Intelligent network card management and control method, device and system and electronic equipment

Abstract

The invention discloses an intelligent network card management and control method, device and system and electronic equipment, and relates to the technical field of computers, the method comprises the steps that temporary data of an intelligent network card is reserved and cached in a preset persistent cache region, when a system of target computing equipment is restarted or shut down, the data in the preset persistent cache region cannot be cleared, and the data in the target computing equipment cannot be cleared; when the intelligent network card enters the initialization stage after the system is started, data recovery is performed on the intelligent network card based on the reserved data cached in the preset persistent cache region, so that rapid initialization of the intelligent network card is realized, the initialization time of the intelligent network card is shortened, and important data of the intelligent network card is prevented from being lost.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, system, and electronic device for managing and controlling an intelligent network card. Background Art

[0002] With the rapid development of cloud computing, edge computing, and high-performance computing scenarios, Smart NICs have become core components of computing devices such as servers due to their integrated high-performance data processing, network offloading, and security acceleration capabilities.

[0003] In the prior art, servers and other computing devices clear temporary data from all peripherals, including Smart NICs, when the system is restarted or shut down. However, due to the complex configuration of Smart NICs, reconfiguring the Smart NICs during system startup and initialization takes a significant amount of time. This not only prolongs the initialization time but also risks losing important data on the Smart NICs, reducing their security. Summary of the Invention

[0004] The present application provides a smart network card management and control method, device, system and electronic device to at least solve the problem in the related art that the initialization time of the smart network card is prolonged and there is a risk of losing important data of the smart network card.

[0005] This application provides a smart network card management and control method, including: When the smart network card in the target computing device meets a preset data retention condition, obtaining temporary data of the smart network card; Cache the temporary data of the Smart NIC as retained data in a preset persistent cache area; When the smart network card enters the initialization phase, data of the smart network card is restored based on the retained data cached in the preset persistent cache area, so that the smart network card completes the initialization.

[0006] This application also provides a smart network card management and control device, including: An acquisition module, configured to acquire temporary data of the smart network card in the target computing device when the smart network card meets a preset data retention condition; A cache module is used to cache the temporary data of the smart network card as retained data in a preset persistent cache area; The management and control module is used to restore data of the smart network card based on the retained data cached in the preset persistent cache area when the smart network card enters the initialization phase, so that the smart network card completes the initialization.

[0007] The application also provides a smart network card management and control system, including: a smart network card, a bus competition terminal and a management and control device; The smart network card connects the bus competition end and the control device through a preset bus; The bus competition end and the control device access the smart network card by competing for the control of the preset bus; The control device controls the smart network card based on any of the above-mentioned smart network card control methods.

[0008] The present application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for implementing the steps of any of the above-mentioned smart network card management and control methods when executing the computer program.

[0009] The present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned smart network card management and control methods are implemented.

[0010] The present application also provides a computer program product, including a computer program, which implements the steps of any of the above-mentioned smart network card management and control methods when executed by a processor.

[0011] Through this application, since the temporary data of the smart network card is retained and cached in a preset persistent cache area, the data in the preset persistent cache area will not be cleared when the target computing device is restarted or shut down. When the smart network card enters the initialization phase after the system is turned on, the data of the smart network card is restored based on the retained data cached in the preset persistent cache area, thereby achieving rapid initialization of the smart network card, shortening the initialization time of the smart network card, and avoiding the loss of important data of the smart network card. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0013] Figure 1 A flowchart of the smart network card management and control method provided in an embodiment of the present application; Figure 2 A flowchart of an exemplary smart network card management and control method provided in an embodiment of the present application; Figure 3 A flowchart of another exemplary smart network card management and control method provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of the intelligent network card management and control device provided in an embodiment of the present application; Figure 5A schematic diagram of the structure of the intelligent network card management and control system provided in an embodiment of the present application; Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0014] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0015] It should be noted that, in the description of this application, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. The terms "first," "second," etc., in this application are used to distinguish similar objects, and are not used to describe a particular order or sequence.

[0016] With the rapid development of cloud computing, edge computing, and high-performance computing, Smart NICs (such as the NVIDIA BlueField series and MBF2H332A-AENOT) have become core components of server systems due to their integrated high-performance data processing, network offload, and security acceleration capabilities. In server remote management scenarios, the Baseboard Management Controller (BMC), a hardware management unit independent of the operating system, is responsible for status monitoring, fault diagnosis, and initialization control of peripherals such as Smart NICs. Traditional BMC firmware design primarily caters to the management needs of standard NICs, with default logic including: Clears temporary status data of all peripherals when the system restarts or shuts down. The BMC executes the power-on self-test (POST) process in parallel with the Basic Input Output System (BIOS), directly accessing shared hardware resources (such as PCIe controllers and chipsets). It identifies network cards and stores information based on physical slot location or device enumeration order.

