Management platform, management method, program, and system

Abstract

Provided is a management platform including: a management unit configured to manage a plurality of distributed platforms that can execute a plurality of MEC applications; a metric acquisition unit configured to acquire, for each combination of the plurality of distributed platforms and the plurality of MEC applications, a metric including the number of connected terminals indicating the number of user terminals connected to the MEC applications; a region association information acquisition unit configured to acquire, for each of the plurality of distributed platforms, region association information associated with a region in which the distributed platform is disposed; and a demand prediction unit configured to predict, for each combination of the plurality of distributed platforms and the plurality of MEC applications, future demand based on the metric and the region association information. The distributed platform can also have a function of RAN control and a function of AI processing (RIC (RAN Intelligent Controller)).

Description

Technical Field

[0001] This invention relates to management platforms, management methods, programs, and systems. Background Technology

[0002] Patent Document 1 describes a configuration in which a management node manages distributed file and object storage for files used in an application in an accessible manner. The distributed file and object storage can access files managed in storage at other locations. The management node has a processor configured to specify the application's access status to files and, based on the access status, control a cache of distributed file and object storage based on its own location for files managed in storage at other locations used in the application before the application is executed.

[0003] Prior art literature

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Application Publication No. 2023-069701 Summary of the Invention

[0006] The technical problem that the invention aims to solve

[0007] Previously, in MEC (Multi-access Edge Computing) applications, the execution region was typically static. MEC application operators selected the region for deployment based on latency and fault tolerance considerations. However, in the case of distributed platforms such as data centers (DCs) with limited regional scale, or individual MEC AI applications requiring large-scale computing resources, statically allocating resources across all distributed platforms for all MEC AI applications cannot effectively and flexibly utilize computing resources. Therefore, predicting in advance the amount of MEC application demand in each region is crucial.

[0008] Solutions for solving technical problems

[0009] In the system described in this embodiment, for example, future demand forecasts for each region of each MEC application are made using past RAN (Radio Access Network) metrics and region association information associated with regions provided with past RAN. The system can use machine learning models for demand forecasting. RAN metrics may include the number of connected UEs (User Equipment), throughput, latency, and handover history for each distributed platform / each MEC application. Region association information may include event information and weather. Items of the predicted demand may include the future number of connected UEs and the future required throughput. For example, a time-series machine learning model can be used to learn and predict demand items for each distributed platform / each MEC application based on various RAN metrics / region association information for each distributed platform / each MEC application, and this service can be provided on a management platform that manages multiple distributed platforms.

[0010] As a specific example, the system 10 involved in this embodiment includes: a distributed platform having RAN control functions for controlling the RAN and AI processing functions for performing AI processing; and a management platform for managing multiple distributed platforms, wherein the management platform performs demand forecasting for each combination of multiple distributed platforms and multiple MEC applications. The management platform can adjust the deployment status of MEC applications or perform communication control based on the forecast results.

[0011] As types of AI processing, we can list AI processing associated with RAN control (sometimes referred to as RAN control AI processing) and AI processing not associated with RAN control (sometimes referred to as non-RAN control AI processing).

[0012] As an example of AI processing in RAN control, the RIC (RAN Intelligent Controller) can be cited. The RIC is a technology that uses AI to optimize RAN radio resources and automate RAN operations. RICs include Non-RT RICs (Non-Real Time RICs) and Near-RT RICs (Near-Real Time RICs). Non-RT RICs are sometimes also called Centralized RICs. Non-RT RICs reside within the SMO (Service Management and Orchestration) which manages and orchestrates the RAN. Non-RT RICs generate and notify RAN control-related policies and send information to Near-RT RICs. For example, a Non-RT RIC generates a learned model of RAN control by performing machine learning using data collected from the RAN and sends it to a Near-RT RIC. Near-RT RICs are sometimes also called Distributed RICs. Compared to Non-RT RICs, Near-RT RICs are located near RAN nodes (RUs, DUs, and CUs) and perform RAN node control and resource control. Near-RT RICs also offer higher real-time processing compared to Non-RT RICs. For example, Near-RT RICs perform inference processing associated with RAN control using a learned model obtained from a Non-RT RIC. RAN control AI processing is not limited to RICs.

[0013] Non-RAN control AI processing can be what is known as MEC AI applications. Non-RAN control AI processing includes arbitrary AI learning and inference processing that is not associated with RAN control.

[0014] According to one embodiment of the present invention, a management platform is provided. The management platform may include a management unit that manages multiple distributed platforms located in various locations, each capable of executing multiple MEC applications. The management platform may include a measurement acquisition unit that acquires a metric, including a number of connected terminals representing the number of user terminals connected to the MEC applications, for each combination of the multiple distributed platforms and the multiple MEC applications. The management platform may include a regional association information acquisition unit that acquires regional association information associated with the region where each of the multiple distributed platforms is configured. The management platform may include a demand forecasting unit that forecasts future demand based on the metric and the regional association information, for each combination of the multiple distributed platforms and the multiple MEC applications.

