A mobile base station computing power management method and related device

Abstract

The present application discloses a mobile base station computing power management method and related devices, which relate to the field of software technology, monitor the system load of the mobile base station, and predict the load change trend of the mobile base station based on the historical system load closest to the current moment; if the load change trend is no-load, determine the first computing task to be executed, allocate the first container resource matching its task type to the first computing task in the resource pool of the mobile base station, and execute the container instance of the first computing task based on the first container resource; if the load change trend is overload, determine the second computing task in execution, release the second container resource occupied by the second computing task, and recycle the computing power resources of the resource pool by adjusting the capacity size of the resource pool. Under the premise of ensuring the normal operation of communication services, the present application maximizes the use of idle computing power resources of mobile base stations, efficiently processes computing tasks, optimizes resource utilization, and reduces the operating cost of edge computing.

Description

Technical Field

[0001] The present application relates to the field of software technology, and in particular to a mobile base station computing power management method and related devices. Background Art

[0002] With the widespread deployment of 4G and 5G radio access networks (RAN), the computing resources of mobile base stations are mainly used for baseband processing and communication protocol execution. However, mobile base stations are often in a low-load state during non-peak hours, which leads to idle resources.

[0003] Since mobile base station nodes are huge in number and have the advantage of being close to users and data sources, making full use of the computing resources of mobile base stations can not only improve resource utilization efficiency, but also effectively reduce network bandwidth consumption and reduce the transmission delay of edge computing tasks.

[0004] Therefore, how to efficiently utilize the idle computing resources of mobile base stations, especially in supporting Internet of Things (IoT) services and edge computing tasks, has become an urgent problem to be solved. Summary of the invention

[0005] In view of the above problems, the present application provides a mobile base station computing power management method and related devices to achieve the purpose of efficiently utilizing the idle computing power resources of the mobile base station. The specific scheme is as follows:

[0006] A first aspect of the present application provides a mobile base station computing power management method, the mobile base station computing power management method comprising:

[0007] Monitoring the system load of the mobile base station, and predicting the load change trend of the mobile base station based on the historical system load closest to the current moment;

[0008] If the load change trend is no load, determine a first computing task to be executed, allocate a first container resource matching its task type to the first computing task in a resource pool of the mobile base station, and execute a container instance of the first computing task based on the first container resource;

[0009] If the load change trend is overload, determine the second computing task being executed, release the second container resources occupied by the second computing task, and reclaim the computing power resources of the resource pool by adjusting the capacity size of the resource pool.

[0010] In a possible implementation, predicting the load change trend of the mobile base station according to the historical system load closest to the current moment includes:

[0011] The load forecasting model is obtained by training the basic model based on the time series analysis model;

[0012] Inputting a first time series corresponding to the historical system load into the load prediction model, and outputting a second time series through prediction by the load prediction model, wherein the second time series can represent the target system load of the mobile base station at a future moment;

[0013] The load variation trend is determined according to the target system load.

[0014] In a possible implementation, allocating a first container resource matching the task type to the first computing task in the resource pool of the mobile base station includes:

[0015] Determine whether there is a target resource pool in the resource pool whose level matches the task type of the first computing task;

[0016] If so, monitor the remaining resource status of the target resource pool; when the remaining resource status indicates that the target resource pool has idle resources, allocate the first container resource to the first computing task in the target resource pool; when the remaining resource status indicates that the target resource pool does not have idle resources, put the first computing task into a waiting queue corresponding to the target resource pool, schedule tasks for the waiting queue, and when the first computing task is scheduled, return to execute the step of monitoring the remaining resource status of the target resource pool;

[0017] If not, create the target resource pool according to the task type of the first computing task, and allocate the first container resource to the first computing task in the target resource pool.

[0018] In a possible implementation, the scheduling of tasks on the waiting queue includes:

[0019] Obtaining a weight factor, expected benefit, maximum resource requirement, maximum waiting time, and execution time of each third computing task in the waiting queue, wherein the weight factor represents the priority of the third computing task;

[0020] Taking maximizing total revenue as the goal and the capacity of the target resource pool as the constraint, a scheduling model including weight factors, expected revenue, maximum resource requirements, maximum waiting time and execution time of each third computing task is established;

[0021] The scheduling model is solved by a linear programming relaxation method, and the solution of the scheduling model is converted into an integer solution, wherein the integer solution represents the third type of computing tasks to be scheduled.

[0022] In a possible implementation, the releasing the second container resources occupied by the second computing task includes:

[0023] According to the task type of the second computing task, the container instance of the second computing task is migrated or closed to release the second container resources.

[0024] In a possible implementation, the mobile base station computing power management method further includes:

[0025] Determining a current load state of the mobile base station according to a current system load at a current moment;

[0026] If the current load state is overload, cell switching or call access control is performed on the communication services accessing the mobile base station.

[0027] A second aspect of the present application provides a mobile base station computing power management device, the mobile base station computing power management device comprising:

[0028] A load monitoring module, used to monitor the system load of the mobile base station and predict the load change trend of the mobile base station based on the historical system load closest to the current moment;

[0029] an idle management module, configured to, if the load change trend is idle, determine a first computing task to be executed, allocate a first container resource matching its task type to the first computing task in a resource pool of the mobile base station, and execute a container instance of the first computing task based on the first container resource;

[0030] An overload management module is used to determine the second computing task being executed if the load change trend is overload, release the second container resources occupied by the second computing task, and reclaim the computing power resources of the resource pool by adjusting the capacity size of the resource pool.

[0031] The third aspect of the present application provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements the mobile base station computing power management method of the above-mentioned first aspect or any implementation method of the first aspect.