[0017] However, due to the complex configuration of the smart network card, when the system is turned on to initialize the smart network card, it takes a lot of time to reconfigure the smart network card, which not only prolongs the initialization time of the smart network card, but also poses the risk of losing important data of the smart network card, thereby reducing the security of the smart network card.

[0018] In order to solve the above technical problems, the embodiments of the present application provide a method, device, system and electronic device for managing and controlling a smart network card. The method includes: obtaining temporary data of the smart network card in a target computing device when the smart network card meets a preset data retention condition; caching the temporary data of the smart network card as retained data in a preset persistent cache area; and when the smart network card enters the initialization phase, restoring data of the smart network card based on the retained data cached in the preset persistent cache area to complete the initialization of the smart network card. In the method provided by the above scheme, since the temporary data of the smart network card is retained and cached in the preset persistent cache area, the data in the preset persistent cache area will not be cleared when the target computing device is restarted or shut down. When the smart network card enters the initialization phase after the system is turned on, the data of the smart network card is restored based on the retained data cached in the preset persistent cache area, thereby achieving rapid initialization of the smart network card, shortening the initialization time of the smart network card, and avoiding the loss of important data of the smart network card.

[0019] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0020] The present invention provides a method for managing and controlling a Smart NIC (Smart Network Interface Card) (SNIC) in a computing device, such as a server, using a BMC. The method is used to retain temporary data on the Smart NIC (Smart Network Interface Card) on the computing device and quickly initialize the Smart NIC. The embodiments of the present invention are implemented by electronic devices, such as servers, desktop computers, laptops, tablet computers, and other electronic devices capable of managing and controlling Smart NICs.

[0021] like Figure 1 FIG. 1 is a flow chart of a smart network card management and control method provided in an embodiment of the present application, the method comprising: Step 101: When a smart network card in a target computing device meets a preset data retention condition, temporary data of the smart network card is obtained.

[0022] Temporary data for SmartNICs includes at least SmartNIC configuration information, sensor data, and firmware logs. SmartNIC configuration information includes PCIe interface configuration information, built-in data processing unit (DPU) configuration information, and current register values. Sensor data includes the physical and operational status data of the SmartNIC. Physical status data may include temperature, power supply voltage, and power consumption, while operational status data includes DPU load rate, network throughput, and link status information. Firmware logs include SmartNIC startup logs, operational logs, and exceptions.

[0023] It should be noted that since the method provided in the embodiment of the present application is applied to the BMC, the BMC monitors the status of the smart network card in real time, and when the current status of the smart network card presets the data retention condition, it triggers the smart network card temporary data acquisition process to obtain the temporary data of the smart network card.

[0024] Step 102: Cache the temporary data of the smart network card as reserved data in a preset persistent cache area.

[0025] Specifically, the BMC marks the acquired temporary data as retained data and writes it to a preset non-volatile storage area (such as the eMMC storage area). This storage area is an independent partition allocated in the eMMC (such as / dev / mmcblk0p5), using the EXT4 file system format, specifically for storing smart network card data. Among them, configuration information such as the current value of the register can be saved in the form of a binary image (such as reg_backup.bin). The configuration information also includes PCIe configuration space, DPU register snapshots, etc.; sensor data and firmware logs can be serialized and stored as CSV files (such as sensor_log_20231001.csv) by timestamp, recording historical data such as temperature, power consumption, and link status.

[0026] Step 103 : When the smart network card enters the initialization phase, data of the smart network card is restored based on the retained data cached in the preset persistent cache area, so that the smart network card completes initialization.

[0027] Among them, the smart network card takes NVIDIA BlueField-2 MBF2H332A-AENOT as an example, which supports PCIe Gen4x16 interface and integrates Arm Cortex-A72 DPU core and ConnectX-6 Dx network controller.

[0028] Specifically, after the system restarts, the Basic Input / Output System (BIOS) completes the power-on self-test (POST) and releases control of hardware resources. The BMC triggers the SmartNIC initialization process. The BMC reads the retained data from the eMMC persistent cache and restores it to the SmartNIC. By directly reusing the retained data, the SmartNIC quickly completes initialization and returns to its pre-shutdown state without having to re-execute the complex configuration process.

[0029] Based on the above embodiment, as an implementable manner, in one embodiment, the method further includes: Step 201: When a system restart instruction or a system shutdown instruction of a target computing device is received, the physical presence status and power supply status of the smart network card are obtained; Step 202 : Determine whether the smart network card meets a preset data retention condition based on the physical presence status and power supply status of the smart network card.