[0015] In the management platform, the measurement acquisition unit can acquire, for each combination of the plurality of distributed platforms and the plurality of MEC applications, the measurement also includes the communication status between the MEC application and the user terminal. The measurement acquisition unit can also acquire, for each combination of the plurality of distributed platforms and the plurality of MEC applications, the measurement also includes at least one of the throughput and communication latency between the MEC application and the user terminal.

[0016] In any of the management platforms, the metric acquisition unit can acquire the metric, which also includes a history of HO (Householding Organization) representing the history of user terminal switching during connection with the MEC application, according to each combination of the plurality of distributed platforms and the plurality of MEC applications.

[0017] In any of the management platforms, the regional association information acquisition unit can acquire the regional association information, including weather information and event information of the region where the distributed platform is configured, for each of the plurality of distributed platforms.

[0018] The management platform of any one of the claims may include: a storage unit that stores the metrics and regional association information of each combination of the plurality of distributed platforms and the plurality of MEC applications, acquired by the metric acquisition unit and the regional association information acquisition unit, as learning data; and a learning execution unit that generates a machine learning model that predicts the demand of each combination of the plurality of distributed platforms and the plurality of MEC applications based on the metrics and regional association information of each combination of the plurality of distributed platforms and the plurality of MEC applications by executing machine learning using the learning data stored in the storage unit, wherein the demand prediction unit may input the metrics and regional association information of each combination of the plurality of distributed platforms and the plurality of MEC applications for each of the past multiple periods into the machine learning model to predict the demand of each combination of the plurality of distributed platforms and the plurality of MEC applications for future periods. The machine learning model can take the metrics and regional association information of each combination of the multiple distributed platforms and the multiple MEC applications as input, and take the number of connected terminals of each combination of the multiple distributed platforms and the multiple MEC applications as output. The demand forecasting unit can take the metrics and regional association information of each combination of the multiple distributed platforms and the multiple MEC applications for each past period as input to the machine learning model, and obtain the number of connected terminals output from the machine learning model as the demand for each combination of the multiple distributed platforms and the multiple MEC applications for future periods.

[0019] The management platform of any one of the above may further include: an adjustment unit, which adjusts the deployment status of the respective MEC applications of the multiple distributed platforms based on the forecast results of the demand forecasting unit.

[0020] According to one embodiment of the present invention, a management method is provided, the management method being executed by a management platform, the management platform managing multiple distributed platforms configured in various locations, each capable of executing multiple MEC applications. The management method may include: a metric acquisition phase, wherein for each combination of the multiple distributed platforms and the multiple MEC applications, the metric acquisition phase acquires a metric including the number of connected terminals representing the number of user terminals connected to the MEC applications. The management method may include: a region association information acquisition phase, wherein for each of the multiple distributed platforms, the region association information associated with the region where the distributed platform is configured is acquired. The management method may include a demand forecasting phase, wherein for each combination of the multiple distributed platforms and the multiple MEC applications, the demand forecasting phase forecasts future demand based on the metric and the region association information.

[0021] According to one embodiment of the present invention, a program is provided for causing a computer to execute the management method.

[0022] According to one embodiment of the present invention, a system is provided that includes the management platform and the plurality of distributed platforms.

[0023] It should be noted that the above summary of the invention does not list all the essential features of the invention. Furthermore, sub-combinations of these feature groups can also constitute an invention. Attached Figure Description

[0024] Figure 1 An example of System 10 is shown in general.

[0025] Figure 2 This is an explanatory diagram used to illustrate the processing content based on the management platform 100.

[0026] Figure 3 This is an explanatory diagram used to illustrate the processing content based on the management platform 100.

[0027] Figure 4 This is an illustrative diagram used to explain learning based on the management platform 100.

[0028] Figure 5 This is a simplified example of the functional configuration of the management platform 100.

[0029] Figure 6 An example of the hardware configuration of a computer 1200 that functions as a management platform 100 or a distributed platform 200 is shown in a simplified manner. Detailed Implementation

[0030] The present invention will now be described through embodiments thereof; however, these embodiments are not intended to limit the invention as defined in the claims. Furthermore, the combinations of features described in the embodiments are not all necessary for the solution of the invention.

[0031] Figure 1 An example of system 10 is schematically illustrated. System 10 includes a management platform 100 and multiple distributed platforms 200. In system 10, the management platform 100 and the multiple distributed platforms 200 can coordinate to perform control and AI processing of RAN 310. Mobile communication services are provided to UE 30 through RAN 310. UE 30 may be an example of a user terminal.