[0032] A fourth aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0033] The memory is used to store computer programs;

[0034] The processor is used to execute the computer program so that the electronic device can implement the mobile base station computing power management method of the above-mentioned first aspect or any implementation method of the first aspect.

[0035] A fifth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the mobile base station computing power management method of the above-mentioned first aspect or any implementation method of the first aspect.

[0036] By means of the above technical scheme, the present application provides a mobile base station computing power management method and related devices, which monitor the system load of the mobile base station and predict the load change trend of the mobile base station based on the historical system load closest to the current moment; if the load change trend is no-load, determine the first computing task to be executed, allocate the first container resource matching its task type to the first computing task in the resource pool of the mobile base station, and execute the container instance of the first computing task based on the first container resource; if the load change trend is overload, determine the second computing task in execution, release the second container resource occupied by the second computing task, and recycle the computing power resources of the resource pool by adjusting the capacity size of the resource pool. The present application predicts its load change trend by monitoring the system load of the mobile base station. When the load change trend is no-load, container resources are allocated to computing tasks and their container instances are executed. When the load change trend is overload, the container resources of the computing tasks can be released and the capacity size of the resource pool can be adjusted to recycle its computing power resources. Under the premise of ensuring the normal operation of communication services, this application maximizes the use of idle computing resources of mobile base stations, efficiently processes computing tasks, optimizes resource utilization, and reduces the operating costs of edge computing. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0038] Figure 1 A flow chart of a mobile base station computing power management method provided in an embodiment of the present application;

[0039] Figure 2 A partial flow chart of a mobile base station computing power management method provided in an embodiment of the present application;

[0040] Figure 3 A schematic diagram of another part of the flow chart of a mobile base station computing power management method provided in an embodiment of the present application;

[0041] Figure 4 A schematic diagram of another part of the flow chart of a mobile base station computing power management method provided in an embodiment of the present application;

[0042] Figure 5A schematic diagram of the structure of a mobile base station computing power management device provided in an embodiment of the present application;

[0043] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0044] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0045] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0046] The terms "first", "second" etc. in the specification of the application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable in appropriate circumstances, and this is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0047] A large number of mobile base stations have been deployed in the Radio Access Network (RAN), which constitute the backbone of the mobile communication network. Currently, the computing resources of these base stations are mainly used for baseband processing and the execution of communication protocols. Since mobile communication services show a significant tidal effect, that is, the load fluctuates in different time periods, many mobile base stations are often in a low-load state during non-peak hours, resulting in a large amount of computing resources being idle.

[0048] The inventors have found through research that the existing solutions assume that mobile base stations are stable computing resources. The main research is how to schedule computing tasks in a multi-level computing system between cloud and edge terminals, and between mobile base stations, taking into account user geographic location, user service characteristics, node computing status, and computing task types. However, the primary task of a mobile base station is to ensure the normal operation of communication services, and computing resources are mainly used for baseband processing and the execution of communication protocols. Due to the dynamic changes in communication service traffic, mobile base stations are not stable computing resources. This dynamic characteristic makes it a major challenge to efficiently manage and utilize the computing resources of mobile base stations in practical applications.

[0049] In this regard, the present application provides a mobile base station computing power management method and related devices, which, under the premise of ensuring the normal operation of communication services, maximize the use of idle computing power resources of mobile base stations to run computing tasks (including IoT services and edge computing tasks), thereby optimizing the utilization efficiency of mobile base station computing power resources and reducing the operator's investment cost in edge computing computing power resources. Under the premise of ensuring the normal operation of communication tasks, the present application maximizes the use of idle computing power resources of mobile base stations, and realizes the optimal utilization of computing power resources and efficient scheduling of computing tasks through multi-level resource pool technology and dynamic load management. The mobile base station computing power management method of the embodiment of the present application is described in detail below in conjunction with the accompanying drawings.

[0050] See also Figure 1 , Figure 1 A flow chart of a mobile base station computing power management method provided in an embodiment of the present application. Figure 1 As shown, a mobile base station computing power management method provided in an embodiment of the present application may include steps S10 to S30, and these steps are described in detail below.

[0051] S10, monitoring the system load of the mobile base station, and predicting the load change trend of the mobile base station according to the historical system load closest to the current moment.

[0052] In the embodiment of the present application, when the communication services are running, they rely on the existing operating system and do not need to rely on containerization or resource pool technology. When the communication service load is low, it will accept user and external requests to intelligently schedule computing tasks. In this regard, the system load of the mobile base station can be monitored in real time to determine the occupancy of the computing power resources of the mobile base station, such as the occupancy of CPU, memory, storage, network bandwidth, etc. In this way, the system load of the mobile base station at different times can be obtained.

[0053] The historical system load at multiple historical moments closest to the current moment is retrieved, and the historical system load can be predicted based on the time series analysis tool to obtain the load change trend of the mobile base station at future moments.

[0054] In actual applications, the resource usage of mobile base stations can be analyzed in real time based on advanced monitoring models. Once abnormal conditions such as resource overload are found, the alarm mechanism is automatically triggered to notify the administrator to intervene in time to ensure system stability and prevent potential service interruptions.

[0055] In a possible implementation, a load forecasting model is trained based on a time series analysis tool to predict the load change trend. Figure 2 , Figure 2 A partial flow chart of a mobile base station computing power management method provided in an embodiment of the present application. Figure 2As shown, an embodiment of the present application provides a mobile base station computing power management method, wherein step S10 "predicts the load change trend of the mobile base station based on the historical system load closest to the current moment" may include steps S101 to S103, and these steps are described in detail below.

[0056] S101, training is performed using the time series analysis model as a basic model to obtain a load prediction model.

[0057] S102, inputting a first time series corresponding to the historical system load into a load prediction model, and outputting a second time series through prediction by the load prediction model, wherein the second time series can represent the target system load of the mobile base station at a future moment.