[0030] Specifically, the target computing device's baseboard management controller (BMC), a hardware management unit independent of the operating system, monitors system commands in real time. When the BMC receives a system restart command (such as a remote restart command sent via the IPMI protocol) or a system shutdown command, it triggers a SmartNIC status check to determine the SmartNIC's physical presence and power supply status. Based on these two conditions, the BMC then determines whether the SmartNIC meets the preset data retention requirements.

[0031] Specifically, in one embodiment, whether the smart network card is in a normal operating state can be determined based on the physical presence status and power supply status of the smart network card; if it is determined that the smart network card is in a normal operating state, it is determined that the smart network card meets the preset data retention condition.

[0032] Among them, the BMC in the embodiment of the present application adopts the ASPEED AST2600 series, integrates the ARM Cortex-A7 processor, has an independently running management subsystem, and communicates with the Complex Programmable Logic Device (CPLD) through the IPMB (Intelligent Platform Management Bus).

[0033] Specifically, if Figure 2 The figure shows a flow chart of an exemplary smart network card management and control method provided by an embodiment of the present application. When the baseboard management controller (BMC) receives a system restart command or system shutdown command from the target computing device, it communicates with the complex programmable logic device (CPLD) via the I2C bus. The CPLD is responsible for real-time acquisition of the physical signals of the smart network card. The physical presence status is detected by the level of the PRSNT# pin of the smart network card. A low level indicates that the network card is physically connected to the device slot, that is, in place. The power supply status is detected by the PWR_GOOD signal of the 3.3V power supply. A high level indicates that the network card power supply is stable, that is, the power-on is normal.

[0034] The CPLD converts the collected status signals into binary flags (e.g., Present = 1 indicates presence, Power_Good = 1 indicates normal power supply) and stores them in its own register (e.g., address 0x20). The BMC then reads these registers to obtain the physical presence and power supply status of the SmartNIC. If both status flags are 1 (i.e., Present & Power_Good = 1), the SmartNIC is physically present and powered normally. This indicates that the SmartNIC is in a valid operating state, and its temporary data has retention value. Therefore, the preset data retention conditions are met. The CPLD then retrieves the temporary data and writes it as retained data to the preset persistent cache. After data retention is complete, the system is shut down. If either flag is 0 (e.g., Present = 0 due to removal of the NIC, or Power_Good = 0 due to a power outage), the retention conditions are not met. The default logic prevents data retention and the system is shut down.

[0035] Specifically, through dual detection of physical presence and power supply status, invalid data retention of removed or powered-off smart network cards is avoided, redundant operations of the BMC are reduced, and resource utilization of the BMC is improved.

[0036] Based on the above embodiment, as an implementable manner, in one embodiment, the method further includes: Step 301: Monitor the firmware running status information of the smart network card in real time; Step 302 : When the firmware running status information of the smart network card indicates that the smart network card is in a high-load running state, determine that the smart network card meets a preset data retention condition.

[0037] Specifically, the baseboard management controller (BMC) accesses the SmartNIC to collect real-time information about the SmartNIC's firmware operating status. The SmartNIC's built-in firmware periodically reports its operating parameters to the BMC, which then uses its built-in status monitoring function to analyze and record these parameters in real time, obtaining operational data such as the current load rate of the SmartNIC's data processing units (DPUs).

[0038] Furthermore, the BMC can include a built-in threshold comparator to pre-set a threshold for determining when a SmartNIC is operating under high load (e.g., DPU load rate > 80%). When the BMC detects that the SmartNIC's firmware operating status reaches this threshold, it determines that the SmartNIC is operating under high load. At this point, the BMC determines that the SmartNIC meets the data retention criteria. Because SmartNICs operating under high load carry a higher risk of failure, to prevent critical configuration and data loss due to SmartNIC failures, the system enters state retention preparation mode even if no system restart or shutdown commands are received, ensuring that critical configuration and data are not lost during high load conditions.

[0039] Based on the above embodiment, as an implementable approach, in one embodiment, temporary data of the Smart NIC is cached as retained data in a preset persistent cache area, including: Step 1021: encrypt the retained data to obtain encrypted retained data. Step 1022: Cache the encrypted retained data into a preset persistent cache area.

[0040] Specifically, to prevent malicious tampering of retained data, the BMC can encrypt the retained data using encryption algorithms such as AES-256-GCM through the hardware acceleration engine, and then generate a SHA-256 checksum value stored in the same partition to form a data integrity verification mechanism. At the same time, the BMC prohibits the normal network card initialization process from accessing the partition through the file system lock flock() and the write protection flag (WR_PROTECT=1), preventing the reserved data of the smart network card from being overwritten. When the smart network card data needs to be restored, the BMC reads the encrypted retained data from the eMMC persistent cache, restores the original data through the hardware decryption engine, and verifies the SHA-256 checksum value to ensure the data is valid.