[0032] RAN310 can be a virtualized vRAN, and system 10 can perform vRAN control. RAN310 can also be a physical RAN, and system 10 can perform physical RAN control. In this embodiment, the case where RAN310 is a vRAN will be used as an example for explanation.

[0033] The AI ​​processing performed by System 10 may include RAN control AI processing (sometimes referred to as RAN_AI). The AI ​​processing performed by System 10 may also include non-RAN control AI processing (sometimes referred to as non-RAN_AI).

[0034] The distributed platform 200 can be a data center deployed in various locations. The distributed platform 200 can be comprised of multiple devices. The distributed platform 200 can be implemented on a virtualization platform based on multiple devices. The distributed platform 200 can also be implemented using a single device. That is, the distributed platform 200 can also be a distributed device. The distributed platform 200 can function as a BBU (Baseband Unit), and the wireless base station 300 can function as an RRU (Remote Radio Unit). The distributed platform 200 can install a CU (User Unit). The distributed platform 200 can install a DU (User Utility Unit). The distributed platform 200 can install a UPF (User Plane Function).

[0035] The management platform 100 can be a data center that manages multiple distributed platforms 200. The management platform 100 can be configured across multiple devices. The management platform 100 can be implemented on a virtualization platform based on multiple devices. The management platform 100 can also be implemented through a single device. That is, the management platform 100 can also be a management device.

[0036] Sometimes the management platform 100 is referred to as the Core Brain, and the distributed platform 200 is referred to as the Regional Brain. It should be noted that... Figure 1 The example shown illustrates a scenario where one layer of management platform 100 is configured below management platform 100, but it is not limited to this. Distributed platform 200 can also be multi-layered. For example, in the case where two layers of distributed platforms 200 are configured below management platform 100, sometimes management platform 100 is referred to as the Core Brain, the lower-level distributed platform 200 is referred to as the Regional Brain, and the even lower-level distributed platform 200 is referred to as the Sub-Regional Brain.

[0037] In the presence of a Core Brain, a Regional Brain, and a Sub Regional Brain, the demand forecasting process involved in this implementation can be performed by either the Sub Regional Brain or the Regional Brain.

[0038] In the distributed platform 200, one or more CPUs (Central Processing Units) can be configured. One or more GPUs (Graphics Processing Units) can also be configured in the distributed platform 200. Multiple superchips can be configured in the distributed platform 200, connected to the CPUs and GPUs via interconnects. This interconnect can be a memory-coherent interconnect capable of high-frequency bandgap and low latency. Thus, the distributed platform 200 can have both CPU and GPU resources as computing resources.

[0039] Figure 2 , Figure 3 This is an explanatory diagram used to illustrate the processing content of the management platform 100. Figure 2 , Figure 3 In the example shown, multiple distributed platforms 200 execute MEC applications, and multiple UEs 30 connect to their MEC applications. The MEC application can be an application running on the MEC. The MEC application does not have to be a RAN control application like a RIC, but rather an application that connects to the UE 30 and provides arbitrary services to the UE 30 and its owner. Figure 2 An example illustrating the conditions during the day. Figure 3 An example illustrating the conditions at night.

[0040] The individual needs of the MEC applications on the multiple distributed platforms 200 may vary depending on their location. For example, in factory areas, the demand for MEC applications such as Computer Vision (CV) is considered high. Furthermore, in office areas, the demand for MEC applications such as Natural Language Processing (NLP) is considered high. Additionally, in financial areas, the demand for MEC applications such as Structured Data Analysis is considered high. Moreover, in residential areas, the demand for MEC applications such as Natural Language Processing and Cloud Gaming (CGaming) is considered high. These needs may fluctuate daily, hourly, or depending on weather, events, etc., in the areas where the distributed platforms 200 are deployed.

[0041] As a specific example, consider: Figure 2 , 3 As shown, during the day, the demand for MEC applications with high latency requirements, such as image recognition, increases in factory areas, while the demand for MEC applications with low latency requirements, such as natural language systems, increases in residential areas. At night, compared to the daytime, the demand for MEC applications in factory areas decreases, while the demand for MEC applications with high latency requirements, such as cloud gaming, increases in residential areas.

[0042] Considering this situation, for example, the following control measures could be taken: During the day, financial AIMEC applications are deployed on distributed platform 200 near the financial district, and natural language processing (NLP) MEC applications are deployed on distributed platform 200 near the residential district. Conversely, at night, cloud gaming MEC applications are deployed on distributed platform 200 near the residential district, and recommendation systems and learning instances are deployed on distributed platform 200 near the office area. This would help optimize the deployment of MEC applications across multiple distributed platforms 200.

