchronous Receiver / Transmitter (UART) to the controller for signal transmission, enabling multiplexing (MUX) functionality. This allows the controller to switch between the system serial port and the network card's serial port. Specifically, when out-of-band upgrades or debugging are needed, the controller can switch the serial port to the network card (typically, servers only have system serial ports and BMC serial ports, and cannot access the smart network card's serial port).

[0091] In one exemplary embodiment, the server further includes a controller; the method further includes the following steps S41-S42:

[0092] Step S41: When the current working mode of the network card is to be switched to the target working mode, the firmware file in the network card is updated to the firmware file corresponding to the target working mode through the controller. The working mode of the network card includes the target working mode.

[0093] In some embodiments, after the network card completes the firmware (FW) update, the controller will disconnect the network card's power supply, and the new operating mode of the network card will take effect after power is restored.

[0094] Step S42: Control the enable pin of the electronic fuse to close via the controller.

[0095] It should be noted that by controlling the enable pin of the electronic fuse to be turned off by the controller, the network card can be restarted, allowing it to complete the mode switch. Simultaneously, this ensures circuit safety during the network card's operating mode switch, preventing unexpected situations such as excessive current or short circuits during the update process. As a controllable protective device, the electronic fuse can cut off the circuit when necessary, protecting the equipment from damage.

[0096] It should be noted that by updating the firmware file in the network card through the above steps, the network card can switch from its current operating mode to the target operating mode. This means that the network card can flexibly adjust its operating mode according to actual needs to adapt to different application scenarios or optimize performance. The controller is responsible for updating the firmware file in the network card to the firmware corresponding to the target operating mode, ensuring the stability and efficiency of the network card and improving the performance and stability of the entire system.

[0097] In an exemplary embodiment, the server further includes: a controller, a programmable device, and a field-replaceable unit; the method further includes the following steps S51-S52:

[0098] Step S51: When the current working mode of the network card is to be switched to the target working mode, the code in the field replaceable unit is updated to the code corresponding to the target working mode through the controller;

[0099] In some embodiments, the controller refreshes the field replaceable unit of the server, changing the current model code to the code of the corresponding network card mode. After completing the firmware file refresh of the network card and the field replaceable unit refresh, the controller will notify the programmable device of the network card mode change, requiring the power supply and heat dissipation control logic to follow the new working mode of the new network card.

[0100] In some embodiments, after the server is restarted or restarted while powered on, the network card completes the mode switch and runs the new service;

[0101] Step S52: After the firmware file in the network card and the code in the field replaceable unit have been updated, the controller sends an indication message to the programmable device, wherein the indication message is used to indicate that the operating mode of the network card of the programmable device has been switched.

[0102] In some embodiments, the controller updates the code in the field-replaceable units (FRUs) to the code corresponding to the target operating mode. FUs typically refer to hardware components that are easy to replace and upgrade, such as modules on a network interface card (NIC). Updating the code of these units ensures their compatibility with the target operating mode, thereby improving the overall performance and stability of the NIC. The next step is only performed after both the firmware file in the NIC and the code in the FUs have been updated. This ensures that all critical components of the NIC are correctly configured, avoiding potential problems caused by asynchronous updates. The controller sends an indication message to the programmable device, informing it that the NIC's operating mode has been switched. Programmable devices (such as FPGAs, CPLDs, etc.) are typically configured to implement complex logic functions, and they need to adjust their behavior according to the NIC's operating mode. Sending the indication message ensures that the programmable device responds promptly to changes in operating mode, thus maintaining the normal operation of the entire system. Through these steps, the NIC's operating mode switching process becomes more controllable and reliable. The synchronous updates and indications of the firmware file, FUs code, and programmable devices ensure a smooth transition when the NIC switches operating modes, avoiding system crashes or performance degradation due to incompatibility or misconfiguration.