[0041] Based on the above embodiment, as an implementable manner, in one embodiment, obtaining temporary data of the smart network card includes: Step 1011, determining whether the preset bus is in the target state; Step 1012: When the preset bus is in the target state, access the smart network card through the preset bus to obtain temporary data of the smart network card.

[0042] The target state includes an idle state.

[0043] It should be noted that in target computing devices such as servers, the SmartNIC, BMC, and BIOS access the default bus (LPC bus) through shared resources. Before the BMC needs to access the SmartNIC through the default bus, it must first determine whether the default bus is idle. This means that the default bus is not occupied by other components and there is no data transmission or resource access operation.

[0044] Specifically, the BMC monitors the status signals of the preset bus or exchanges status information with the BIOS to determine whether other components are currently accessing devices such as the SmartNIC through the bus. If the bus is not occupied, it determines that the preset bus is in the target state (idle state).

[0045] Specifically, in one embodiment, the access phase of the bus competing end to the shared resource may be obtained; and according to the access phase of the bus competing end to the shared resource, it is determined whether the preset bus is in the target state.

[0046] The bus competing end accesses the preset bus through shared resources.

[0047] It should be noted that the bus contention end includes the BIOS, and the shared resources include the PCIe Switch chip (such as Broadcom PEX8796) and the 9641 chip (responsible for power management and sensor aggregation). The shared resources are connected to the BMC through the LPC bus.

[0048] Specifically, the BMC maintains a mapping table of POST phases and resource occupancy, associating the BIOS access phase with the shared resource occupancy status, allowing for real-time updates of the BIOS's shared resource occupancy status (access phase). If the BIOS is in the shared resource access phase, the mapping table marks it as "resource exclusive," indicating that the bus is currently occupied by the BIOS and is not in the target state. If the BIOS is in the shared resource access phase, the mapping table marks it as "resource released," indicating that the BIOS has stopped accessing shared resources and that the bus is currently in the idle target state. The BMC determines whether the bus is idle by comparing the BIOS's current access phase with the mapping table. When the BIOS completes POST and releases resources, the bus is considered to be in the target state.

[0049] Specifically, the embodiment of the present application obtains the access stage of the bus competing end such as BIOS to the shared resources, determines whether the preset bus is occupied by the bus competing end such as BIOS, avoids forced access when the bus competing end monopolizes the resources, and thus avoids bus read and write conflicts of the temporary data of the smart network card, thereby ensuring the integrity and accuracy of the temporary data finally obtained.

[0050] Specifically, in one embodiment, a control right release signal of the bus competing end may be detected to obtain a control right release signal detection result; and the access phase of the bus competing end to the shared resource is determined based on the control right release signal detection result.

[0051] The control release signal is an indicator that the BIOS releases the control of shared resources. It usually exists in the form of a hardware flag or a shared register status and is used to notify other components (BMC) that they can access shared resources.

[0052] Specifically, if the detection result is not released, it is determined that the BIOS is in the resource access stage, at which time the shared resources are exclusively occupied by the BIOS; if the detection result is released, it is determined that the BIOS is in the resource release stage, at which time the BIOS has completed hardware initialization and no longer occupies shared resources.

[0053] Specifically, in one embodiment, it is determined whether the bus competing end has timed out and has not released its control right; if the bus competing end has timed out and has not released its control right, the bus competing end is forced to suspend access to the shared resource so that the preset bus enters the target state.

[0054] Specifically, by determining whether the bus competing end has timed out and has not released its control right, it is possible to avoid affecting the BMC's acquisition of smart network card temporary data and smart network card data recovery operations due to the bus competing end occupying the preset bus for a long time.

[0055] Specifically, in one embodiment, when the control right of a bus competing end is not released after a timeout, a conflict log of the bus competing end is generated.

[0056] The conflict log includes at least the control right acquisition time, the control right occupation time and the operation record during the control right occupation time of the bus contention terminal.

[0057] Specifically, since a bus contender's failure to release control after a timeout may be due to a fault, a conflict log is generated when a bus contender's failure to release control occurs. By recording the time a bus contender acquires control, the time it holds control, and the operations performed during that time, subsequent technicians can analyze the conflict log to identify the fault type and cause, thereby improving maintenance efficiency.