[0043] However, in order to adjust the deployment of MEC applications, it is difficult to predict when and where the demand for which type of MEC application will be high, which is unrealistic. The management platform 100 involved in this embodiment collects various information and uses the collected information to predict future demand according to each combination of multiple distributed platforms 200 and multiple MEC applications.

[0044] exist Figure 2 , Figure 3 In the example shown, management platform 100 includes SMO 102, which performs demand forecasting using a combination of multiple distributed platforms 200 and multiple MEC applications employing demand forecasting model 104. It should be noted that management platform 100 can also perform demand forecasting without SMO 102.

[0045] SMO102 can collect a metric 250 for each of the multiple MEC applications executed in multiple distributed platforms 200, including the number of connected terminals representing the number of UEs 30 connected to the MEC application.

[0046] exist Figure 2 , 3In the example shown, SMO102 collects a metric 250 for each of the multiple MEC applications executed by multiple distributed platforms 200, including the number of connected terminals representing the number of UEs 30 connected to the MEC application. Metric 250 may also include throughput between the UEs 30 and the MEC application. Metric 250 may also include communication latency between the UEs 30 and the MEC application. Metric 250 may also include the handover history of the UEs 30 during the use of the MEC application. Metric 250 may also include information beyond these related to the needs of the MEC application.

[0047] In addition, SMO 102 collects region association information 450 from system 400. Region association information 450 is information associated with the region where distributed platform 200 is configured. Region association information 450 may include weather information indicating the weather in the region where distributed platform 200 is configured. System 400 may include servers providing weather information for various locations, from which SMO 102 can collect weather information. Region association information 450 may also include event information related to events occurring in the region where distributed platform 200 is configured. System 400 may include servers providing event information for various locations, from which SMO 102 can collect event information.

[0048] exist Figure 2 In the example shown, MEC_APP_A202 is executed in the distributed platform 200 corresponding to the factory area, and MEC_APP_B204 is executed in the distributed platform 200 corresponding to the residential area. SMO102 collects a metric 250 including the number of connected terminals indicating the number of UEs 30 connected to MEC_APP_A202, and a metric 250 including the number of connected terminals indicating the number of UEs 30 connected to MEC_APP_B204. Furthermore, SMO102 collects area association information 450 from system 400.

[0049] SMO102 generates or updates the demand forecasting model 104 by using collected metrics 250 and regional correlation information 450, and is able to predict the daytime demand increase of MEC_APP_A202 in the factory area and the demand increase of MEC_APP_B204 in the residential area.

[0050] exist Figure 3In the example shown, MEC_APP_C206 is executed in the distributed platform 200 corresponding to the factory area, and MEC_APP_D208 is executed in the distributed platform 200 corresponding to the residential area. SMO102 collects a metric 250 including the number of connected terminals indicating the number of UEs 30 connected to MEC_APP_C206, and a metric 250 including the number of connected terminals indicating the number of UEs 30 connected to MEC_APP_D208. Furthermore, SMO102 collects area association information 450 from system 400.

[0051] SMO102 generates or updates the demand forecasting model 104 by using collected metrics 250 and regional correlation information 450, and is able to predict the increase in demand at night in the factory area MEC_APP_C206 and the increase in demand in the residential area MEC_APP_D208.

[0052] Figure 4 This is an illustrative diagram used to explain learning based on the management platform 100. Figure 4 In the example shown, the management platform 100 generates a demand forecasting model 104, which takes as input the metrics 250 of each combination of multiple distributed platforms 200 and multiple MEC applications for each predetermined period in the past, and the regional association information 450, and outputs the number of connected UEs and the required throughput for each combination of multiple distributed platforms 200 and multiple MEC applications in the future.

[0053] As a predetermined period, such as 15 minutes, it can be set, and it can also be changed after being set.

[0054] The demand forecasting model 104 can be any model as long as it can take the metric 250 and the region association information 450 as inputs and take a portion of the metric 250, such as the number of connected UEs and throughput, as outputs. For example, the demand forecasting model 104 can be a neural network that takes past metric 250 and region association information 450 as input layers, takes a portion of future metric 250 as output layers, and includes one or more intermediate layers.

[0055] Figure 5 This is a simplified example of the functional configuration of the management platform 100. The management platform 100 includes a storage unit 110, a management unit 112, a measurement acquisition unit 114, a region-related information acquisition unit 116, a learning execution unit 118, a demand forecasting unit 120, a forecast result output unit 122, and an adjustment unit 124. It should be noted that the management platform 100 may not possess all of these components.

[0056] Management unit 112 manages multiple distributed platforms 200. Management unit 112 can manage the status of multiple distributed platforms 200. Management unit 112 can send various instructions to multiple distributed platforms 200, etc.

