Server integrated firmware system and task processing method
Through the RISC-V architecture modularization of BIOS and BMC and integration into the same system platform, the inefficiency problem in traditional server design is solved, efficient task processing and resource management is achieved, and the needs of cloud computing and big data processing is adapted.
Patent Information
- Application Number
- CN202410947264.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-07-15
AI Technical Summary
The separate design of BIOS and BMC in traditional server design leads to inefficiency, waste of resources and maintenance complexity, and the communication efficiency between BIOS and BMC is limited, making it difficult to meet the needs of cloud computing and big data processing.
Modularize BIOS and BMC using RISC-V architecture and integrate it into the same embedded system platform. It receives task instructions through integrated control modules, coordinates BIOS and BMC modules to perform tasks, and uses hardware resource library and API interface for resource management. , realize information interaction and communication.
It improves processing speed and system performance, reduces resource reuse and storage requirements, enhances system response speed and execution accuracy, and adapts to cloud computing and big data processing needs.
Smart Images

Figure CN119045986A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of firmware engineering, and in particular to a server integrated firmware system and a task processing method. Background Art
[0002] With the rapid development of cloud computing and big data technology, the demand for server hardware is also changing. Lightweight servers, due to their portability, rapid deployment and temporary nature, have shown great application potential in many fields, especially in military and mobile computing.
[0003] Traditional server designs are configured with separate BIOS and baseboard management controllers (BMCs). Although this separation design provides flexibility to a certain extent, it also has problems such as inefficiency, resource waste, and high maintenance complexity. For example, communication between BIOS and BMC is usually carried out through a low-speed interface, which limits the speed and efficiency of information exchange. In addition, each independent module has its own memory and processing resources, which may lead to resource duplication and waste. Summary of the invention
[0004] The embodiments of the present disclosure provide a server integrated firmware system and a task processing method, aiming to solve the problems existing in the above-mentioned background technology.
[0005] In order to solve the above technical problems, the present disclosure is implemented as follows:
[0006] In a first aspect, an embodiment of the present disclosure provides a server integrated firmware system, the system comprising:
[0007] An integrated control module is used to receive and process task instructions from a target application, call a corresponding BIOS module and / or BMC module to execute the target task according to the task instruction, and return the execution result to the target application;
[0008] The embedded system platform integrates multiple BIOS modules and BMC modules. The BIOS module and the BMC module are obtained by modularizing the BIOS and the BMC according to task functions through a processor of the RISC-V architecture.
[0009] Optionally, the system further comprises a hardware resource library and an API interface, wherein the hardware resource library is used to integrate and manage hardware resources on the server, and the BIOS module and the BMC module access and operate the hardware resource library through the API interface;
[0010] The hardware resource library is integrated with a resource manager, and the resource manager is used to monitor the operating load, hardware performance characteristics and user-defined policies of the server, and allocate and manage the resources of the hardware resource library.
[0011] Optionally, the embedded system platform is configured with a plurality of processor modules, and the processor modules are obtained by modularizing a processor of a RISC-V architecture according to processor functions;
[0012] According to the task requirements of the BIOS module and the BMC module, a processor module suitable for the task requirements is selected and combined from a plurality of processor modules.
[0013] Optionally, the system further includes a management interface and a user interface, wherein the management interface integrates the task functions of the BIOS module and the BMC module, and the task functions include but are not limited to at least one of the following: hardware configuration management, device driver loading, firmware update, system status monitoring, and fault warning;
[0014] The various task functions of the BIOS module and the BMC module are displayed through the user interface.
[0015] Optionally, the embedded system platform is configured with a message queue for providing a channel for information interaction between the BIOS module and the BMC module;
[0016] When the BIOS is started, the BIOS module obtains the real-time hardware status information in the message queue by querying the message queue, and sends the BIOS operation information to the message queue, wherein the real-time hardware status information is obtained by the BMC module and sent to the message queue, and the BIOS operation information includes the startup progress and error log of the BIOS;
[0017] The embedded system platform is configured with an event notification mechanism for real-time communication between the BIOS module and the BMC module;
[0018] The event notification mechanism is configured to obtain real-time hardware status information of the corresponding resources when the BMC module detects a change in the status of a resource in the hardware resource library or receives a management instruction for a resource from a user, and send the real-time hardware status information to the BIOS module through the message queue.
[0019] Optionally, the embedded system platform is configured with a message scheduling algorithm;
[0020] The message scheduling algorithm is configured to dynamically sort the information in the message queue based on the priority of each message in the message queue, and to schedule and process information with higher priority when executing the target task. The priority of the information in the message queue is defined based on the nature of the information and the degree of impact on the system.
[0021] Optionally, the embedded system platform is configured with a machine learning algorithm, which is configured to input the current real-time hardware status information in the message queue into a historical training model, and predict the change trend of the real-time hardware status information based on the output of the historical training model. The historical training model is pre-trained using the real-time hardware status information historically collected in the message queue.