[0103] In an exemplary embodiment, after determining the current operating mode of the network interface card, the method further includes the following steps S61-S62:

[0104] Step S61: Determine the server resource allocation strategy corresponding to the current working mode;

[0105] It's important to note that the server's internal resource requirements change depending on the smart network interface card's (NIC) operating mode. For example, in data processing unit (DMU) mode, the NIC may require more access to storage, network, and CPU resources to enable its independent computing, networking, and storage offloading functions. In network interface card (NIC) mode, these requirements decrease, as the NIC primarily handles network transmission and consumes fewer server resources.

[0106] In some embodiments, the resource requirements of the smart network interface card (NIC) in different modes are first understood, including CPU time, memory, storage space, and network bandwidth. Secondly, based on the NIC's mode requirements, a resource allocation strategy is formulated. For example, when the NIC switches to data processing unit mode, the strategy includes enhancing the NIC's CPU core access permissions, reserving a portion of memory and storage space for the NIC, and optimizing network traffic allocation to ensure the NIC can efficiently process and transmit data. Furthermore, considering server stability and security, a resource redundancy strategy is formulated to ensure that the critical functions of the server and the NIC can still operate normally even under conditions of resource scarcity or failure.

[0107] Step S62: Allocate the server's internal resources based on the server resource allocation strategy.

[0108] Once the resource allocation strategy is determined, the server needs to dynamically reallocate internal resources according to this strategy. This may involve the following operations: CPU resource allocation: Adjusting the allocation of CPU cores to ensure that the smart NIC can access sufficient computing resources for data processing in data processing unit mode; Memory and storage resource allocation: Reserving or releasing memory and storage resources according to the needs of the smart NIC to ensure that the smart NIC can operate efficiently in different modes; Network resource allocation: Optimizing network bandwidth allocation to ensure that the smart NIC can independently provide network services in data processing unit mode without affecting the performance of other network devices; Hardware resource pooling: In data processing unit mode, the smart NIC may need to pool the server's internal hardware resources (such as PCIe devices and network interfaces) to support flexible allocation and sharing of hardware resources; Resource monitoring and adjustment: Dynamically monitoring the resource usage of the server and the smart NIC, and adjusting resources according to real-time load to achieve optimal performance and efficiency.

[0109] Through steps S61 and S62, the server can intelligently adapt to the resource requirements of the smart NIC in different modes and dynamically adjust resource allocation. This ensures that both the server and the smart NIC operate at their optimal state, improving overall system performance and resource utilization, while also enhancing system flexibility and scalability. This resource allocation and adjustment strategy is a crucial component of smart NIC mode switching technology and plays a key role in achieving efficient collaborative work between the smart NIC and the server.

[0110] In an exemplary embodiment, after determining the current operating mode of the network interface card, the method further includes the following steps S71-S73:

[0111] Step S71: Obtain the target model, which is set to predict network card failures based on the network card's historical operating data, wherein the network card includes the target network card;

[0112] Step S72: Using the target model, predict the failure probability of the target network card at the target time based on the operating data of the target network card, and obtain the target probability, where the target time is the time after the current time;

[0113] Step S73: If the target probability is greater than the preset probability, after determining that the target network card has failed, send an early warning alarm message, wherein the alarm warning message is used to indicate that the probability of the target network card failing at the target time is greater than the preset probability.

[0114] In this embodiment, the above method can be used to predict the failure of the target network card and avoid the target network card from suddenly failing at a certain time.

[0115] Obviously, the embodiments described above are only some embodiments of this application, and not all embodiments. To better understand the above method, the following description, in conjunction with embodiments, illustrates the process, but is not intended to limit the technical solutions of the embodiments of this application. Optionally, taking a BMC as the controller and a CPLD as the programmable device as an example:

[0116] This application proposes a power supply method for a network interface card (NIC) and a server that is compatible with both modes of a smart NIC. Through hardware design and software logic, the mode switching of the smart NIC can be made more convenient, reducing the debugging and design work of the server when changing the smart NIC mode, and improving efficiency and reliability.