[0058] S103, determining a load change trend according to the target system load.

[0059] In the embodiment of the present application, by predicting the load change trend of the mobile base station in advance, resources are reasonably allocated and the resource pool capacity is dynamically adjusted to avoid affecting the operation of communication services, so that measures can be taken in advance to alleviate peak pressure and optimize resource utilization.

[0060] The inventors have found through research that the system load of mobile base stations has the characteristics of periodicity, trend, burst, randomness, autoregression, spatial correlation and user behavior dependence, which reflect the temporal and spatial dynamic changes of system load, such as daily and weekly peaks, long-term growth trends, short-term peaks caused by special events, random fluctuations, the impact of past load on current load, load transfer between adjacent base stations, and the direct impact of user behavior on load. These complex characteristics need to be considered in load prediction and resource management to ensure efficient and stable operation of mobile base stations.

[0061] Time series analysis tools can identify patterns, trends, seasonality, periodicity and other characteristics in time series and predict future data. In the embodiment of the present application, the NeuralProphet time series analysis model can be used as the basic model. Each time series consists of multiple components, and each component contributes an additive component to the prediction result. Most components can also be scaled according to the trend to produce a multiplicative effect. Time series analysis model It can be expressed by the following formula (1):

[0062] (1)

[0063] in, It's time The trend of (i.e., the trend component), It's time The periodic impact of (i.e., the periodic component), It's time The event and holiday effects (i.e., event components), is a known exogenous variable in the future at time The regression effect of is based on past observations The autoregressive effect (i.e., the autoregressive component) of It's in time The regression effects of lagged observations of exogenous variables.

[0064] In the embodiment of the present application, time series analysis technology is used to analyze and predict the dynamic changes of mobile base station loads, taking into account components such as trends, periodicity, events, and autoregression to capture the complex change pattern of mobile base station loads, thereby improving the accuracy and reliability of load prediction. Reflects the long-term change trend of mobile base station load; periodic component Capture the periodic fluctuations of mobile base station load, such as daily, weekly, and annual load changes; event component It is used to model the impact of special events on mobile base station load, such as holidays, special events, etc. Among them, event markers mark the time point when the event occurs in the data, and the event impact model defines the short-term and long-term impact of the event on the load; the autoregressive component Models the impact of past load values ​​on current load. It uses past load values ​​as model input, captures the autocorrelation of the load, and sets the autoregressive order to determine how many past load values ​​are used.

[0065] Assume that the error in time series analysis Uniform distribution ,in, Represents the mean, represents the standard deviation, and is a byproduct of time series analysis. For uniform distribution , the relationship between the mean and standard deviation is: , , through the above relationship, we can solve and : , .

[0066] If the prediction error range is Within , the prediction is defined as reliable. Uniform distribution , so the error is Evenly distributed between. Because the error exist Evenly distributed between exist Probability within is the ratio of the length of this interval to the length of the overall interval. The length is . Because the overall interval length is , and the length of the credible interval is , then each prediction error exist Probability .

[0067] The negative binomial distribution describes the probability that, in a series of independent Bernoulli trials, Number of trials required for success The cumulative distribution function formula can be expressed as follows:

[0068] (2)

[0069] In order to ensure independent Bernoulli trials to achieve Second error exist Probability , we can use the cumulative distribution function to find the appropriate Value. Need to find Make , this value can be obtained by combining historical forecast data.

[0070] Assume y Indicates The system load of the mobile base station at the moment is predicted using the prediction model to predict the system load of the mobile base station at the next k moments to obtain the predicted load sequence Based on historical data, calculate the predicted system load at the next k moments and historical system load The absolute difference between: . Define the threshold ,if , then count Increase once. Count target ,in, It is obtained through the above algorithm.

[0071] In practical applications, system load prediction at future times is performed k-step forecast, record the number of forecasts that exceed the threshold Count If here In the forecast, there are Second-rate , then the multi-level resource pool adjustment process is triggered to adjust the computing power resources in different types of resource pools according to a certain ratio.

[0072] Based on this, in the embodiment of the present application, historical data can be collated to generate a sample data set suitable for training the model, specifically including the system load of the executed tasks. Then, the sample data set is used to train the time series analysis model, and the trained model is verified and tested to ensure its accuracy and reliability, and the verified and tested model is saved as a load prediction model for use in actual load prediction.

[0073] In practical applications, the system load of the mobile base station can be monitored in real time, and the first time series corresponding to multiple historical system loads closest to the current moment can be input into the load prediction model. The load prediction model can be used to predict the second time series corresponding to the target system load at the future moment. Furthermore, the target system load can be statistically analyzed to determine the magnitude of its load change, so as to identify whether there is a trend of no load or overload.

[0074] In practical applications, the above prediction and analysis process can be repeated to determine the probability of the load change amplitude of the target system load exceeding the threshold in multiple predictions, so that the trend with the highest probability of occurrence can be used as the subsequent load change trend, so as to take different treatment measures. By training and using the load prediction model, combined with real-time monitoring and historical data analysis, accurate prediction of the system's future load is achieved. According to the prediction results, the execution status of the resource pool and computing tasks can be dynamically adjusted to ensure efficient use of resources under load changes and avoid overload or resource waste. Based on this, the intelligent management level of the system can be improved, ensuring efficient use of resources and stable operation of the system.

[0075] In addition, in the embodiments of the present application, historical data can also be collected, stored, cleaned, integrated and archived. Historical data can be stored in a local or remote file system, the data can be cleaned to ensure quality, historical data from different sources can be integrated to form a data set with a standardized and unified format, and old data can be archived regularly to optimize storage efficiency and access speed.