[0058] For example, Figure 3The figure shows a flowchart of another exemplary Smart NIC management and control method provided by an embodiment of the present application. In addition to obtaining access to a preset bus during the temporary data acquisition phase, access to the preset bus is also required during the Smart NIC initialization phase. Taking the target computing device system startup as an example, the BMC first completes its own initialization and startup to prepare for subsequent hardware monitoring and interaction. The BMC continuously monitors the BIOS's POST (Power-On Self-Test) process status. POST is the hardware detection and initialization process after the BIOS boots up. The BMC obtains real-time information about this process through hardware interfaces (such as the LPC bus) to determine hardware readiness. The BMC determines whether the BIOS has entered the POST phase. If not, the BMC continues to monitor in a loop. If it has, it triggers a subsequent soft lock operation. Before the BIOS completes the POST phase, the BMC initiates a soft lock operation (via atomic instructions or semaphores) on shared resources to prevent the BMC from actively accessing the Smart NIC. Upon detecting that the BIOS has released control (e.g., the POST completion flag POST_Complete = 1), the BMC releases the lock and initiates the Smart NIC initialization and monitoring thread. The BMC uses the Token Ring protocol to negotiate a shared resource access token with the BIOS. At the same time, only the BMC or BIOS holding the token can access shared resources. Based on this protocol, the BMC exchanges tokens with the BIOS through shared registers to ensure that when the BMC obtains access rights, the BIOS does not compete for resources. The BMC continues to monitor the BIOS status to determine whether the POST has ended. If not, it determines whether the bus competing end has timed out and has not released the control rights. If the occupied time exceeds 500ms, that is, if the control release signal is not received for more than 500ms, the BMC triggers an SM1 interrupt to suspend the current BIOS operation (forcibly suspending the bus competing end's access to shared resources) and record a conflict log for subsequent troubleshooting and tracing. If it is determined that the BIOS has completed the POST stage or has forcibly suspended the bus competing end's access to shared resources, the BMC performs a soft lock release operation and starts the smart network card initialization process to achieve fast initialization of the smart network card based on the retained data. Before the system is shut down, the Token Holder (the permission status record in the Token Ring protocol) is also written to the preset persistent cache area. It is used to restore the resource access permission status and resource access order after the system is started up, ensuring the timing consistency of the entire system hardware collaboration, that is, ensuring the timing consistency of the smart network card data recovery and BIOS operation.

[0059] Based on the above embodiment, as an implementable approach, in one embodiment, data is restored for each Smart NIC in sequence based on the retained data cached in a preset persistent cache area according to the retained data cache order, so that each Smart NIC completes initialization in sequence, including: Step 1031: If there is more than one Smart NIC, when the Smart NIC enters the initialization phase, the reserved data cache order corresponding to each Smart NIC is obtained in the preset persistent cache area; Step 1032 : In accordance with the reserved data cache order, based on the reserved data cached in the preset persistent cache area, data is restored for each smart network card in sequence, so that each smart network card is initialized in sequence.

[0060] Specifically, if there are multiple SmartNICs, the BMC can assign a cache sequence number to each SmartNIC when caching retained data and establish a link between the cache sequence number and the SmartNIC's unique identifier. When the system restarts and the SmartNIC enters the initialization phase, the BMC determines the cache sequence number of the corresponding retained data based on the SmartNIC's unique identifier. Then, based on the cache order of the retained data represented by the cache sequence number, the BMC sequentially restores data from each SmartNIC, avoiding resource conflicts caused by simultaneous initialization of multiple SmartNICs.

[0061] Specifically, in one embodiment, the configuration registers of each smart network card can be configured with parameters in sequence according to the smart network card configuration information cached in the preset persistent cache area in accordance with the retained data cache order; the sensor data and firmware logs cached in the preset persistent cache area can be stored in the target data cache area of ​​each smart network card in accordance with the retained data cache order, so that each smart network card can complete initialization in sequence.

[0062] Among them, data recovery includes configuration register parameter configuration and target data buffer area data storage.

[0063] Specifically, the BMC accesses the configuration registers of each SmartNIC in sequence and writes the configuration information into them. Configuration registers are hardware-level storage units for SmartNICs, storing the configuration parameters that determine SmartNIC functionality. After completing the configuration register parameter configuration, the BMC writes the sensor data and firmware logs to the target data cache of the corresponding SmartNIC, such as the SmartNIC's built-in SRAM.

[0064] Specifically, in one embodiment, when a common network card is provided in the target computing device, a retained data filtering process is performed on the initialization process of the common network card.

[0065] Specifically, to prevent the cached reserved data from being mistakenly written into the common network card, the reserved data is filtered during the initialization process of the common network card to ensure the security of the reserved data of the smart network card.

[0066] Specifically, in one embodiment, an anomaly detection model can also be constructed based on the retained data cached in the preset persistent cache area; during the operation of the smart network card, fault prediction of the smart network card is performed based on the anomaly detection model to obtain a fault prediction result of the smart network card.