[0117] As shown in Figure 3, in order to realize the mode switching of the smart network card, the switching method and device proposed in this application consists of the following modules:

[0118] (1) A power supply circuit that automatically switches according to the different mode requirements of a smart network card;

[0119] (2) A heat dissipation control strategy suitable for different modes of smart network cards;

[0120] (3) A sideband control signal circuit for communication between a smart network card and a server and for out-of-band debugging of the smart network card;

[0121] (4) A process and method for switching modes of smart network cards.

[0122] In this architecture, firstly, an FRU chip is designed to identify the smart network interface card (NIC) mode. The BMC can read / write information within the FRU, indicating whether the smart NIC is currently in NIC or DPU mode. Secondly, it needs to be able to switch between different power supplies to the NIC based on its mode, without requiring manual physical changes to the power connections by maintenance personnel. In NIC mode, it provides Core power (power to the server in S0 state); in DPU mode, it provides AUX (Auxiliary) power (power to the server in S5 state). Finally, a thermal management module is included. In NIC mode, the smart network interface card (NIC) operates only in the server's S0 state, so the server fan only needs to operate in S0 state. However, in DPU mode, the NIC operates in the server's S5 state, so fan speed adjustment is required after the server is plugged in with AC power. The entire process does not require manual adjustment of the cooling strategy after changing modes. Next is the sideband signal control module. In NIC mode, the NIC typically only requires one I2C connection for out-of-band management. In DPU mode, however, the NIC is more powerful and often acts as a host device to control the server. Therefore, it needs to provide I2C, GPIO (General Purpose Input / Output), and UART signals for communication with the server and for debugging the NIC. Thus, a circuit compatible with both modes is required for sideband control signals. Finally, based on this device, a specific process and method are used to switch the NIC's modes, reducing manual debugging and design work, and improving efficiency and reliability. During the process, the BMC and CPLD play a management and control role in completing the NIC mode switching process.

[0123] The above is the server architecture topology proposed in this application. Based on this, the following options are available:

[0124] 1. A power supply circuit that automatically switches according to the different mode requirements of a smart network card;

[0125] Different modes of smart network interface cards (NICs) have different power supply requirements. In DPU mode, the smart NIC can be considered an independent master device, operating independently when the server is not in operation (S5 state). In NIC mode, the smart NIC, like a regular NIC, is typically a slave device, operating when the server is powered on (S0 state). This results in different power supply requirements: NIC mode requires the server to supply power to the NIC in S0 state (server powered on); DPU mode requires the server to supply power to the NIC even in S5 state (server powered off).

[0126] The power supply circuit is shown in Figure 4. The 12V output voltage from the Power Supply Unit (PSU) is connected to the motherboard's power connector after being isolated by an Efuse. By opening and closing the Efuse, the power connector can be controlled to supply power to the network card in either S5 or S0 state. The CPLD has an internal logic judgment unit that controls the Efuse's on / off state.

[0127] 1) After the machine is powered on, the BMC completes initialization, determines whether the current smart network card is in NIC mode or DPU mode by reading the entire machine's FRU, and sends the information to the CPLD via the I2C bus;

[0128] 2) When the CPLD recognizes that the smart network card is in NIC mode, the CPLD controls the EN pin of the Efuse to open in the S0 state of the machine and supplies power to the smart network card through the power connector.

[0129] 3) When the CPLD recognizes that the smart network card is in DPU mode, the CPLD controls the EN pin of the Efuse to open in the S5 state of the machine and supplies power to the smart network card through the power connector.

[0130] 4) The BMC can also control the opening and closing of Efuse through the CPLD. It is set to power off and restart after upgrading the firmware of the smart network card so that the new firmware takes effect.

[0131] The above design enables automatic switching of power supply modes for different network cards.

[0132] II. A heat dissipation control strategy suitable for different modes of smart network cards;

[0133] This heat dissipation strategy addresses the different heat dissipation requirements of different modes of smart network cards, while also preventing the risk of the fan not spinning due to the heat dissipation strategy code freezing.