[0057] The metric acquisition unit 114 acquires metrics 250 according to each combination of multiple distributed platforms 200 and multiple MEC applications. The metric acquisition unit 114 stores the acquired metrics 250 in the storage unit 110.

[0058] The metric 250 may include the number of connected terminals representing the number of UEs 30 connected to the MEC application. The metric acquisition unit 114 acquires the number of connected terminals, for example, from the MEC application. The metric acquisition unit 114 acquires the number of connected terminals, for example, by collecting information about the MEC applications connected to the UEs 30 from multiple UEs 30. The metric acquisition unit 114 acquires the number of connected terminals, for example, from the core network of the mobile communication system.

[0059] Metric 250 can include the communication status between the MEC application and UE30.

[0060] For example, metric 250 includes the throughput between UE30 and the MEC application. Metric acquisition unit 114 can, for example, acquire the throughput between UE30 and the MEC application from the distributed platform 200 that is executing the MEC application. Metric acquisition unit 114 can also acquire the throughput between UE30 and the MEC application from UE30.

[0061] For example, metric 250 includes the communication latency between UE30 and the MEC application. Metric acquisition unit 114 can, for example, acquire the communication latency between UE30 and the MEC application from the distributed platform 200 executing the MEC application. Metric acquisition unit 114 can acquire the communication latency between UE30 and the MEC application from UE30.

[0062] Measurement 250 may also include the handover history of UE30 during its connection to the MEC application. Measurement acquisition unit 114 can acquire the handover history of UE30 during its connection to the MEC application by acquiring handover information of UE30 and information of the MEC application to which UE30 is connected. Measurement acquisition unit 114 can acquire handover information of UE30 from distributed platform 200. Measurement acquisition unit 114 can acquire handover information of UE30 from the core network of the mobile communication system. Measurement acquisition unit 114 can also acquire handover information of UE30 from UE30. Measurement acquisition unit 114 can acquire information of the MEC application to which UE30 is connected from distributed platform 200. Measurement acquisition unit 114 can also acquire information of the MEC application to which UE30 is connected from UE30.

[0063] The region association information acquisition unit 116 acquires region association information 450 for each of the plurality of distributed platforms 200, which is associated with the region configured with the distributed platform 200. The region association information acquisition unit 116 stores the acquired region association information in the storage unit 110. The storage unit 110 can store the metrics 250 and the region association information 450 for each combination of the plurality of distributed platforms 200 and the plurality of MEC applications as learning data.

[0064] The regional association information 450 may include weather information representing the weather in the region where the distributed platform 200 is configured. The regional association information acquisition unit 116 may acquire weather information from a server that provides weather information for various locations. The regional association information acquisition unit 116 may also acquire weather information from a weather management server that collects and manages weather information for various locations.

[0065] The regional association information 450 may include event information related to events implemented in regions where the distributed platform 200 is configured. The regional association information acquisition unit 116 may acquire event information from a server that provides event information from various locations. The regional association information acquisition unit 116 may also acquire event information from an event management server that collects and manages event information from various locations.

[0066] The learning execution unit 118 generates a machine learning model that predicts the demand for each combination of multiple distributed platforms 200 and multiple MEC applications by executing machine learning using learning data stored in the storage unit 110. This model is based on metrics 250 and regional association information 450 for each combination of multiple distributed platforms 200 and multiple MEC applications. The learning execution unit 118 stores the generated machine learning model in the storage unit 110.

[0067] The learning execution unit 118 can generate a machine learning model that can predict the future number of connected terminals by learning from the number of connected terminals. The learning execution unit 118 can generate a machine learning model that can predict future throughput by learning from throughput. The learning execution unit 118 can generate a machine learning model that can predict future communication latency by learning from communication latency. The learning execution unit 118 can generate a machine learning model that can predict the mobility trends of users connected to a certain MEC application by learning from the handover history. For example, during the morning hours of Monday, if there is a handover history of many user terminals connected to MEC application A switching from the wireless base stations 300 under the first distributed platform 200 to the wireless base stations 300 under the second distributed platform 200 while connected to MEC application A, this handover history can be used for learning. Therefore, during the morning hours of Monday, if there are many user terminals connected to MEC application A under the wireless base stations of the first distributed platform 200, a machine learning model can be generated that can predict an increase in demand for MEC application A in the second distributed platform 200.

[0068] The learning execution unit 118 can generate machine learning models that take into account the different predictions made by the MEC application being used due to different weather conditions by using weather information for learning. For example, in a certain area, if there are more user terminals connected to MEC application A on sunny days and more user terminals connected to MEC application B on rainy days, a machine learning model can be generated as follows: taking this trend into account, for the distributed platform 200 in that area, it is predicted that the demand for MEC application A will increase on sunny days and the demand for MEC application B will increase on rainy days.