[0067] Specifically, since the preset persistent cache stores historical data on the Smart NIC, this data contains long-term operational status characteristics of the Smart NIC, such as temperature fluctuation patterns, power consumption trends, and DPU load. Therefore, it serves as the foundation for building an anomaly detection model. During Smart NIC operation, the BMC collects the Smart NIC's current status data in real time and inputs it into the anomaly detection model. The model compares the current status data with historical normal characteristics. If any characteristics exceed the normal range, it outputs a fault prediction result, such as the risk of heat dissipation failure or DPU overload.

[0068] The smart network card management and control method provided in the embodiment of the present application obtains the temporary data of the smart network card when the smart network card in the target computing device meets the preset data retention conditions; caches the temporary data of the smart network card as retained data in a preset persistent cache area; and when the smart network card enters the initialization phase, restores the data of the smart network card based on the retained data cached in the preset persistent cache area to complete the initialization of the smart network card. The method provided by the above scheme retains the temporary data of the smart network card and caches it in the preset persistent cache area. When the target computing device restarts or shuts down the system, the data in the preset persistent cache area will not be cleared. When the smart network card enters the initialization phase after the system is turned on, the data of the smart network card is restored based on the retained data cached in the preset persistent cache area, thereby achieving rapid initialization of the smart network card, shortening the initialization time of the smart network card and avoiding the loss of important data of the smart network card. In addition, by avoiding access conflicts on bus contention terminals such as BIOS, the reliability and stability of smart network card management and control are further improved. In addition, by building an anomaly detection model, early warning of hardware failure risks is provided, optimizing operation and maintenance efficiency.

[0069] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method.

[0070] The embodiments of the present application also provide a smart network card management and control device for executing the smart network card management and control method provided in the above embodiments.

[0071] like Figure 4 FIG2 is a schematic diagram of the structure of the intelligent network card management and control device provided in an embodiment of the present application. The intelligent network card management and control device 40 includes: an acquisition module 401, a cache module 402 and a management and control module 403.

[0072] Among them, the acquisition module is used to obtain the temporary data of the smart network card when the smart network card in the target computing device meets the preset data retention conditions; the cache module is used to cache the temporary data of the smart network card as retained data in the preset persistent cache area; the management and control module is used to restore the data of the smart network card based on the retained data cached in the preset persistent cache area when the smart network card enters the initialization phase, so that the smart network card completes the initialization.

[0073] Specifically, in one embodiment, the device further includes: A judgment module, configured to obtain the physical presence status and power supply status of the smart network card when receiving a system restart instruction or a system shutdown instruction of the target computing device; According to the physical presence status and power supply status of the smart network card, it is determined whether the smart network card meets the preset data retention condition.

[0074] Specifically, in one embodiment, the judgment module is specifically configured to: Determine whether the Smart NIC is operating normally based on its physical presence and power supply status. When it is determined that the smart network card is in a normal operating state, it is determined that the smart network card meets a preset data retention condition.

[0075] Specifically, in one embodiment, the judgment module is further configured to: Real-time monitoring of the firmware running status of the smart network card; When the firmware running state information of the smart network card indicates that the smart network card is in a high-load running state, it is determined that the smart network card meets the preset data retention condition.

[0076] Specifically, in one embodiment, the cache module is specifically configured to: Performing data encryption processing on the retained data to obtain encrypted retained data; Cache the encrypted retained data into the preset persistent cache area.

[0077] Specifically, in one embodiment, the acquisition module is specifically configured to: Determine whether the preset bus is in the target state; When the preset bus is in the target state, accessing the smart network card through the preset bus to obtain temporary data of the smart network card; The target state includes an idle state.

[0078] Specifically, in one embodiment, the acquisition module is specifically configured to: The stage of obtaining access to shared resources from the bus competing end; According to the access phase of the bus competing end to the shared resource, determining whether the preset bus is in the target state; The bus competing end accesses the preset bus through shared resources.

[0079] Specifically, in one embodiment, the acquisition module is specifically configured to: Detecting a control right release signal at a bus contention end to obtain a control right release signal detection result; According to the control right release signal detection result, the access phase of the bus contention end to the shared resource is determined.

[0080] Specifically, in one embodiment, the acquisition module is further configured to: Determine whether the bus contention end has a control right timeout and has not been released; If the control right of the bus competing end is not released within a timeout period, the bus competing end is forced to suspend access to the shared resource so that the preset bus enters the target state.

[0081] Specifically, in one embodiment, the acquisition module is further configured to: When the control right of the bus competing end times out and is not released, a conflict log of the bus competing end is generated; The conflict log includes at least the control right acquisition time, the control right occupation time and the operation record during the control right occupation time of the bus contention terminal.