[0134] After AC power-on, a typical server is in S5 state, not powered on. Only the BMC and CPLD are active; the network card in NIC mode is inactive in this state, resulting in minimal overall heat dissipation and no need for fans. In contrast, a DPU in S5 state needs to operate, generating significant heat, requiring fan intervention for cooling. Therefore, the overall cooling solution must consider the different needs of these two modes and automatically adapt to different cooling scenarios.

[0135] As shown in Figure 5, the BMC is the control unit for the heat dissipation strategy, which controls the fan via a CPLD. The fan speed is controlled by a Pulse Width Modulation (PWM) signal, and the fan's on / off state is controlled by an enable (EN) signal. The CPLD acts as an auxiliary control unit for redundant fan control.

[0136] (1) After AC power-on, because the CPLD is hardware, it completes initialization before the BMC. The CPLD cannot obtain whether the current smart network card is in DPU mode or NIC mode through the BMC. In order to ensure that the smart network card will not overheat and crash due to working in DPU mode before the BMC takes over the heat dissipation control, the CPLD enables the fan by default and controls the fan speed to 30% until the BMC completes initialization and obtains the current working mode of the smart network card;

[0137] (2) After BMC completes initialization, it sends a heartbeat to CPLD to notify CPLD that BMC has started normal operation. BMC reads FRU information via I2C to determine the current network card mode, and then calls the corresponding network card cooling algorithm. It takes over the control of the fan, continuously monitors the network card temperature, and controls the fan to cool the machine according to the cooling algorithm;

[0138] (3) During operation, the CPLD continuously monitors the BMC's heartbeat signal. When the heartbeat signal is absent, it indicates that the BMC is stuck. The CPLD then takes over the fan control, adjusting the fan speed to 80% to prevent the system from overheating and crashing. This continues until the BMC returns to normal, at which point the CPLD takes over the cooling control again.

[0139] The above design enables automatic adaptation of heat dissipation strategies for different operating modes of smart network cards.

[0140] 3. A sideband control signal circuit for communication between a smart network card and a server, and for out-of-band debugging of the smart network card;

[0141] In NIC mode, out-of-band communication between the server and the network card is only via the I2C bus, used by the BMC to access the network card and obtain network card information, temperature, and other health status data. In DPU mode, the smart network card can operate in the server's S5 state and can act as a host to control the server. Therefore, in addition to traditional I2C, other sideband control signals are required to communicate with the server. In this architecture, the sideband control signal circuit also includes a Reset signal connected to the BMC for resetting the BMC in DPU mode; a UART signal connected to the BMC for debugging and log capture of the smart network card; and an RJ45 management network connected to the BMC's management port for managing the server in DPU mode. Therefore, the sideband signal control circuit in this solution needs to be compatible with the smart network card's requirements in different modes and can automatically switch between modes. The following architectures are possible:

[0142] (1) The I2C bus and UART bus are first connected to the CPLD, and then connected to the BMC through the CPLD;

[0143] (2) Implement UART bus MUX functionality in the CPLD. Switching between the system serial port and the smart network card's serial port can be achieved through BMC control. That is, when out-of-band upgrades or debugging of the smart network card model are needed, the serial port can be switched to the smart network card via the BMC. (Typically, servers only have a system serial port and a BMC serial port, and cannot access the smart network card's serial port).

[0144] (3) The Reset signal of the smart network card is connected to the BMC through the CPLD. When the smart network card acts as the master device in DPU mode, it can reset the server system.

[0145] (4) The RJ45 management network of the smart network card is connected to the BMC management network port via a network cable. In DPU mode, it is used for the smart network card to manage the server.

[0146] IV. A process and method for switching modes of a smart network interface card;

[0147] The smart network interface card (NIC) switching mode and process described in this article are implemented based on the architecture of the device described above, as follows:

[0148] (1) Stop the current server's running services (if there are no services, proceed to step (2));

[0149] (2) Refresh the corresponding firmware for "NIC mode" or "DPU mode" of the smart network card through BMC;

[0150] (3) After the smart network card FW is refreshed, the BMC controls the power supply of the smart network card to disconnect. After power is restored, the new mode of the smart network card will take effect.