[0076] S20: If the load change trend is no-load, determine the first computing task to be executed, allocate a first container resource matching its task type to the first computing task in the resource pool of the mobile base station, and execute the container instance of the first computing task based on the first container resource.

[0077] In the embodiment of the present application, a user or system can submit a computing task. When the load change trend of the mobile base station is no-load, a first computing task to be executed can be determined therefrom, and a corresponding first container resource is allocated to it from the resource pool of the mobile base station according to the task type of the first computing task. Of course, the resource pool of the mobile base station is a multi-level resource pool, and the level of each resource pool in the multi-level resource pool matches the task type of the computing task, and the first container resource is allocated from the resource pool whose level matches the task type of the first computing task.

[0078] After allocating the first container resource to the first computing task, by deploying the container instance of the first computing task, the container instance of the first computing task can call the first container resource to execute the first computing task. During the task execution process, the system updates the resource pool status, manages the task status of the first computing task, monitors, records and adjusts it, and finally feeds back the execution result to the user or the system.

[0079] In actual applications, users can submit tasks, view task status, and obtain task execution results through the interactive interface and API interface. Users can configure task parameters and resource requirements through a direct and intuitive interface or API to ensure that task requirements are accurately communicated and processed.

[0080] In the embodiment of the present application, multi-level resource pool technology is used to schedule computing tasks of different task types in their respective resource pools, ensuring that each type of computing task has a fair chance to be executed. This strategy not only improves resource utilization efficiency, but also avoids performance degradation caused by resource competition. By allocating dedicated resource pools to computing tasks of different task types, it can be ensured that critical tasks obtain sufficient resource support when needed without being interfered with by other non-critical tasks.

[0081] Combined with container technology, the configuration management and migration convenience of tasks and their operating environments can be further improved. Containerization technology encapsulates each task and its dependent environment in an independent container, ensuring the consistency and portability of tasks in different environments. Whether migrating tasks from one resource pool to another or scheduling between different physical or virtual machines, container technology can simplify the process and improve the flexibility and adaptability of the system.

[0082] Setting resource quotas and resource limits is a key measure to ensure stable system operation. By setting specific resource quotas for each resource pool and each container, you can prevent certain tasks from occupying too many computing resources, such as CPU, memory, storage, and network bandwidth, thereby affecting the operation of other tasks. Fine-grained resource limits, such as limiting CPU usage and memory usage, can further fine-tune resource allocation and usage. This resource management strategy not only improves resource utilization, but also effectively prevents system overload problems caused by a single task.

[0083] In addition, in an embodiment of the present application, the process of the container instance based on the first container resource for the first computing task may include the entire process of the mobile base station from hardware detection and resource registration, to task analysis and program customization, and then to the deployment and operation of computing tasks. First, the mobile base station detects its hardware type and specifications when it starts, and registers the information to the central resource management system. Then, according to the needs of computing tasks, customize and optimize the task program, generate container image files suitable for different hardware types, then the mobile base station downloads the corresponding image file from the central image library, configures the operating environment and deploys the container instance, and finally starts the container instance to execute the computing task, and monitors the running status and resource usage of the computing task in real time to ensure the smooth completion of the computing task.

[0084] Specifically, when the mobile base station is started, the system detects its hardware type (such as CPU, GPU, NPU, etc.) and specifications (such as the number of cores, frequency, memory size, etc.), and registers this information in the central resource management system. According to the computing tasks to be executed, the demand for hardware resources is analyzed, and the computing task program is customized and optimized for different hardware types (such as CUDA programming for GPUs, AI frameworks for NPUs, etc.) to ensure that it can run efficiently on the target hardware, and the customized container image is stored in the central image library.

[0085] The task image files of various hardware types are stored in the central image library. The mobile base station downloads the corresponding image file from the central image library according to the registered hardware type to ensure that the downloaded task image is suitable for the current hardware type. Configure the operating environment in the mobile base station, install the necessary drivers and dependent libraries, ensure that the task program in the image file can run normally, use container technology (such as Docker) to deploy the downloaded image file to the container instance, and start the container instance to execute the computing task program. After starting the container instance, monitor the running status and resource usage of the computing task in real time to ensure the successful completion of the task.

[0086] It should be noted that in this application, the computing power resources of the mobile base station can come from the computing power resources inherent in the mobile base station, or from the computing power board integrated on the mobile base station. Various computing power chips must support container deployment, support multi-instance operation, and have resource quota and restriction functions. The container image is provided by the customer, and the customer needs to understand the hardware type and develop the computing task program accordingly.

[0087] Computing power virtualization and containerization combined with resource pools can achieve efficient task scheduling and management. Tasks are scheduled in independent resource pools, and different types of tasks can get fair execution opportunities. The application of computing power virtualization and containerization makes task configuration management and migration more convenient. At the same time, through resource quotas and resource restrictions, it can prevent excessive resource occupation and ensure system stability and efficiency.

[0088] Currently, mainstream chip manufacturers have realized virtualization and containerization support for computing chips through a variety of advanced technical means, which facilitates the effective management and utilization of computing resources. The present invention is based on but not limited to the following technologies to realize the virtualization and containerization capabilities of various computing chips. The following describes the container instance deployment of CPU, GPU, NPU, FPGA and DSP respectively:

[0089] The main methods for CPU to support multiple virtual instances include hardware-assisted virtualization technology, Hypervisor, and container technology. Hardware-assisted virtualization technology provides basic virtualization instruction support and improves virtualization performance. Hypervisor uses these hardware extensions to provide efficient resource isolation and management. Container technology implements lightweight virtualization through isolation at the operating system layer. Combined with resource management and isolation technologies such as cgroups and namespaces, these methods jointly achieve efficient management and resource utilization of multiple virtual instances.