[0022] Optionally, the system further comprises a fast boot module, wherein the fast boot module is integrated with a Boot boot function and is used to configure the boot process of the BIOS as a Linux Boot kernel boot process;
[0023] The fast boot module is configured with a fast memory mapping mechanism, and the fast memory mapping mechanism is configured to map a file with a large amount of data required in the startup process of the BIOS to a memory address space;
[0024] The system further comprises a network remote boot module, which is used to provide a remote wake-up function, send a remote wake-up request through a network, and remotely manage the configuration of the fast boot module through a network.
[0025] Optionally, the system further comprises a wake-up control module, configured to monitor wake-up requests from different sources on the server;
[0026] The wake-up control module is configured with a wake-up decision algorithm, which is configured to determine the current power state of the system based on the current power state and load conditions of the system. When the current power state of the system is a low power state, the service indicated by the wake-up request with a higher priority is preferentially awakened. The priority of the wake-up request is defined based on the urgency of the wake-up request and the degree of impact on the system.
[0027] In a second aspect, an embodiment of the present disclosure provides a task processing method, which is applied to a server integrated firmware system, and the method includes:
[0028] receiving task instructions from a target application;
[0029] Calling the corresponding BIOS module and / or BMC module to execute the target task according to the task instruction, and allocating resources required for executing the target task to the BIOS module and / or BMC module, wherein the BIOS module and the BMC module are obtained by modularizing the BIOS and the BMC according to the task functions through a processor with a RISC-V architecture;
[0030] The execution result of the target task is returned to the target application.
[0031] The technical solution provided by the embodiments of the present disclosure brings at least the following beneficial effects:
[0032] The present disclosure utilizes the RISC-V architecture processor to modularize the BIOS and BMC, so that the functionalized modules can perform task processing and information exchange more efficiently, which not only improves the processing speed, but also reduces the limitations of low-speed interface communication, thereby enhancing the overall performance of the system. By modularly integrating the BIOS and BMC functions onto the same embedded system platform, duplicate resource usage and storage requirements are reduced. In addition, the introduction of the integrated control module allows task instructions to be directly received from the target application, and coordinates the BIOS module and the BMC module to perform tasks. The integrated firmware design of the present disclosure improves the response speed of instruction processing and the accuracy of execution, allowing the server to adapt to the needs of cloud computing and big data processing more quickly. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0034] Figure 1 is a structural block diagram of a server integrated firmware system provided by an embodiment of the present disclosure;
[0035] Figure 2 It is a schematic diagram of the design architecture of a server integrated firmware system provided by an embodiment of the present disclosure;
[0036] Figure 3 It is a step diagram of a task processing method provided by an embodiment of the present disclosure. DETAILED DESCRIPTION
[0037] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.
[0038] In the context of the rapid development of modern information technology, the efficient operation of server hardware systems has become the key to improving computing power and data processing speed. In order to meet these needs, the embodiments of the present disclosure provide a server integrated firmware system, which aims to significantly improve the performance and resource management efficiency of the server through a highly integrated and modular design. Figure 1is a structural block diagram of a server integrated firmware system provided by an embodiment of the present disclosure, such as Figure 1 As shown, the system comprises:
[0039] The integrated control module 101 is used to receive and process task instructions from a target application, call a corresponding BIOS module and / or BMC module to execute a target task according to the task instructions, and return the execution result to the target application.
[0040] The integrated control module is mainly used to receive and process task instructions from the target application in the system, and execute the target task by calling the corresponding BIOS module and / or BMC module, and finally return the execution result to the target application. The target application refers to various applications or services running on the operating system. The target application can be a user interaction interface, a system management tool, and a server monitoring software. The target application sends a task instruction to the integrated control module to perform a specific hardware management or system operation task. The task instruction is a specific operation request sent by the target application to the integrated control module, which contains the specific tasks to be performed and their parameters, such as adjusting the fan speed, monitoring the processor temperature, updating the firmware, etc. The task instruction is a clear operation command used to instruct the integrated control module which BIOS modules and / or BMC modules need to be called to complete a specific task. The target task refers to the specific operation that needs to be performed according to the task instruction. The target task may include but is not limited to: hardware initialization, system monitoring, hardware configuration adjustment, and firmware update. After receiving the task instruction, the integrated control module parses the instruction content of the task instruction, determines the BIOS module (such as boot sequence configuration, hardware initialization) and / or BMC module (such as system monitoring, fan control) that needs to be called according to the parsing result, and calls the corresponding BIOS module and / or BMC module. The BIOS module and / or BMC module executes the target task and returns the execution result of the target task, such as operation success, failure or specific monitoring data, to the integrated control module. The integrated control module organizes the results and returns them to the target application for further processing.
[0041] The embedded system platform 102 integrates multiple BIOS modules and BMC modules. The BIOS modules and the BMC modules are obtained by modularizing the BIOS and the BMC according to task functions through a processor of the RISC-V architecture.