[0151] (4) BMC refreshes the server's FRU, changing the current machine model code to the code corresponding to the network card mode;

[0152] (5) After completing the network card FW refresh and FRU refresh, the BMC notifies the CPLD that the network card mode has changed and the power supply and heat dissipation control logic is performed according to the new mode.

[0153] (6) After the server is restarted or restarted while powered on, the smart network card completes the mode switch and runs the new service.

[0154] It should be noted that this application also has the following advantages: the proposed device includes modules such as network card mode recognition, power supply mode switching, heat dissipation control, and sideband control signals, which can meet the needs of smart network cards in different working modes; after the smart network card is switched to different modes through the switching process proposed above, the server can automatically adapt, reducing the debugging and design work of the server when changing the smart network card mode, and improving efficiency and reliability.

[0155] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the related technology, can be embodied in the form of a software product. This computer software product is stored in a non-volatile readable storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0156] This embodiment also provides a server for implementing the above embodiments and optional implementations; details already described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the modules described in the following embodiments are preferably implemented in software, hardware implementations, or a combination of software and hardware, are also possible and contemplated.

[0157] Figure 7 is a structural block diagram of a server according to an embodiment of the present application. The server includes:

[0158] Management module 702, electronic fuse 704, power connector 706, network card 708;

[0159] One end of the electronic fuse 704 is electrically connected to the power supply unit, and the other end of the electronic fuse 704 is electrically connected to the power supply connector 706. When the enable pin of the electronic fuse 704 is turned on, the electronic fuse 704 allows current to pass through, and the power supply connector 706 allows the network card 708 to be provided with core power or auxiliary power.

[0160] The management module 702 is configured to determine the current operating mode of the network interface card 708, wherein the operating modes of the network interface card 708 include: network interface card mode and data processing unit mode; when the current operating mode is determined to be network interface card mode, the enable pin of the electronic fuse 704 is enabled when the server is running; when the current operating mode is determined to be data processing unit mode, the enable pin of the electronic fuse 704 is enabled when the server is powered off.

[0161] The aforementioned server determines the current operating mode of the network interface card (NIC). When the NIC is in network interface card mode, it controls the enable pin of the electronic fuse to be turned on while the server is running. When the NIC is in data processing unit mode, it controls the enable pin of the electronic fuse to be turned on while the server is off. This allows the server to meet the power supply requirements of the NIC in different modes online, avoiding the need for disassembly and debugging of the server hardware during NIC mode switching. This improves the efficiency of NIC mode switching and solves the problem of low efficiency caused by the server's inability to adapt to different NIC modes online during NIC mode switching.

[0162] In an exemplary embodiment, the management module 702 further includes a controller and a programmable device, wherein the controller is configured to determine the current operating mode of the network interface card and send the current operating mode to the programmable device; the programmable device is configured to, when the current operating mode is determined to be network interface card mode, control the enable pin of the electronic fuse to be turned on when the server is running; and when the current operating mode is determined to be data processing unit mode, control the enable pin of the electronic fuse to be turned on when the server is powered off.

[0163] In one exemplary embodiment, the controller is also configured to control the enable pin of the electronic fuse to close when the power to the network card is to be disconnected.

[0164] In an exemplary embodiment, the server further includes: a fan; a programmable device, further configured to obtain a first preset rotation speed before determining the network card's operating mode, and control the fan to rotate at the first preset rotation speed; the controller is further configured to determine a network card heat dissipation strategy corresponding to the network card's operating mode after determining the network card's operating mode, thereby obtaining a target network card heat dissipation strategy; and to determine the fan's rotation speed according to the target network card heat dissipation strategy, and control the fan to rotate at the rotation speed.

[0165] In one exemplary embodiment, the controller is further configured to send a heartbeat signal to the programmable device at each preset time interval; the programmable device is further configured to obtain a second preset speed through the programmable device and control the fan to rotate at the second preset speed when the heartbeat signal sent by the controller is not monitored.