[0090] GPU virtualizes a physical GPU into multiple virtual GPU instances. Each virtual GPU instance can be assigned to an independent virtual machine or container to provide independent computing and graphics processing capabilities. It enables multi-user sharing and dynamic management of GPU resources and supports a wide range of graphics and computing applications.

[0091] NPU combines hardware and software virtualization technologies to achieve efficient resource utilization and task isolation. Hardware partitioning technology allows a physical NPU to be divided into multiple independent virtual instances, each with its own computing resources, memory, and bandwidth. Specific methods include logical partitioning and time partitioning. Logical partitioning is to divide the computing units, memory, and other resources inside the NPU into several logical partitions during hardware design. Each logical partition can be used as an independent virtual instance. Time partitioning is a method of allocating time slices to allow multiple virtual instances to share the same physical NPU in time. Within a time slice, only one virtual instance is allowed to access NPU resources.

[0092] FPGAs support virtualization by partitioning resources into multiple independent logic blocks, each of which can be programmed and used independently.

[0093] DSP virtualizes resources into multiple virtual DSP instances and schedules resources through the virtualization management layer.

[0094] In a possible implementation, the first container resource may be allocated to the first computing task through task scheduling. Figure 3 , Figure 3 This is another partial flow chart of a mobile base station computing power management method provided in an embodiment of the present application. Figure 3 As shown, an embodiment of the present application provides a mobile base station computing power management method, wherein step S20 "allocating a first container resource matching the task type to a first computing class task in the resource pool of the mobile base station" may include steps S201 to S203, and these steps are described in detail below.

[0095] S201, determine whether there is a target resource pool in the resource pool whose level matches the task type of the first computing task. If yes, execute step S202; if not, execute step S203.

[0096] S202, monitor the remaining resource status of the target resource pool; when the remaining resource status indicates that the target resource pool has idle resources, allocate a first container resource for the first computing class task in the target resource pool; when the remaining resource status indicates that the target resource pool does not have idle resources, put the first computing class task into a waiting queue corresponding to the target resource pool, schedule tasks for the waiting queue, and when the first computing class task is scheduled, return to execute the step of monitoring the remaining resource status of the target resource pool.

[0097] S203: Create a target resource pool according to the task type of the first computing task, and allocate a first container resource to the first computing task in the target resource pool.

[0098] In the embodiment of the present application, the level of the resource pool can be matched according to the task type of the first computing task to determine the source of resources to be allocated. Specifically, it is determined whether there is a target resource pool in the resource pool whose level matches the first computing task.

[0099] If the target resource pool exists, the remaining resource status in the target resource pool is monitored to determine whether there are free resources in the target resource pool; if the target resource pool has free resources, the first container resource is allocated to the first computing class task in the target resource pool; if the target resource pool does not have free resources, it is detected whether the first computing class task can wait. If the first computing class task can wait, the first computing class task is placed in the waiting queue corresponding to the target resource pool. After task scheduling, the first container resource is reallocated to the task when the target resource pool has free resources. Otherwise, if the first computing class task cannot wait, a message of task failure is returned.

[0100] If the target resource pool does not exist, it can be determined whether the current system load at the current moment is less than the corresponding load threshold; if it is less, a new resource pool is created as the target resource pool, and the first container resource is allocated to the first computing class task in the target resource pool; if it is not less, a task failure message is returned.

[0101] In a possible implementation, tasks are scheduled effectively in the waiting queue to ensure maximum benefits. Figure 4 , Figure 4 This is another partial flow chart of a mobile base station computing power management method provided in an embodiment of the present application. Figure 4 As shown, a mobile base station computing power management method provided by an embodiment of the present application, wherein "scheduling tasks for the waiting queue" in step S202 may include steps S2021 to S2023, and these steps are described in detail below.

[0102] S2021, obtaining the weight factor, expected benefit, maximum resource requirement, maximum waiting time and execution time of each third computing task in the waiting queue, where the weight factor represents the priority of the third computing task.

[0103] S2022, with the goal of maximizing total revenue and the capacity of the target resource pool as a constraint, establish a scheduling model that includes the weight factors, expected revenue, maximum resource requirements, maximum waiting time and execution time of each third computing task.

[0104] S2023, using a linear programming relaxation method to solve the scheduling model, and converting the solution of the scheduling model into an integer solution, where the integer solution represents the third type of computing task to be scheduled.

[0105] In the embodiment of the present application, the waiting queue corresponding to each resource pool in the multi-level resource pool usually has Computing tasks (i.e., third-class computing tasks) are waiting to be scheduled and run. Each third-class computing task has a different priority, expected benefit, maximum resource requirement, maximum waiting time, and execution time. The priority can be converted into a weight factor to calculate the weighted benefit. If the third-class computing task is not executed within the maximum waiting time, the task fails and there is no benefit. Multiple tasks can run in parallel, but the total amount of the resource pool is limited.

[0106] The goal of task scheduling in the target resource pool is to maximize the total benefit under these constraints. This is a problem of finding the extreme value of the linear objective function under the constraints. The mathematical model shown in the following formula (3) can be established as the scheduling model:

[0107] (3)

[0108] The constraints shown in the following formulas (4), (5), (6), and (7) are satisfied:

[0109] (4)

[0110] (5)

[0111] (6)

[0112] (7)

[0113] in, Indicates the total amount of tasks of the third computing category; Indicates The priority weight factor of the third computing task ( ); Indicates The priority weight factor of the third computing task ( ); Indicates The expected return of the third computing task ( ); Indicates The maximum resource requirement of the third computing task ( ); Indicates The maximum waiting time of the third computing task ( ); Indicates The execution time of the third computing task ( ); Indicates the capacity of the target resource pool, that is, the total amount limit; is a binary variable, indicating Whether a third computing task is selected for execution ( When it is selected for execution, is not selected for execution); Indicates The start time of the third computing task; Represents the total scheduling period; represents an indicator function, when In the third computing task It is 1 if the execution time is within the range, otherwise it is 0.