[0042] RISC-V is an open source instruction set architecture (ISA), and its design goal is to provide a simple, flexible, scalable and free-to-use architecture. The present disclosure selects the RISC-V architecture to implement the integration of server firmware, mainly based on the following considerations: First, the modular design of RISC-V allows the selection and combination of different processor modules according to specific needs, such as integer operation modules, floating-point operation modules, storage interface modules, etc., to meet various needs from low-power embedded systems to high-performance computing servers; Second, the reduced instruction set design of RISC-V improves the operating efficiency of the system and can provide efficient computing capabilities; Third, the open source nature of RISC-V avoids licensing fees and reduces costs. According to different task functions, the BIOS and BMC are modularly designed to obtain multiple BIOS modules and BMC modules. The flexibility and scalability of the RISC-V processor are used to integrate the above BIOS modules and BMC modules into a unified embedded system platform. Optionally, advanced microcontrollers or SOCs can also be used for fusion. In addition, in order to give full play to the advantages of RISC-V processors and provide support for operating systems, the present invention modifies and optimizes the RISC-V architecture, including kernel scheduling and memory management, to improve the performance stability and response speed of the operating system.
[0043] BIOS (Basic Input / Output System) is responsible for hardware initialization, hardware detection and system booting when the computer starts. After BIOS modularization, each BIOS module is responsible for specific tasks, such as hardware initialization module, boot loading module, etc. On the RISC-V architecture, the BIOS module can be modularly designed according to different task functions, such as memory initialization, device detection, boot loading, etc. BMC (Baseboard Management Controller) is responsible for server hardware management, monitoring and remote management. Similarly, after BMC modularization, each module is responsible for specific management tasks, such as hardware monitoring module, fan control module, remote management module, etc. On the RISC-V architecture, the BMC module can also be modularly designed according to different task functions, such as hardware monitoring, system alarm, remote control, etc. By integrating BIOS and BMC modules into the embedded system platform on the RISC-V architecture, the two modules can share memory space and peripheral resources to improve system resource utilization. The integrated design also makes the system architecture more compact, effectively reduces the complexity of hardware layout and resource waste, and improves the overall system efficiency. It should be noted that since the task functions of BIOS and operating system OS are highly overlapped, modularizing the functions of BIOS is equivalent to integrating the operating system as well. Therefore, the embodiment of the present disclosure is equivalent to providing a deeply integrated design solution of BIOS, BMC and operating system OS based on RISC-V architecture.
[0044] The present disclosure utilizes the RISC-V architecture processor to modularize the BIOS and BMC, so that the functionalized modules can perform task processing and information exchange more efficiently, which not only improves the processing speed, but also reduces the limitations of low-speed interface communication, thereby enhancing the overall performance of the system. By modularly integrating the BIOS and BMC functions onto the same embedded system platform, duplicate resource usage and storage requirements are reduced. In addition, the introduction of the integrated control module allows task instructions to be directly received from the target application, and coordinates the BIOS module and the BMC module to perform tasks. The integrated firmware design of the present disclosure improves the response speed of instruction processing and the accuracy of execution, allowing the server to adapt to the needs of cloud computing and big data processing more quickly.
[0045] Exemplarily, the system also includes a hardware resource library and an API interface, wherein the hardware resource library is used to integrate and manage hardware resources on the server, and the BIOS module and the BMC module access and operate the hardware resource library through the API interface; the hardware resource library is integrated with a resource manager, and the resource manager is used to monitor the operating load, hardware performance characteristics and user-defined policies of the server, and allocate and manage the resources of the hardware resource library.
[0046] In order to achieve efficient management and operation of the server, the present invention designs a unified resource library and API interface, so that the BIOS module and the BMC module can access and operate the same set of hardware resource libraries. The unified resource library is a global resource pool that contains server hardware configuration information, device drivers, firmware updates, etc. The unified resource library integrates and manages all hardware resources of the server, such as processors, memory, storage devices, and network interfaces. The API (Application Programming Interface) is a unified interface. Through the API interface, the BIOS module and the BMC module can access and operate the resources in the unified resource library. The API interface provides a standardized access path for the BIOS module and the BMC module, simplifying the development and maintenance process.
[0047] Under the architecture of a unified resource library, BMBIOS (for ease of expression and understanding, the disclosed embodiment defines the functional integration of the BIOS module and the BMC module as BMBIOS) is not only responsible for initializing the hardware and booting the operating system, but also acts as a manager of hardware resources. When the system starts, BMBIOS initializes the processor, memory and other key hardware components, loads and boots the operating system. BMBIOS comprehensively manages various hardware resources of the system through the resource manager. The resource manager can monitor the usage of hardware resources in real time, and intelligently adjust resource allocation according to system load, hardware performance characteristics and user policies. When a target application requires more computing resources, the resource manager will temporarily allocate idle processor cores to the target application to ensure its smooth operation. When the system load is reduced, the resource manager will recycle resources in a timely manner for use by other applications or system services. When the system memory usage is close to the upper limit, the resource manager can ensure the stable operation of the system by optimizing memory allocation and releasing unused memory blocks.
[0048] Exemplarily, the embedded system platform is configured with multiple processor modules, and the processor modules are obtained by modularizing a processor of a RISC-V architecture according to processor functions; according to the task requirements of the BIOS module and the BMC module, a processor module suitable for the task requirements is selected and combined from the multiple processor modules.