[0082] Specifically, in one embodiment, the management and control module is specifically configured to: In the case where the Smart NIC is not unique, when the Smart NIC enters the initialization phase, the reserved data cache order corresponding to each Smart NIC is obtained in the preset persistent cache area; According to the reserved data cache order, based on the reserved data cached in the preset persistent cache area, data of each smart network card is restored in sequence, so that each smart network card is initialized in sequence.

[0083] Specifically, in one embodiment, the retained data includes at least smart network card configuration information, sensor data, and firmware logs.

[0084] Specifically, in one embodiment, the management and control module is specifically configured to: According to the order of retaining data cache, and based on the smart network card configuration information cached in the preset persistent cache area, the configuration registers of each smart network card are configured in sequence; According to the order of retaining data cache, the sensor data and firmware logs cached in the preset persistent cache area are stored in the target data cache area of ​​each smart network card, so that each smart network card can complete initialization in sequence; Among them, data recovery includes configuration register parameter configuration and target data buffer area data storage.

[0085] Specifically, in one embodiment, the management and control module is further configured to: In the case that a common network card is provided in the target computing device, the initialization process of the common network card is subjected to the reserved data filtering process.

[0086] Specifically, in one embodiment, the management and control module is further configured to: Build an anomaly detection model based on the retained data cached in the preset persistent cache area; During the operation of the smart network card, fault prediction of the smart network card is performed based on the anomaly detection model to obtain a fault prediction result of the smart network card.

[0087] For the description of the features in the embodiment corresponding to the intelligent network card management and control device, please refer to the relevant description of the embodiment corresponding to the intelligent network card management and control method, and will not be repeated here.

[0088] The embodiments of the present application also provide a smart network card management and control system for executing the smart network card management and control method provided in the above embodiments.

[0089] like Figure 5 FIG2 is a schematic diagram of the structure of the intelligent network card management and control system provided by an embodiment of the present application. The intelligent network card management and control system includes: an intelligent network card, a bus competition terminal and a management and control device.

[0090] Among them, the smart network card connects the bus competition end and the control device through a preset bus; the bus competition end and the control device access the smart network card by competing for the control right of the preset bus; the control device controls the smart network card based on the smart network card control method provided by the above embodiment. Including the control device adopts BMC, the BMC includes an eMMC storage area, the eMMC storage area is non-volatile, and the eMMC storage area can be used as a preset persistent cache area, such as Figure 5 As shown, the system further includes a CPLD, which is used to provide the BMC with the physical presence status and power supply status of the smart network card.

[0091] For the description of the features in the embodiment corresponding to the intelligent network card management and control system, please refer to the relevant description of the embodiment corresponding to the intelligent network card management and control method, and will not be repeated here.

[0092] The embodiment of the present application also provides an electronic device, such as Figure 6 As shown, it is a structural diagram of an electronic device provided in an embodiment of the present application, including a processor 10 and a memory 20, in which a computer program is stored. The processor 10 is configured to run the computer program to execute the steps in any of the above-mentioned smart network card management and control method embodiments.

[0093] An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program, wherein the computer program is configured to execute the steps of any of the above-mentioned smart network card management and control method embodiments when running.

[0094] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0095] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of any of the above-mentioned smart network card management and control method embodiments are implemented.

[0096] An embodiment of the present application also provides another computer program product, including a non-volatile computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps in any of the above-mentioned smart network card management and control method embodiments.

[0097] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0098] The above is a detailed introduction to the intelligent network card management and control method, device, system and electronic device provided by this application. This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method of this application and its core idea. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of this application, several improvements and modifications can be made to this application, and these improvements and modifications also fall within the scope of protection of the claims of this application.

Claims

1. A method for managing and controlling an intelligent network card, characterized in that: include: When the smart network card in the target computing device meets the preset data retention condition, obtaining temporary data of the smart network card; Cache the temporary data of the smart network card as reserved data in a preset persistent cache area; When the smart network card enters the initialization phase, data recovery is performed on the smart network card based on the retained data cached in the preset persistent cache area, so that the smart network card completes initialization.

2. The intelligent network card management and control method according to claim 1, characterized in that: The method further comprises: When receiving a system restart instruction or a system shutdown instruction of the target computing device, obtaining the physical presence status and power supply status of the smart network card; According to the physical presence status and power supply status of the smart network card, it is determined whether the smart network card meets the preset data retention condition.

3. The intelligent network card management and control method according to claim 2, characterized in that: The determining, based on the physical presence status and power supply status of the smart network card, whether the smart network card meets a preset data retention condition includes: Determining whether the smart network card is in a normal operating state according to the physical presence state and power supply state of the smart network card; When it is determined that the smart network card is in a normal operating state, it is determined that the smart network card meets a preset data retention condition.

4. The intelligent network card management and control method according to claim 1, characterized in that: The method further comprises: Real-time monitoring of the firmware running status information of the smart network card; When the firmware running state information of the smart network card indicates that the smart network card is in a high-load running state, it is determined that the smart network card meets a preset data retention condition.