[0069] The learning execution unit 118 can generate a machine learning model by using event information for learning, which can make different predictions about the MEC application used, taking into account the presence or absence of an event and the different types of events held. For example, in a certain region, if there are more user terminals connected to MEC application A on days without events, more user terminals connected to MEC application B on days with music-related events, and more user terminals connected to MEC application C on days with sports-related events, the machine learning model can be generated as follows: taking this trend into account, for the distributed platform 200 in that region, it can be predicted that the demand for MEC application A will increase on days without events, the demand for MEC application B will increase on days with music events, and the demand for MEC application C will increase on days with sports-related events.

[0070] The learning execution unit 118 can generate a machine learning model by using multiple or all of the following information for learning: date and time, number of connected terminals, throughput, communication latency, handover history, weather information, and event information. This model can comprehensively consider this information to predict the needs of each combination of multiple distributed platforms 200 and multiple MEC applications.

[0071] The learning execution unit 118 generates a machine learning model as follows: taking as input measures 250 and regional association information 450 for each combination of multiple distributed platforms 200 and multiple MEC applications from multiple past periods, and outputting at least a portion of the measures 250 for future periods. This period can be in units such as 15 minutes, hours, days, or any other unit.

[0072] As a specific example, the learning execution unit 118 generates a machine learning model as follows: taking the number of connected terminals, throughput, communication latency, and handover history and regional association information 450 of each combination of multiple distributed platforms 200 and multiple MEC applications in each of the past multiple periods as input, and taking the number of connected terminals and throughput of future periods as output.

[0073] The demand forecasting unit 120 forecasts future demand based on metrics 250 and regional correlation information 450, according to each combination of multiple distributed platforms 200 and multiple MEC applications.

[0074] For example, the demand forecasting unit 120 inputs the metrics 250 and regional association information 450 for each combination of multiple distributed platforms 200 and multiple MEC applications from each of multiple past periods into a machine learning model generated by the learning execution unit 118 to predict the demand for each combination of multiple distributed platforms 200 and multiple MEC applications in future periods. For example, the demand forecasting unit 120 inputs the metrics 250 and regional association information 450 for each combination of multiple distributed platforms 200 and multiple MEC applications from each of multiple past periods into a machine learning model to obtain the number of connected terminals output from the machine learning model as the demand for each combination of multiple distributed platforms 200 and multiple MEC applications in future periods.

[0075] The demand forecasting unit 120 can input the metrics 250 of each combination of multiple distributed platforms 200 and multiple MEC applications for each of the past multiple periods and the regional association information 450 into the machine learning model to obtain the number of connected terminals and throughput output from the machine learning model as the predicted number of connected terminals and the required throughput for each combination of multiple distributed platforms 200 and multiple MEC applications for future periods.

[0076] The forecast result output unit 122 outputs the forecast results of the demand forecast unit 120. The forecast result output unit 122 can display the forecast results of the demand forecast unit 120. For example, the forecast result output unit 122 displays the forecast results of the demand forecast unit 120 on a display provided by the management platform 100. The forecast result output unit 122 can also send the forecast results of the demand forecast unit 120. For example, the forecast result output unit 122 sends the forecast results of the demand forecast unit 120 to an external device.

[0077] Based on the forecast results of the demand forecasting unit 120, the adjustment unit 124 adjusts the deployment status of the respective MEC applications on the multiple distributed platforms 200. For example, based on the future demand of each combination of the multiple distributed platforms 200 and the multiple MEC applications predicted by the demand forecasting unit 120, the adjustment unit 124 may deploy MEC applications on the multiple distributed platforms 200 respectively, change the applications deployed among the multiple distributed platforms 200, or change the amount of resources allocated to the MEC applications on the multiple distributed platforms 200 respectively.

[0078] Figure 6An example of the hardware configuration of a computer 1200 that functions as a management platform 100 or a distributed platform 200 is shown in schematic form. A program installed on the computer 1200 enables the computer 1200 to function as one or more "parts" of the apparatus according to this embodiment, or to perform operations associated with the apparatus or the one or more "parts" according to this embodiment, and / or to perform processes or stages of processes according to this embodiment. Such a program can be executed by the CPU 1212 to cause the computer 1200 to perform specified operations associated with several or all of the blocks in the flowcharts and block diagrams described in this specification.

[0079] The computer 1200 based on this embodiment includes a CPU 1212, a GPU 1213, a RAM 1214, and a graphics controller 1216, which are interconnected via a host controller 1210. The computer 1200 also includes input / output units such as a communication interface 1222, a storage device 1224, a DVD drive 1226, and an IC card driver, which are connected to the host controller 1210 via an input / output controller 1220. The DVD drive 1226 can be a DVD-ROM drive or a DVD-RAM drive, etc. The storage device 1224 can be a hard disk drive or a solid-state drive, etc. The computer 1200 also includes conventional input / output units such as a ROM 1230 and a keyboard, which are connected to the input / output controller 1220 via an input / output chip 1240.