[0049] Similar to the functional modularization of BIOS and BMC, the present disclosure utilizes the modular design of RISC-V to combine different functional modules of the processor to form a system platform with multiple processor modules. As mentioned above, a notable feature of the RISC-V architecture is its modular design. This design allows different processor modules to be selected and combined according to actual needs, including but not limited to the following: integer operation module, responsible for basic arithmetic operations, such as integer calculations such as addition, subtraction, multiplication and division; floating-point operation module, processing floating-point operations, suitable for application scenarios requiring high-precision calculations, such as scientific computing and graphics processing; storage interface module, managing interfaces with memory and storage devices, and implementing data read and write operations; accelerator module, optimized for specific tasks, such as image processing, encryption and decryption, etc. For example, in a low-power embedded system, only integer operation modules and storage interface modules can be selected to reduce power consumption and cost, while in a high-performance computing server, floating-point operation modules and multiple accelerator modules can be selected to improve computing power and efficiency.
[0050] Exemplarily, the system also includes a management interface and a user interface, wherein the management interface integrates the task functions of the BIOS module and the BMC module, and the task functions include but are not limited to at least one of the following: hardware configuration management, device driver loading, firmware update, system status monitoring, and fault warning; the various task functions of the BIOS module and the BMC module are displayed through the user interface.
[0051] The disclosed embodiment also provides a unified management interface that integrates the functions of BIOS and BMC, and displays and operates through a unified user interface, thereby achieving convenience and intuitiveness in system management. Specifically, the unified management interface needs to deeply integrate the various functional modules of the BIOS module and the BMC module in the underlying design to ensure that they can exchange information efficiently and accurately. These task functions include but are not limited to the following: hardware configuration management, device driver loading, firmware update, system status monitoring, and fault warning. The user interface is designed to provide users with an intuitive and easy-to-use system management platform. In the user interface, the main interface displays an overview of the system, such as CPU usage, memory occupancy, and system temperature. Users can enter the detailed management page by clicking the corresponding module to view or operate the various task functions of the BIOS module and the BMC module.
[0052] Exemplarily, the embedded system platform is configured with a message queue for providing a channel for information interaction between the BIOS module and the BMC module; when the BIOS is started, the BIOS module obtains real-time hardware status information in the message queue by querying the message queue, and sends BIOS operation information to the message queue, wherein the real-time hardware status information is obtained by the BMC module and sent to the message queue, and the BIOS operation information includes the startup progress and error log of the BIOS.
[0053] The message queue mechanism provides an orderly, cross-module information exchange channel between the BIOS module and the BMC module. During the BIOS startup process, by querying the message queue, you can obtain real-time hardware status information sent by the BMC module. At the same time, the BIOS module can also send information such as the startup progress and error log to the message queue for the BMC module to analyze and process. When the BIOS is started, obtaining real-time hardware status information is crucial for system initialization and configuration. This information includes but is not limited to processor temperature, fan speed, and memory usage. For example, when the BIOS is started, the BIOS module queries the processor temperature information provided by the BMC module through the message queue. If the temperature is too high, the corresponding BIOS module can take corresponding measures, such as adjusting the fan speed or reducing the processor frequency, to avoid system overheating. During the BIOS startup process, BIOS operation information such as startup progress and error logs will be generated. The BIOS operation information is sent to the corresponding BMC module through the message queue, thereby helping the BMC understand the startup status of the BIOS and provide support when necessary.
[0054] The embedded system platform is configured with an event notification mechanism for real-time communication between the BIOS module and the BMC module; the event notification mechanism is configured to obtain real-time hardware status information of the corresponding resources when the BMC module detects that the status of the resources in the hardware resource library has changed, or receives a management instruction for the resources from the user, and sends the real-time hardware status information to the BIOS module through the message queue.
[0055] In embedded system platforms, the event notification mechanism provides an efficient channel for real-time communication between BIOS modules and BMC modules. The event notification mechanism ensures that when BMC detects a change in hardware status or receives a user's management instruction, it can quickly notify the relevant information to the BIOS module of the relevant function, so that the BIOS module can respond in time, adjust the system configuration or perform specific tasks. When BMC detects a change in the status of certain resources in the hardware resource library, it will send the real-time hardware status information of the corresponding resources to the BIOS through the message queue through the event notification mechanism. Suppose BMC detects that the processor temperature suddenly rises to a critical value. BMC will immediately trigger an event notification and send this temperature information to the BIOS module of the relevant function. After receiving the notification, BIOS can take corresponding measures, such as increasing the fan speed or reducing the processor load, to ensure that the system runs within a safe temperature range. When BMC receives management instructions for hardware resources from users, such as system restart, firmware update, fan speed adjustment, etc., it passes the real-time hardware status information of the corresponding resources to the BIOS module of the relevant function through the message queue through the event notification mechanism, so that BIOS can perform the operations indicated by the execution management instructions.
[0056] Exemplarily, the embedded system platform is configured with a message scheduling algorithm; the message scheduling algorithm is configured to dynamically sort the information in the message queue based on the respective priorities of each information in the message queue, and when executing the target task, schedule and process information with higher priority. The priority of the information in the message queue is defined according to the nature of the information and the degree of impact on the system.