5. The intelligent network card management and control method according to claim 1, characterized in that: The step of caching the temporary data of the smart network card as reserved data in a preset persistent cache area includes: Performing data encryption processing on the retained data to obtain encrypted retained data; The encrypted retained data is cached in a preset persistent cache area.

6. The intelligent network card management and control method according to claim 1, characterized in that: The obtaining of temporary data of the smart network card includes: Determine whether the preset bus is in the target state; When the preset bus is in the target state, accessing the smart network card through the preset bus to obtain temporary data of the smart network card; The target state includes an idle state.

7. The intelligent network card management and control method according to claim 6, characterized in that: The determining whether the preset bus is in the target state includes: The stage of obtaining access to shared resources from the bus competing end; Determining whether the preset bus is in a target state according to the access phase of the bus competing end to the shared resource; The bus competing end accesses the preset bus through the shared resource.

8. The intelligent network card management and control method according to claim 7, characterized in that: The stage of obtaining access to shared resources by the bus competing end includes: Detecting a control right release signal of the bus contention terminal to obtain a control right release signal detection result; The access phase of the bus contention end to the shared resource is determined according to the control right release signal detection result.

9. The intelligent network card management and control method according to claim 8, characterized in that: The method further comprises: Determine whether the bus competing end has a control right that has timed out and not been released; If the control right of the bus competing end is not released after a timeout, the access of the bus competing end to the shared resource is forcibly suspended so that the preset bus enters the target state.

10. The intelligent network card management and control method according to claim 9, characterized in that: The method further comprises: In the case where the control right of the bus competing end is not released after a timeout, generating a conflict log of the bus competing end; The conflict log at least includes the control right acquisition time, control right occupation time and operation records of the bus competing end during the control right occupation time.

11. The intelligent network card management and control method according to claim 1, characterized in that: When the smart network card enters the initialization phase, restoring data on the smart network card based on the retained data cached in the preset persistent cache area so that the smart network card completes initialization includes: In the case where the smart network card is not unique, when the smart network card enters the initialization phase, obtaining the reserved data cache order corresponding to each smart network card in the preset persistent cache area; According to the reserved data caching order, based on the reserved data cached in the preset persistent cache area, data recovery is performed on each of the smart network cards in sequence, so that each of the smart network cards completes initialization in sequence.

12. The intelligent network card management and control method according to claim 11, characterized in that: The retained data includes at least smart network card configuration information, sensor data, and firmware logs.

13. The intelligent network card management and control method according to claim 12, characterized in that: The method of sequentially restoring data for each of the smart network cards according to the reserved data cache order and based on the reserved data cached in the preset persistent cache area, so that each of the smart network cards completes initialization in sequence, includes: According to the reserved data cache order, and based on the smart network card configuration information cached in the preset persistent cache area, sequentially configure the configuration registers of each smart network card; According to the reserved data cache order, the sensor data and firmware logs cached in the preset persistent cache area are stored in the target data cache area of ​​each smart network card, so that each smart network card is initialized in sequence; The data recovery includes configuration register parameter configuration and target data buffer area data storage.

14. The intelligent network card management and control method according to claim 1, characterized in that: The method further comprises: In the case that a common network card is provided in the target computing device, a retained data filtering process is performed on the initialization process of the common network card.

15. The intelligent network card management and control method according to claim 1, characterized in that: The method further comprises: Building an anomaly detection model based on the retained data cached in the preset persistent cache area; During the operation of the smart network card, fault prediction is performed on the smart network card based on the anomaly detection model to obtain a fault prediction result of the smart network card.

16. An intelligent network card management and control device, characterized in that: include: An acquisition module, configured to acquire temporary data of a smart network card in a target computing device when the smart network card meets a preset data retention condition; A cache module, configured to cache the temporary data of the smart network card as retained data in a preset persistent cache area; The management and control module is used to restore data of the smart network card based on the retained data cached in the preset persistent cache area when the smart network card enters the initialization phase, so that the smart network card completes the initialization.

17. An intelligent network card management and control system, characterized in that: include: Smart network cards, bus competition terminals and control equipment; The smart network card is connected to the bus competition end and the control device via a preset bus; The bus competition terminal and the control device access the smart network card by competing for the control right of the preset bus; The management and control device manages and controls the smart network card based on the smart network card management and control method according to any one of claims 1 to 15.

18. An electronic device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the intelligent network card management and control method according to any one of claims 1 to 15 when executing the computer program.

19. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the intelligent network card management and control method according to any one of claims 1 to 15 are implemented.

20. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the smart network card management and control method according to any one of claims 1 to 15 are implemented.

Images