[0080] CPU 1212 operates according to the program stored in ROM 1230 and RAM 1214, thereby controlling each unit. Graphics controller 1216 obtains image data generated by CPU 1212 from the frame buffer provided in RAM 1214 or from itself, and displays the image data on display device 1218.

[0081] Communication interface 1222 communicates with other electronic devices via a network. Storage device 1224 stores programs and data used by CPU 1212 within computer 1200. DVD drive 1226 reads programs or data from DVD-ROM 1227, etc., and provides them to storage device 1224. IC card driver reads programs and data from IC card and / or writes programs and data to IC card.

[0082] ROM 1230 stores the boot program executed by computer 1200 during activation, and / or programs dependent on the hardware of computer 1200. Input / output chip 1240 can also connect various input / output units to input / output controller 1220 via USB port, parallel port, serial port, keyboard port, mouse port, etc.

[0083] The program is provided by a computer-readable storage medium such as a DVD-ROM 1227 or an IC card. The program is read from the computer-readable storage medium, installed in a storage device 1224, RAM 1214, or ROM 1230 (also examples of computer-readable storage media), and executed by the CPU 1212. The information processing described within this program is read by the computer 1200, resulting in cooperation between the program and the aforementioned types of hardware resources. The apparatus or method can be configured to perform information manipulation or processing according to the use of the computer 1200.

[0084] For example, when communication is performed between computer 1200 and external devices, CPU 1212 can execute a communication program loaded in RAM 1214, and based on the processing described in the communication program, command communication processing to communication interface 1222. Under the control of CPU 1212, communication interface 1222 reads transmission data stored in the transmission buffer area provided in recording medium such as RAM 1214, storage device 1224, DVD-ROM 1227, or IC card, and sends the read transmission data to the network, or writes received data received from the network into the receive buffer area provided on the recording medium, etc.

[0085] Furthermore, the CPU 1212 can read all or a portion of files or databases stored in external recording media such as storage devices 1224, DVD drives 1226 (DVD-ROM 1227), and IC cards from the RAM 1214, and perform various types of processing on the data in the RAM 1214. The CPU 1212 can then write the processed data back to the external recording medium.

[0086] Various types of information, such as programs, data, tables, and databases, can be stored in the recording medium and processed. The CPU 1212 can perform various types of processing on data read from the RAM 1214, including various types of operations specified by the program's command sequence as described throughout this disclosure, information processing, conditional judgment, conditional branching, unconditional branching, information retrieval, replacement, etc., and write the results back to the RAM 1214. Furthermore, the CPU 1212 can retrieve information from files, databases, etc., within the recording medium. For example, when multiple entries, each having an attribute value associated with a second attribute, are stored in the recording medium, the CPU 1212 can retrieve from these multiple entries an entry that matches the condition specified by the first attribute value, read the attribute value of the second attribute stored in that entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0087] The programs or software modules described above can be stored on or near computer-readable storage media on computer 1200. Alternatively, a recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can be used as a computer-readable storage medium, thereby providing the program to computer 1200 via the network.

[0088] In this embodiment, the boxes in the flowcharts and block diagrams may represent stages of a process for performing an operation or "parts" of a device that performs the operation. The designated stages and "parts" may be installed by dedicated circuitry, programmable circuitry supplied along with computer-readable commands stored on a computer-readable storage medium, and / or a processor supplied along with computer-readable commands stored on a computer-readable storage medium. Dedicated circuitry may include digital and / or analog hardware circuitry, and may also include integrated circuits (ICs) and / or discrete circuitry. Programmable circuitry may include reconfigurable hardware circuitry, including logic products, logical sums, XOR, negation logical products, negation logical sums, and other logical operations, flip-flops, registers, and memory elements such as field-programmable gate arrays (FPGAs) and programmable logic arrays (PLAs).

[0089] Computer-readable storage media can include any tangible device capable of storing commands that can be executed by a suitable device. As a result, a computer-readable storage medium having commands stored therein comprises an executable product including commands for creating units for performing operations specified in a flowchart or block diagram. Examples of computer-readable storage media include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, etc. More specific examples of computer-readable storage media include floppy disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), optical disc read-only memory (CD-ROM), digital multipurpose disc (DVD), Blu-ray disc, memory stick, integrated circuit card, etc.

[0090] Computer-readable commands may include any one of source code or object code described in any combination of one or more programming languages, including assembly commands, command set architecture (ISA) commands, machine commands, machine-dependent commands, microcode, firmware commands, state setting data, or object-oriented programming languages ​​such as Smalltalk (registered trademark), JAVA (registered trademark), C++, and conventional procedural programming languages ​​such as the "C" programming language or similar programming languages.