[0057] Each message in the message queue has its own priority. The priority is defined based on the nature of the message and the extent of its impact on the system. For example: the hardware temperature alarm is high priority; the CPU usage report is medium priority; the periodic status check is low priority. The messages in the message queue are dynamically sorted according to their priority, so that when the system processes these messages, it always gives priority to those high-priority messages. The dynamic sorting algorithm can be implemented using efficient sorting algorithms such as heap sort and quick sort. When the embedded system platform needs to perform the target task, the message scheduling algorithm will schedule and process the message according to the priority of the message. Assume that there are three messages in the message queue:
[0058] Message A, hardware overtemperature alarm (high priority);
[0059] Message B, CPU usage report (medium priority);
[0060] Message C, regular status check (low priority); Assuming that the system receives a new message D captured by the BMC module during the process of processing message A, power supply voltage abnormality alarm (highest priority), the message scheduling algorithm will immediately insert message D to the front of the message queue to ensure that it is processed first. The queue order becomes: message D> message A> message B> message C. Ensure that the most important messages can be processed first when system resources are limited.
[0061] Exemplarily, the embedded system platform is configured with a machine learning algorithm, which is configured to input the current real-time hardware status information in the message queue into a historical training model, and predict the change trend of the real-time hardware status information based on the output of the historical training model. The historical training model is pre-trained using the real-time hardware status information historically collected in the message queue.
[0062] In the embedded system platform, the machine learning algorithm is configured to process the real-time hardware status information in the message queue. The machine learning algorithm predicts the change trend of the system status by analyzing the current real-time data and historical data, thereby optimizing the decision-making process and response capability of the system. First, a historical training model needs to be built in advance. The historical training model is obtained by training the real-time hardware status information collected historically in the message queue. Optionally, the real-time hardware status information in the message queue is first collected, and then the collected data is cleaned to remove noise and invalid data to ensure data quality. Then useful features are extracted from the original data. For example, the rate of temperature change, the fluctuation pattern of the CPU load, etc. Finally, the extracted feature data is used to train the machine learning algorithm to generate a trained historical training model. Assume that the system has collected the following data in the past period of time: at a certain moment, the hardware temperature is 50°C, the CPU utilization rate is 70%, and then the temperature rises to 60°C. At another moment, the hardware temperature is 55°C, the CPU utilization rate is 75%, and then the temperature rises to 65°C. Through the historical training model, the relationship between temperature and CPU utilization is identified, and the future temperature change trend is predicted. During system operation, the machine learning algorithm inputs the current real-time hardware status information in the message queue into the trained historical model. Assume that at the current moment, the following two pieces of real-time hardware status information are added to the message queue: the hardware temperature is 57°C and the CPU usage is 72%. These two pieces of real-time hardware status information are input into the trained model. The historical training model predicts the change trend of the real-time hardware status information based on the input real-time hardware status information and historical data, thereby predicting the change trend of the system status. This prediction result can help the system respond and adjust in advance. Based on the input real-time data, the model predicts that the hardware temperature may rise to 65°C in the next 10 minutes. If the temperature exceeds a certain critical value, it may cause the system to overheat and affect normal operation. Based on the prediction results, the system can take measures in advance. For example, reduce the CPU load, increase the fan speed, send warning information, etc. to avoid potential problems.
[0063] The machine learning algorithm configured in the embedded system platform realizes intelligent management of the system by analyzing and predicting real-time hardware status information and historical data. It can identify potential risks in advance, optimize the allocation and use of system resources, and improve the overall performance and stability of the system.
[0064] Exemplarily, the system also includes a quick boot module, which integrates a Boot boot function and is used to configure the BIOS startup process as a Linux Boot kernel startup process; the quick boot module is configured with a fast memory mapping mechanism, and the fast memory mapping mechanism is configured to map files with large amounts of data required in the BIOS startup process to the memory address space; the system also includes a network remote boot module, which is used to provide a remote wake-up function, send a remote wake-up request through the network, and remotely manage the configuration of the quick boot module through the network.
[0065] The fast boot module integrates the basic boot functions of the traditional BIOS and deeply integrates the boot process into the LinuxBoot kernel boot process. Traditional BIOS usually performs a series of self-check steps (such as memory test, hardware initialization, etc.), which may take a long time in some cases. In order to optimize the boot time, the fast boot module removes unnecessary self-check steps and directly loads the Linux kernel to initialize the system and start the operating system. The fast memory mapping mechanism is designed to speed up the loading of large data files during the boot process. During the boot process, a large amount of configuration files, driver programs and other data need to be read and loaded. By directly mapping this data to the memory address space, the data access speed can be significantly improved and the boot time can be reduced. Suppose a 100MB operating system image file needs to be loaded during the boot process. In the traditional way, it may be necessary to read the disk multiple times, but through the fast memory mapping mechanism, the file can be mapped to the memory at one time, reducing disk IO operations and improving loading efficiency.