[0166] In an exemplary embodiment, the network interface card (NIC) has a first network port, and the controller has a second network port, which are connected via a network cable. The NIC is also configured to, when in data processing unit mode, acquire data sent by the controller or send data to the controller based on the network communication connection between the NIC and the controller.

[0167] In an exemplary embodiment, the network interface card (NIC) has a data line for transmitting a reset signal between itself and the programmable device, and the programmable device has a data line for transmitting a reset signal between itself and the controller. The NIC is further configured to perform fault diagnosis on the controller based on the operating data sent by the controller when the NIC obtains the operating data based on the network communication connection between the NIC and the controller; and to send a reset signal to the programmable device and instruct the programmable device to forward the reset signal to the controller when a controller fault is determined based on the operating data, wherein the reset signal is used to reset the controller.

[0168] In an exemplary embodiment, the network interface card (NIC) and the programmable device have an integrated circuit interconnect bus and / or a universal asynchronous transceiver (UART) bus, and the programmable device and the controller have an integrated circuit interconnect bus and / or a UART bus; the controller is further configured to send data to the NIC or receive data sent by the NIC based on the integrated circuit interconnect bus and / or the UART bus.

[0169] In one exemplary embodiment, the controller is further configured to update the firmware file in the network card to the firmware file corresponding to the target operating mode when the current operating mode of the network card is to be switched to the target operating mode, wherein the operating mode of the network card includes the target operating mode; and to control the enable pin of the electronic fuse to be turned off.

[0170] In one exemplary embodiment, the server further includes: a field-replaceable unit; a controller, further configured to update the code in the field-replaceable unit to the code corresponding to the target operating mode when the current operating mode of the network interface card is to be switched to the target operating mode; and to send indication information to the programmable device when the firmware file in the network interface card and the code in the field-replaceable unit are updated, wherein the indication information is used to indicate that the operating mode of the network interface card of the programmable device has been switched.

[0171] In an exemplary embodiment, the management module 702 is further configured to determine a server resource allocation strategy corresponding to the current working mode of the network card after determining the current working mode; and to allocate internal resources of the server based on the server resource allocation strategy.

[0172] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0173] Embodiments of this application also provide a non-volatile computer-readable storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above method embodiments when it is run.

[0174] In some embodiments, in this embodiment, the computer program described above can be configured to perform the following steps:

[0175] S1, determine the current working mode of the network card, where the working mode of the network card includes: network interface card mode and data processing unit mode;

[0176] S2, when the current working mode is network interface card mode, the enable pin of the electronic fuse is turned on when the server is running;

[0177] S3, when the current working mode is data processing unit mode, controls the enable pin of the electronic fuse to be turned on when the server is powered off.

[0178] In one exemplary embodiment, the aforementioned non-volatile computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0179] An embodiment of this application also provides an electronic device, as shown in FIG8, which includes a memory 808 and a processor 804. The memory 808 stores a computer program, and the processor 804 is configured to execute the steps in any of the above method embodiments through the computer program.

[0180] In some embodiments, in this embodiment, the processor 804 described above can be configured to perform the following steps via a computer program:

[0181] S1, determine the current working mode of the network card, where the working mode of the network card includes: network interface card mode and data processing unit mode;

[0182] S2, when the current working mode is network interface card mode, the enable pin of the electronic fuse is turned on when the server is running;

[0183] S3, when the current working mode is data processing unit mode, controls the enable pin of the electronic fuse to be turned on when the server is powered off.

[0184] The optional examples in this embodiment can refer to the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0185] In some embodiments, those skilled in the art will understand that the structure shown in FIG8 is merely illustrative and does not limit the structure of the electronic device described above. For example, the electronic device may also include more or fewer components (such as network interfaces) than shown in FIG8, or have a different configuration than that shown in FIG8.