[0114] Since this is a 0-1 integer programming problem, there is usually no analytical solution. Therefore, the embodiment of the present application adopts a linear programming relaxation method to solve it. Relax to continuous variables, solve the relaxed solution of the linear programming, and then convert the relaxed solution into an integer solution through heuristic methods or branch and bound method. The solution steps are as follows:

[0115] 1) Define the total number of tasks and the weight factor, expected benefit, maximum resource requirement, maximum waiting time, execution time, target resource pool capacity and scheduling period of each third computing task;

[0116] 2) Construct a mathematical model of the objective function and constraints;

[0117] 3) Choose a suitable linear programming solver, commonly used ones include PuLP, GLPK, Gurobi, CPLEX, etc.;

[0118] 4) Input the task parameters and run the solver to obtain the optimal solution;

[0119] 5) Post-process the results of the LP solver: round the continuous solutions to ensure that the results are integer solutions;

[0120] 6) Schedule tasks and update the status of the target resource pool based on the solution results.

[0121] In actual applications, you can also manage the entire life cycle of computing tasks, including creation, scheduling, execution, monitoring, and completion. Track the execution status of computing tasks, and collect and store task results after the task is completed. Work with task scheduling to ensure that high-priority computing tasks get resources first and optimize the efficiency of system resource usage.

[0122] S30: If the load change trend is overload, determine the second computing task being executed, release the second container resources occupied by the second computing task, and reclaim the computing power resources of the resource pool by adjusting the capacity of the resource pool.

[0123] In an embodiment of the present application, when the load change trend of the mobile base station is overloaded, the second computing type task being executed in the current multi-level resource pool of the mobile base station can be checked, the second container resources occupied by the second computing type task can be released, and the capacity of the resource pool where the second computing type task is located can be adjusted to recover the computing power resources allocated by the mobile resource to the resource pool, so as to cope with the future load requirements of the mobile base station.

[0124] In a possible implementation, the container instance of the second computing task can be migrated or closed according to the task type of the second computing task to release the second container resources. Specifically, in the embodiment of the present application, the task type of the second computing task can be divided into Internet of Things tasks, real-time computing tasks and non-real-time computing tasks. Correspondingly, the multi-level resources of the mobile base station are also mainly divided into the following three categories:

[0125] Resource pool for IoT tasks. The resource pool for IoT tasks accounts for a relatively small proportion of the resources in the entire system. This is because IoT tasks usually have low demand for computing resources and rely more on their periodic and continuous operation characteristics. In order to ensure the continuity and stability of IoT tasks, such tasks should minimize migration. Frequent migration may cause data loss or task interruption, affecting the overall business stability. Therefore, IoT tasks are designed to run relatively fixedly in the resource pool, and try not to migrate even when the system load is high.

[0126] Real-time computing task resource pool: Real-time computing tasks require a large amount of computing resources, such as video surveillance, VR rendering, etc. These tasks usually have extremely high requirements for computing power and response speed. Therefore, the real-time computing task resource pool occupies a large proportion of resources to meet its high-intensity computing needs. When the system load increases, especially when the load of communication tasks increases, real-time computing tasks may be migrated to other resource pools or other mobile base stations to release key resources and ensure priority processing of communication tasks. If there is no suitable migration target, you can choose to shut down some real-time computing tasks to ensure overall stability and priority needs of communication services.

[0127] Resource pool for non-real-time computing tasks: Non-real-time computing tasks, such as AI model training and data backup, are less sensitive to time, so they are usually scheduled when the load is relatively idle. In the embodiment of the present application, resources are allocated to real-time computing tasks and communication tasks first, and only when the resource requirements of these tasks are met will non-real-time computing tasks be allocated to the resource pool for operation. When the system load increases, especially during peak communication periods, non-real-time computing tasks will be shut down or migrated first to free up more resources for real-time computing tasks and communication services.

[0128] In addition, in an embodiment of the present application, computing tasks are deployed and managed through containerization technology, thereby achieving flexible allocation of resources and efficient scheduling of tasks. When migrating the second computing type task, the relevant container instance can be migrated from one mobile base station to a nearby mobile base station to ensure that the container can quickly resume operation on the new node and maintain service continuity and data consistency. The migration process includes the creation of container snapshots, synchronization of network configurations, migration of data volumes, seamless switching of services, and resource reallocation and status synchronization after the migration is completed. Closely cooperate with resource pool management to ensure resource allocation and task status consistency during task migration. In addition, of course, if there is no mobile base station suitable for migration, some container instances of the second computing type tasks can be shut down to release related resources.

[0129] This application uses multi-level resource pools and container technology to achieve efficient task scheduling and management, ensuring the priority of communication services while optimizing resource utilization and system stability for computing tasks.

[0130] In one possible implementation, considering that container migration takes a certain amount of time, and in order to protect certain important computing tasks that are already running from being interrupted, the embodiment of the present application can also temporarily not migrate these key container instances while ensuring that no new container instances are started. In this process, in order not to affect the normal operation of existing communication services, the system can select cell switching or call access control. In this regard, a mobile base station computing power management method provided in an embodiment of the present application also includes the following steps:

[0131] The current load state of the mobile base station is determined according to the current system load at the current moment; if the current load state is overloaded, cell switching or call access control is performed on the communication services accessing the mobile base station.