[0066] The network remote boot module uses the management capabilities of the BMC to provide a remote wake-up function. Administrators can send wake-up commands through the LAN or the Internet to trigger the wake-up operation of the remote computer. This is of great significance for the management of large-scale servers. Administrators can remotely control the power on and off of the server without going to the site in person. The network remote boot module also provides remote configuration management functions, allowing administrators to update boot configurations, load new system images, etc. through the network. This makes system maintenance and updates more flexible and efficient. Suppose an administrator needs to update the operating system image files of multiple servers. If using traditional methods, manual operations need to be performed on each server one by one. With the remote configuration management function, administrators can update the image files of all servers on the network at one time, significantly improving work efficiency.
[0067] Through the combination of the quick boot module and the network remote boot module, the system can not only achieve quick startup, but also be flexibly managed remotely through the network. This integration not only improves the system's response speed, but also provides administrators with convenient management tools, improving the reliability and security of the overall system.
[0068] Exemplarily, the system also includes a wake-up control module for monitoring wake-up requests from different sources on the server; the wake-up control module is configured with a wake-up decision algorithm, and the wake-up decision algorithm is configured to determine the current power state of the system based on the current power state and load conditions of the system. When the current power state of the system is a low power state, the service indicated by the wake-up request with a higher priority is preferentially awakened. The priority of the wake-up request is defined according to the urgency of the wake-up request and the degree of impact on the system.
[0069] The agile wake-up control module is responsible for monitoring wake-up requests from different sources on the server, including hard disk, PCIE device, keyboard, mouse, network, etc. This means that whether it is the activity of hardware devices, user input operations, or network commands, the system can be triggered to wake up. The wake-up control module ensures priority wake-up of key services by managing the priority of wake-up requests. Different wake-up requests are given different priorities according to their urgency and the degree of impact on the system. When multiple wake-up requests arrive at the same time, the wake-up control module will decide which request to respond to first based on the priority of the wake-up requests. Assume that the server receives two wake-up requests: one is the movement of the user's mouse (lower priority) and the other is an important task on the hard disk (higher priority). The agile wake-up control module will prioritize the important task on the hard disk to ensure that the key functions of the system are restored first. The wake-up decision algorithm intelligently determines whether to perform the wake-up operation based on factors such as the current state of the system and the power management strategy. The wake-up decision algorithm evaluates the current power state and system load and decides under what circumstances to perform the wake-up operation. When the system is in a low power state, the wake-up decision algorithm will prioritize the wake-up requests with higher priorities. This is to ensure that the system can continue to perform the most important tasks under limited power conditions. Assume that the server is low on battery power but receives two wake-up requests at the same time: one is a system update notification (lower priority) and the other is a remote work instruction from the network (higher priority). The wake-up decision algorithm will prioritize the remote work instruction from the network based on the current low battery status to ensure that important work tasks can be executed in time.
[0070] Exemplarily, the system further includes a fault recovery module, which is configured to automatically detect faults of the BIOS module and the BMC module and execute an automatic repair process. The automatic repair process includes but is not limited to restoring data from a backup, restarting the corresponding module, and reconfiguring parameter settings of the faulty module. The fault recovery module interacts with the hardware resource library through the API interface to ensure data integrity and continuous operation of the system in the event of a hardware fault or a software fault.
[0071] The fault recovery module monitors the status of the BIOS module and the BMC module in real time. Once an abnormality or failure is detected, it can quickly take measures to repair or recover, thereby minimizing the risk of system downtime and data loss. The fault recovery module works closely with the API interface and hardware resource library in the system, using the information in the hardware resource library and the functions provided by the resource manager to perform fault detection and recovery operations. For example, when a BIOS module failure is detected, the fault recovery module can instruct the resource manager to call the backup data for recovery, or reconfigure the hardware settings to adapt to the current operation requirements.
[0072] In addition, the fault recovery module is also configured to automatically restart the failed module or adjust the configuration settings to ensure that the system can be restored to an optimal state in the shortest time.
[0073] By integrating the fault recovery module into the server integrated firmware system, the system's response speed and processing capabilities to faults can be greatly improved, thereby ensuring the high availability and business continuity of the system, helping to meet the growing reliability and automation needs of data centers, and providing important guarantees for the stable operation of the server integrated firmware system.
[0074] Figure 2 This is a schematic diagram of the design architecture of a server integrated firmware system provided by an embodiment of the present disclosure. As mentioned above, the present disclosure first provides the lowest-level software design, a modular design based on the RISC-V architecture. Its design goal is to provide a simple, flexible, scalable and free-to-use architecture. The modular design of the core processor allows different processor modules to be selected and combined according to specific needs. The modular design of the RISC-V architecture is also reflected in its scalability. Appropriate processor modules can be selected according to different application scenarios to meet various needs from low-power embedded systems to high-performance computing servers. The present disclosure adopts RISC-V as the core processor of the system, and uses its reduced instruction set design and high flexibility to provide the system with efficient computing power.