[0186] The memory 808 can be configured to store software programs and modules, such as the power supply method for the network card and the corresponding program instructions / modules for the server in this embodiment. The processor 804 executes various functional applications and data processing by running the software programs and modules stored in the memory 808, thereby implementing the aforementioned power supply method for the network card. The memory 808 may include high-speed random access memory and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 808 may include memory remotely located relative to the processor 804, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The memory 808 may, but is not limited to, be configured to store system configuration files and other information. As an example, as shown in FIG8, the memory 808 may, but is not limited to, include the management module 702, electronic fuse 704, power connector 706, and network card 708 in the aforementioned network server. In addition, it may include, but is not limited to, other module units in the aforementioned power supply device for the network card, which will not be described in detail in this example.

[0187] In some embodiments, the transmission device 806 described above is configured to receive or transmit data via a network. Examples of network options include wired and wireless networks. In one example, the transmission device 806 includes a network adapter that can be connected to other network devices and a router via a network cable to communicate with the Internet or a local area network. In one example, the transmission device 806 is a radio frequency (RF) module configured to communicate wirelessly with the Internet.

[0188] In addition, the aforementioned electronic device also includes: a display 808; and a connection bus 810, configured to connect the various module components in the aforementioned electronic device.

[0189] Embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0190] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0191] The embodiments described herein also provide a computer program that includes computer instructions stored in a non-volatile computer-readable storage medium; a processor of a computer device reads the computer instructions from the non-volatile computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps in any of the above method embodiments.

[0192] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0193] The above are merely optional embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A power supply method for a network card, characterized in that, Applied to a server, the server includes: an electronic fuse, a power connector, and a network interface card (NIC); one end of the electronic fuse is electrically connected to a power supply unit, and the other end of the electronic fuse is electrically connected to the power connector. When the enable pin of the electronic fuse is turned on, the electronic fuse allows current to flow, and the power connector allows the NIC to be supplied with core power or auxiliary power. include: Determine the current operating mode of the network interface card (NIC), wherein the operating mode of the NIC includes: network interface card mode and data processing unit mode; When the current operating mode is the network interface card mode, the enable pin of the electronic fuse is turned on when the server is running. When the current operating mode is the data processing unit mode, the enable pin of the electronic fuse is turned on when the server is powered off.

2. The method according to claim 1, characterized in that, When the current operating mode is the data processing unit mode, if the server is powered on, the enable pin of the electronic fuse is in the open state.

3. The method according to claim 1, characterized in that, The server also includes: a controller and a programmable device; The method further includes: The controller determines the current operating mode of the network card and sends the current operating mode to the programmable device. When the programmable device determines that the current working mode is the network interface card mode, the enable pin of the electronic fuse is turned on by the programmable device when the server is running. When the programmable device determines that the current operating mode is the data processing unit mode, the programmable device controls the enable pin of the electronic fuse to be turned on when the server is in a powered-off state.

4. The method according to claim 3, characterized in that, The programmable device and the controller have an integrated circuit interconnect bus, and sending the current operating mode to the programmable device includes: The current operating mode is sent to the programmable device via the integrated circuit interconnect bus.

5. The method according to claim 3, characterized in that, The server further includes: a field-replaceable unit, wherein determining the current operating mode of the network interface card via the controller includes: After the server is powered on, the controller completes initialization and reads the field replaceable unit to determine whether the current working mode of the network card is the network interface card mode or the data processing unit mode.

6. The method according to claim 3, characterized in that, The method further includes: When the controller is about to disconnect the power to the network card, the controller controls the enable pin of the electronic fuse to close.

7. The method according to claim 1, characterized in that, The server also includes: a fan; Before determining the working mode of the network card, the method further includes: obtaining a first preset speed and controlling the fan to rotate at the first preset speed; After determining the working mode of the network card, the method further includes: determining a network card heat dissipation strategy corresponding to the working mode of the network card to obtain a target network card heat dissipation strategy; determining the rotation speed of the fan according to the target network card heat dissipation strategy, and controlling the fan to rotate at the rotation speed.

8. The method according to claim 7, characterized in that, The server further includes a controller and a programmable device; wherein the programmable device is configured to control the fan to rotate at the first preset speed, and the controller is configured to control the fan to rotate at the specified speed. The method further includes: The controller sends a heartbeat signal to the programmable device at each preset time interval; After determining the working mode of the network card, without monitoring the heartbeat signal sent by the controller, the programmable device obtains a second preset speed and controls the fan to rotate at the second preset speed.