[0132] In an embodiment of the present application, the current system load is compared with the corresponding load threshold to determine the current load state of the mobile base station. If the current load state is overloaded, it can be determined whether there are any nearby mobile base stations that are idle. If so, a neighboring cell with a lower load is selected, and some user equipment (UE) is selected to switch to the cell to reduce the load of the current baseband processing unit (BBU). This method effectively balances the network load and improves the overall service quality. On the contrary, if there is no such cell, a threshold can be set to limit new call access requests, start call access control, limit new call access, and give priority to ensuring the quality of existing connections. This measure ensures that the communication experience of existing users is not affected.

[0133] Through the above description, the embodiment of the present application provides a method for managing computing power of a mobile base station, which divides the resource pool into multiple levels, and dynamically adjusts the size of the resource pool based on efficient dynamic prediction of base station load and scheduling of computing tasks, and optimizes the operation of scheduling computing tasks on the multi-level resource pool. The present application can maximize the use of idle resources to perform computing tasks while ensuring the stable operation of communication tasks, thereby achieving the optimization of resource utilization and the dynamic balance of business needs. The present application has the following advantages:

[0134] 1) Efficient resource utilization: Through multi-level resource pools and container technology, fine scheduling of different types of tasks can be achieved to improve resource utilization. In addition, container technology ensures the consistency and portability of tasks in different environments, simplifying task migration and scheduling.

[0135] 2) Prioritize communication tasks: Dynamic load management strategies ensure smooth operation of communication services during peak periods. Measures such as call access control and inter-cell switching are taken to prioritize communication service needs and improve user experience.

[0136] 3) Flexible scheduling of computing tasks: Intelligently schedule computing tasks according to task type and system load to achieve flexible resource allocation. It also supports efficient operation of IoT tasks, real-time computing tasks, and non-real-time computing tasks to optimize system resource configuration.

[0137] 4) Fine-grained resource management: By setting specific resource quotas and restrictions, resource allocation can be finely controlled to prevent excessive resource occupation, ensure stable system operation, improve overall resource utilization efficiency, and prevent system overload problems.

[0138] The above introduces a mobile base station computing power management method provided by an embodiment of the present application. The following will introduce a device for executing the above mobile base station computing power management method.

[0139] See also Figure 5 , Figure 5 This is a schematic diagram of the structure of a mobile base station computing power management device provided in an embodiment of the present application. Figure 5 As shown, a mobile base station computing power management device provided in an embodiment of the present application includes:

[0140] The load monitoring module 10 is used to monitor the system load of the mobile base station and predict the load change trend of the mobile base station based on the historical system load closest to the current moment;

[0141] The idle management module 20 is used to determine a first computing task to be executed if the load change trend is idle, allocate a first container resource matching the task type to the first computing task in the resource pool of the mobile base station, and execute the container instance of the first computing task based on the first container resource;

[0142] The overload management module 30 is used to determine the second computing task being executed if the load change trend is overload, release the second container resources occupied by the second computing task, and recycle the computing power resources of the resource pool by adjusting the capacity size of the resource pool.

[0143] In a possible implementation, the load monitoring module 10 for predicting the load change trend of the mobile base station according to the historical system load closest to the current moment is specifically used for:

[0144] The time series analysis model is used as the basic model for training to obtain a load prediction model; the first time series corresponding to the historical system load is input into the load prediction model, and the second time series is output through the load prediction model. The second time series can characterize the target system load of the mobile base station at a future moment; the load change trend is determined according to the target system load.

[0145] In a possible implementation, the idle management module 20 for allocating a first container resource matching the task type to the first computing task in the resource pool of the mobile base station is specifically configured to:

[0146] Determine whether there is a target resource pool in the resource pool whose level matches the task type of the first computing class task; if so, monitor the resource remaining status of the target resource pool; when the resource remaining status indicates that the target resource pool has idle resources, allocate the first container resource to the first computing class task in the target resource pool; when the resource remaining status indicates that the target resource pool does not have idle resources, put the first computing class task into the waiting queue corresponding to the target resource pool, and schedule the task in the waiting queue, and when the first computing class task is scheduled, return to execute the step of monitoring the resource remaining status of the target resource pool; if not, create a target resource pool according to the task type of the first computing class task, and allocate the first container resource to the first computing class task in the target resource pool.

[0147] In a possible implementation, the idle management module 20 for scheduling tasks in the waiting queue is specifically used to:

[0148] The weight factor, expected benefit, maximum resource requirement, maximum waiting time and execution time of each third computing task in the waiting queue are obtained, wherein the weight factor represents the priority of the third computing task. With the goal of maximizing the total benefit and the capacity of the target resource pool as a constraint, a scheduling model is established that includes the weight factor, expected benefit, maximum resource requirement, maximum waiting time and execution time of each third computing task. The scheduling model is solved by the linear programming relaxation method, and the solution of the scheduling model is converted into an integer solution, wherein the integer solution represents the third computing task to be scheduled.

[0149] In a possible implementation, the overload management module 30 for releasing the second container resources occupied by the second computing task is specifically configured to:

[0150] According to the task type of the second computing task, the container instance of the second computing task is migrated or closed to release the second container resources.

[0151] In a possible implementation, the overload management module 30 is further configured to:

[0152] The current load state of the mobile base station is determined according to the current system load at the current moment; if the current load state is overloaded, cell switching or call access control is performed on the communication services accessing the mobile base station.

[0153] It should be noted that the detailed functions of each module in the embodiment of the present application can be found in the corresponding public part of the above-mentioned mobile base station computing power management method embodiment, and will not be repeated here.