[0075] Then there is the integrated design, which is the underlying hardware design of BIOS, BMC and OS. In order to give full play to the advantages of RISC-V processors, the present disclosure implements a new architecture design, deeply integrating BIOS, BMC and operating system. By modularizing BIOS and BMC and integrating these modules on the RISC-V architecture, more efficient task processing and information exchange can be achieved. The present disclosure designs a set of unified resource libraries and API interfaces, so that the BIOS module and the BMC module can access and operate the same set of hardware resource libraries. The hardware resource library contains server hardware configuration information, device drivers, firmware updates and other contents, and the API interface provides a standardized access path, which simplifies the development and maintenance process. The integrated control module is responsible for coordinating the task execution of the BIOS module and the BMC module, and communicates through a unified API interface. Not only does it reduce the waste of hardware resources, but it also improves the utilization of system resources and the accuracy of task execution. The present disclosure deeply integrates BIOS, BMC and the operating system OS into the RISC-V architecture to achieve a more compact and efficient system architecture and form a highly integrated system solution.
[0076] Then the present disclosure provides a unified internal and external management interface design, which is a user-oriented integrated design. The unified management interface integrates the functions of BIOS and BMC, and provides users with a convenient management platform through standardized interfaces and protocols. The unified user interface design provides users with an intuitive and easy-to-use system management platform. In the present disclosure, a unified management interface is crucial to improving the efficiency and convenience of system management. The management interface is designed to not only integrate BIOS and BMC at the bottom layer, but also integrate management functions into a unified user interface, thereby providing users with a more intuitive and more convenient system management and maintenance experience.
[0077] Finally, the agile system wake-up function design is a software system integration design. When the system starts, the fast boot module optimizes the boot process and hardware initialization steps to achieve fast startup. This can shorten the system startup time and improve the user experience. The wake-up control module monitors the system status and receives external wake-up signals, and quickly responds and restores the system operation when the system is in sleep or low power consumption state. The network remote boot module allows the system to receive boot instructions through the network to achieve remote startup and management. It is particularly important for server management in data centers and cloud computing environments, and can significantly improve management efficiency and flexibility. BMBIOS integrates the boot function and OS startup function of traditional BIOS, and no longer distinguishes the handover of the boot program between the two. Linux Boot is used directly to start the whole machine, and the Linux kernel is directly used to initialize the computer system and load the operating system. In addition, BMBIOS also has the management capability of BMC, and can remotely boot the computer system through the network.
[0078] Figure 3 is a step diagram of a task processing method provided by an embodiment of the present disclosure, wherein the method is applied to the server firmware integration system as described above. Figure 3 As shown, the method includes:
[0079] Step S201: receiving a task instruction from a target application.
[0080] Step S202, calling the corresponding BIOS module and / or BMC module to execute the target task according to the task instruction, and allocating resources required to execute the target task to the BIOS module and / or BMC module, wherein the BIOS module and the BMC module are obtained by modularizing the BIOS and the BMC according to the task functions through a processor of the RISC-V architecture.
[0081] Step S203: Return the execution result of the target task to the target application.
[0082] In step S201, the integrated control module of the server integrated firmware system receives a task instruction from the target application. The target application sends a specific operation request to the integrated control module, which includes the specific task to be executed and its parameters.
[0083] Involving step S202, according to different task functions, the BIOS and BMC are modularly designed to obtain multiple BIOS modules and BMC modules, and the flexibility and scalability of the RISC-V processor are utilized to integrate the above BIOS modules and BMC modules into a unified embedded system platform. In this step, according to the received task instruction, the integrated control module parses the instruction content and determines the specific BIOS module and / or BMC module and its function that need to be called. The integrated control module disclosed in the present invention then calls the corresponding modules to perform the target task and allocates the resources required to perform the task to these modules. Assuming that the target application sends a task instruction for the target task, requiring the processor temperature of a server to be monitored, the task instruction includes the identification of the target server and the specific parameters of the monitoring task. After the integrated control module parses the task instruction, it determines that the hardware monitoring function in the BMC module needs to be called, and allocates the necessary resources (such as memory, computing resources) to the BMC module to perform the temperature monitoring task.
[0084] In step S203, after the BIOS module and / or the BMC module completes the target task, the execution result of the task (such as operation success, failure or specific monitoring data) is returned to the integrated control module. The integrated control module organizes these results and returns them to the target application that sent the task instruction. Continuing with the previous example, after the BMC module completes the temperature monitoring task, it returns the temperature data of the processor to the integrated control module. The integrated control module organizes and returns these data to the target application.
[0085] The task processing method provided by the embodiment of the present disclosure significantly improves the performance and resource management efficiency of the server through the process of receiving task instructions, calling the corresponding module to execute the task, and returning the execution result. The combination of the integrated control module, the RISC-V architecture processor and its modular design enables the system to flexibly respond to various task requirements and provides an efficient solution for cloud computing and big data processing.
[0086] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as systems and methods. Therefore, the embodiments of the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present disclosure may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0087] The embodiments of the present disclosure are described with reference to the flowcharts and / or block diagrams of the systems and methods according to the embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing terminal device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1These computer program instructions can also be loaded into a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0088] Although the preferred embodiments of the present disclosure have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present disclosure.