9. The method according to claim 1, characterized in that, The server further includes: a controller; the network card has a first network port, the controller has a second network port, and the first network port and the second network port are connected via a network cable; The method further includes: When the network interface card (NIC) is in the data processing unit mode, data sent by the controller can be obtained or sent to the controller through the NIC based on the network communication connection between the NIC and the controller.

10. The method according to claim 9, characterized in that, The server further includes: a programmable device; a data line for transmitting a reset signal is provided between the network card and the programmable device, and a data line for transmitting the reset signal is provided between the programmable device and the controller; The method further includes: Using the network interface card (NIC), and in the case where the operating data sent by the controller is obtained based on the network communication connection between the NIC and the controller, fault diagnosis of the controller is performed based on the operating data. If a controller malfunction is determined based on the operating data, a reset signal is sent to the programmable device via the network card, and the programmable device is instructed to forward the reset signal to the controller, wherein the reset signal is used to reset the controller.

11. The method according to claim 1, characterized in that, The server further includes: a controller and a programmable device; the network interface card (NIC) and the programmable device are connected by an integrated circuit interconnect bus and / or a universal asynchronous transceiver (UAST) bus, and the programmable device and the controller are connected by an integrated circuit interconnect bus and / or a UAST bus; The method further includes: The controller can send data to or receive data sent by the network card via the integrated circuit interconnect bus and / or universal asynchronous transceiver bus.

12. The method according to claim 1, characterized in that, The server also includes: a controller; The method further includes: When the current working mode of the network card is to be switched to the target working mode, the firmware file in the network card is updated to the firmware file corresponding to the target working mode through the controller, wherein the working mode of the network card includes the target working mode; The controller controls the enable pin of the electronic fuse to close.

13. The method according to claim 12, characterized in that, After updating the firmware file in the network card to the firmware file corresponding to the target operating mode via the controller, the method further includes: The power supply to the network card is disconnected via the controller; The controller re-powers the network card, switching the network card's current operating mode to the target operating mode.

14. The method according to claim 12, characterized in that, The server also includes: a programmable device and a field-replaceable unit; The method further includes: When the current working mode of the network card needs to be switched to the target working mode, the code in the field replaceable unit is updated to the code corresponding to the target working mode through the controller; When the firmware file in the network card and the code in the field replaceable unit are updated, the controller sends an indication message to the programmable device, wherein the indication message is used to indicate that the operating mode of the network card in the programmable device has been switched.

15. The method according to claim 14, characterized in that, The method further includes: Once the firmware file in the network card and the code in the field replaceable unit have been updated, the server is controlled to restart or be rebooted while powered on, switching the current working mode of the network card to the target working mode.

16. The method according to claim 1, characterized in that, After determining the current operating mode of the network card, the method further includes: Determine the server resource allocation strategy corresponding to the current working mode; The internal resources of the server are allocated based on the server resource allocation strategy.

17. A server, characterized in that, include: Management module, electronic fuse, power connector, network card; One end of the electronic fuse is electrically connected to the power supply unit, and the other end of the electronic fuse is electrically connected to the power supply connector. When the enable pin of the electronic fuse is turned on, the electronic fuse allows current to flow, and the power supply connector allows the network card to be supplied with core power or auxiliary power. The management module is configured to determine the current operating mode of the network interface card (NIC), wherein the operating mode of the NIC includes: network interface card mode and data processing unit mode; when the current operating mode is determined to be the network interface card mode, the module controls the enable pin of the electronic fuse to be turned on when the server is running; when the current operating mode is determined to be the data processing unit mode, the module controls the enable pin of the electronic fuse to be turned on when the server is powered off.

18. A non-volatile computer-readable storage medium, characterized in that, The non-volatile computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of claims 1 to 16.

19. 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, it implements the steps of the method described in any one of claims 1 to 16.

20. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 16.

Images