[0154] The present application also provides an electronic device in an embodiment. Figure 6 , Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device in the embodiment of the present application may be a device capable of establishing a communication link with a mobile base station, such as a fixed terminal such as a laptop, PAD (tablet computer), desktop computer, etc. Figure 6 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0155] like Figure 6 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 to a random access memory (RAM) 603. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0156] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a memory card, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0157] An embodiment of the present application also provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any mobile base station computing power management method provided in the embodiment of the present application.

[0158] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any mobile base station computing power management method provided in the embodiment of the present application.

[0159] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0160] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0161] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0162] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A mobile base station computing power management method, characterized in that: The mobile base station computing power management method comprises: Monitoring the system load of the mobile base station, and predicting the load change trend of the mobile base station based on the historical system load closest to the current moment; If the load change trend is no-load, determine a first computing task to be executed, allocate a first container resource matching its task type to the first computing task in the resource pool of the mobile base station, and execute the container instance of the first computing task based on the first container resource; wherein a resource quota is set for each resource pool and each container; wherein the resource pool of the mobile base station is a multi-level resource pool, and the level of each resource pool in the multi-level resource pool matches the task type of the computing task; If the load change trend is overload, determine the second computing task being executed, release the second container resources occupied by the second computing task, and reclaim the computing power resources of the resource pool by adjusting the capacity size of the resource pool; The predicting of the load change trend of the mobile base station according to the historical system load closest to the current moment includes: The time series analysis model is used as the basic model for training to obtain a load forecasting model; the time series analysis model It is expressed by the following formula: in, It's time The trend It's time The cyclical impact of It's time Event and holiday effects, is a known exogenous variable in the future at time The regression effect of is based on past observations The autoregressive effect of It's in time The regression effect of lagged observations of exogenous variables; Inputting a first time series corresponding to the historical system load into the load prediction model, and outputting a second time series through prediction by the load prediction model, wherein the second time series can represent the target system load of the mobile base station at a future moment; The load variation trend is determined according to the target system load.

2. The mobile base station computing power management method according to claim 1, characterized in that: The allocating a first container resource matching the task type to the first computing task in the resource pool of the mobile base station includes: Determine whether there is a target resource pool in the resource pool whose level matches the task type of the first computing task; If so, monitor the remaining resource status of the target resource pool; when the remaining resource status indicates that the target resource pool has free resources, allocate the first container resource to the first computing task in the target resource pool; when the remaining resource status indicates that the target resource pool does not have free resources, put the first computing task into a waiting queue corresponding to the target resource pool, and schedule tasks for the waiting queue, and when the first computing task is scheduled, return to execute the step of monitoring the remaining resource status of the target resource pool; If not, create the target resource pool according to the task type of the first computing task, and allocate the first container resource to the first computing task in the target resource pool.

3. The mobile base station computing power management method according to claim 2, characterized in that: The step of scheduling tasks on the waiting queue includes: Obtaining a weight factor, expected benefit, maximum resource requirement, maximum waiting time, and execution time of each third computing task in the waiting queue, wherein the weight factor represents the priority of the third computing task; Taking maximizing total revenue as the goal and the capacity of the target resource pool as the constraint, a scheduling model including weight factors, expected revenue, maximum resource requirements, maximum waiting time and execution time of each third computing task is established; The scheduling model is solved by a linear programming relaxation method, and the solution of the scheduling model is converted into an integer solution, wherein the integer solution represents the third type of computing tasks to be scheduled.

4. The mobile base station computing power management method according to claim 1, characterized in that: The releasing the second container resources occupied by the second computing task includes: According to the task type of the second computing task, the container instance of the second computing task is migrated or closed to release the second container resources.

5. The mobile base station computing power management method according to claim 1, characterized in that: The mobile base station computing power management method further includes: Determining a current load state of the mobile base station according to a current system load at a current moment; If the current load state is overload, cell switching or call access control is performed on the communication services accessing the mobile base station.

6. A mobile base station computing power management device, characterized in that: The mobile base station computing power management device comprises: A load monitoring module, used to monitor the system load of the mobile base station and predict the load change trend of the mobile base station based on the historical system load closest to the current moment; An idle management module, configured to determine a first computing task to be executed if the load change trend is idle, allocate a first container resource matching its task type to the first computing task in the resource pool of the mobile base station, and execute a container instance of the first computing task based on the first container resource; wherein a resource quota is set for each resource pool and each container; wherein the resource pool of the mobile base station is a multi-level resource pool, and the level of each resource pool in the multi-level resource pool matches the task type of the computing task; an overload management module, configured to, if the load change trend is overload, determine the second computing task being executed, release the second container resources occupied by the second computing task, and reclaim the computing power resources of the resource pool by adjusting the capacity of the resource pool; The load monitoring module predicts the load change trend of the mobile base station according to the historical system load closest to the current moment, including: The time series analysis model is used as the basic model for training to obtain a load forecasting model; the time series analysis model It is expressed by the following formula: in, It's time The trend It's time The cyclical impact of It's time Event and holiday effects, is a known exogenous variable in the future at time The regression effect of is based on past observations The autoregressive effect of It's in time The regression effect of lagged observations of exogenous variables; Inputting a first time series corresponding to the historical system load into the load prediction model, and outputting a second time series through prediction by the load prediction model, wherein the second time series can represent the target system load of the mobile base station at a future moment; The load variation trend is determined according to the target system load.

7. A computer program product, characterized in that It includes computer-readable instructions, and when the computer-readable instructions are executed on an electronic device, the electronic device implements the mobile base station computing power management method as described in any one of claims 1 to 5.

8. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the mobile base station computing power management method as described in any one of claims 1 to 5.

9. A computer storage medium, characterized in that The storage medium carries one or more computer programs, and when the one or more computer programs are executed by an electronic device, the electronic device can implement the mobile base station computing power management method as described in any one of claims 1 to 5.

Images