[0089] Finally, it should be noted that, in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "include..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements. The above describes in detail a server integrated firmware system and a task processing method provided by the present disclosure. The present disclosure is described in detail using specific examples in this article to explain the principles and implementation methods of the present disclosure. The description of the above embodiments is only used to help understand the method and its core idea of the present disclosure; at the same time, for those of ordinary skill in the art, according to the idea of the present disclosure, there will be changes in the specific implementation method and scope of application. In summary, the content of this specification should not be understood as a limitation on the present disclosure.
Claims
1. A server integrated firmware system, characterized in that: The system comprises: An integrated control module is used to receive and process task instructions from a target application, call a corresponding BIOS module and / or BMC module to execute the target task according to the task instruction, and return the execution result to the target application; The embedded system platform integrates multiple BIOS modules and BMC modules. The BIOS module and the BMC module are obtained by modularizing the BIOS and the BMC according to task functions through a processor of the RISC-V architecture.
2. The system according to claim 1, characterized in that The system further comprises a hardware resource library and an API interface, wherein the hardware resource library is used to integrate and manage hardware resources on the server, and the BIOS module and the BMC module access and operate the hardware resource library through the API interface; The hardware resource library is integrated with a resource manager, and the resource manager is used to monitor the operating load, hardware performance characteristics and user-defined policies of the server, and allocate and manage the resources of the hardware resource library.
3. The system according to claim 1, characterized in that The embedded system platform is configured with a plurality of processor modules, and the processor modules are obtained by modularizing a processor of a RISC-V architecture according to processor functions; According to the task requirements of the BIOS module and the BMC module, a processor module suitable for the task requirements is selected and combined from a plurality of processor modules.
4. The system according to claim 1, characterized in that The system further includes a management interface and a user interface, wherein the management interface integrates the task functions of the BIOS module and the BMC module, and the task functions include but are not limited to at least one of the following: hardware configuration management, device driver loading, firmware update, system status monitoring, and fault warning; The various task functions of the BIOS module and the BMC module are displayed through the user interface.
5. The system according to claim 2, characterized in that The embedded system platform is configured with a message queue for providing a channel for information interaction between the BIOS module and the BMC module; When the BIOS is started, the BIOS module obtains the real-time hardware status information in the message queue by querying the message queue, and sends the BIOS operation information to the message queue, wherein the real-time hardware status information is obtained by the BMC module and sent to the message queue, and the BIOS operation information includes the startup progress and error log of the BIOS; The embedded system platform is configured with an event notification mechanism for real-time communication between the BIOS module and the BMC module; The event notification mechanism is configured to obtain real-time hardware status information of the corresponding resources when the BMC module detects a change in the status of a resource in the hardware resource library or receives a management instruction for a resource from a user, and send the real-time hardware status information to the BIOS module through the message queue.
6. The system according to claim 5, characterized in that The embedded system platform is configured with a message scheduling algorithm; The message scheduling algorithm is configured to dynamically sort the information in the message queue based on the priority of each message in the message queue, and to schedule and process information with higher priority when executing the target task. The priority of the information in the message queue is defined based on the nature of the information and the degree of impact on the system.
7. The system according to claim 5, characterized in that The embedded system platform is configured with a machine learning algorithm, which is configured to input the current real-time hardware status information in the message queue into a historical training model, and predict the change trend of the real-time hardware status information based on the output of the historical training model. The historical training model is pre-trained using the real-time hardware status information historically collected in the message queue.
8. The system according to claim 1, characterized in that The system further comprises a fast boot module, wherein the fast boot module is integrated with a Boot boot function and is used to configure the boot process of the BIOS as a Linux Boot kernel boot process; The fast boot module is configured with a fast memory mapping mechanism, and the fast memory mapping mechanism is configured to map a file with a large amount of data required in the startup process of the BIOS to a memory address space; The system further comprises a network remote boot module, which is used to provide a remote wake-up function, send a remote wake-up request through a network, and remotely manage the configuration of the fast boot module through a network.
9. The system according to claim 1, characterized in that The system also includes a wake-up control module for monitoring wake-up requests from different sources on the server; The wake-up control module is configured with a wake-up decision algorithm, which is configured to determine the current power state of the system based on the current power state and load conditions of the system. When the current power state of the system is a low power state, the service indicated by the wake-up request with a higher priority is preferentially awakened. The priority of the wake-up request is defined based on the urgency of the wake-up request and the degree of impact on the system.
10. A task processing method, characterized in that: Applied to the system according to any one of claims 1 to 9, the method comprises: receiving task instructions from a target application; Calling the corresponding BIOS module and / or BMC module to execute the target task according to the task instruction, and allocating resources required for executing the target task to the BIOS module and / or BMC module, wherein the BIOS module and the BMC module are obtained by modularizing the BIOS and the BMC according to the task functions through a processor with a RISC-V architecture; The execution result of the target task is returned to the target application.
Citation Information
Patent Citations
Equipment starting method and device, computer equipment and storage medium
CN112130910A
Integrated development method for RISC-V architecture processor
CN117435171A
System and method for unified firmware managment
US20170131991A1
Cited By
Resource scheduling control method and system for embedded device
CN120123104A
System starting method, device and equipment based on RISC-V server and medium
CN121217720A