[0091] To execute computer-readable commands in order for the processor of a programmable data processing device such as a computer, or for programmable circuitry to generate units for performing operations specified in a flowchart or block diagram, the computer-readable commands are provided via a local area network (LAN), wide area network (WAN) such as the Internet, to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device. Here, a computer can be a PC (personal computer), tablet computer, smartphone, workstation, server computer, general-purpose computer, or special-purpose computer, or it can be a computer system connecting multiple computers. Such a computer system connecting multiple computers is also called a distributed computing system, which is a computer in a broad sense. In a distributed computing system, multiple computers each execute a part of a program, transferring data between computers as needed during program execution, thereby enabling multiple computers to collectively execute the program.

[0092] Examples of processors include computer processors, central processing units (CPUs), processing units, microprocessors, digital signal processors, controllers, and microcontrollers. A computer can have one or more processors. In a multiprocessor system with multiple processors, each processor executes a portion of the program, transferring data between processors as needed, thus enabling multiple processors to collectively execute the program. For example, in multitasking, multiple processors execute a portion of each task by switching tasks according to time slices. In this case, which part of the program each processor executes is dynamically changing. Alternatively, the specific part of the program executed by each processor can be statically determined by understanding the multiprocessor programming.

[0093] The present invention has been described above using embodiments; however, the scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications or improvements can be made to the above embodiments. It is obvious, as described in the claims, that forms with such modifications or improvements can be included within the scope of the present invention.

[0094] It is important to note that the execution order of actions, processes, steps, and stages in the apparatus, systems, programs, and methods shown in the claims, description, and drawings can be arbitrary unless explicitly stated as "before" or "in advance," and the output of a preceding process is not used in a subsequent process. Regarding the flow of actions in the claims, description, and drawings, even if terms such as "firstly," "next," etc., are used for ease of explanation, this does not imply that the actions must be performed in that specific order.

[0095] Explanation of reference numerals in the attached figures

[0096] 10 System, 30 UE, 100 Management Platform, 102 SMO, 104 Demand Forecasting Model, 110 Storage Department, 112 Management Department, 114 Measurement Acquisition Department, 116 Regional Association Information Acquisition Department, 118 Learning Execution Department, 120 Demand Forecasting Department, 122 Forecast Result Output Department, 124 Adjustment Department, 200 Distributed Platform, 202 MEC_APP_A, 204 MEC_APP_B, 206 MEC_APP_C, 208 MEC_APP_D, 250 Measurement, 300 Wireless Base Station, 310 RAN, 400 System, 450 Regional Association Information, 1200 Computer, 1210 Main Controller, 1212 CPU, 1213 GPU, 1214 RAM, 1216 Graphics Controller, 1218 Display Device, 1220 Input / Output Controller, 1222 Communication Interface, 1224 Storage devices, 1226 DVD drive, 1227 DVD-ROM, 1230 ROM, 1240 I / O chip

Claims

1. A management platform, comprising: The management department, located in various locations, manages multiple distributed platforms capable of running multiple MEC applications; The measurement acquisition unit acquires a metric for each combination of the plurality of distributed platforms and the plurality of MEC applications, including a metric representing the number of user terminals connected to the MEC applications. The regional association information acquisition unit acquires regional association information with the region configured with the distributed platform for each of the plurality of distributed platforms; as well as The demand forecasting unit forecasts future demand based on the metrics and the regional association information, according to each combination of the plurality of distributed platforms and the plurality of MEC applications.

2. The management platform according to claim 1, wherein, The metric acquisition unit acquires the metric, which also includes the communication status between the MEC application and the user terminal, according to each combination of the plurality of distributed platforms and the plurality of MEC applications.

3. The management platform according to claim 2, wherein, The metric acquisition unit acquires, according to each combination of the plurality of distributed platforms and the plurality of MEC applications, the metric, which also includes at least one of the throughput and communication latency between the MEC application and the user terminal.

4. The management platform according to claim 2, wherein, The metric acquisition unit acquires the metric, according to each combination of the plurality of distributed platforms and the plurality of MEC applications, a HO history that also represents the history of user terminal switching during connection with the MEC application.

5. The management platform according to claim 1, wherein, The regional association information acquisition unit acquires regional association information for each of the plurality of distributed platforms, including weather information and event information for the region where the distributed platform is configured.

6. The management platform according to any one of claims 1 to 5, comprising: The storage unit stores the metrics and region association information acquired by the metric acquisition unit and the region association information acquisition unit for each combination of the plurality of distributed platforms and the plurality of MEC applications as learning data; and The learning execution unit, by executing machine learning using the learning data stored in the storage unit, generates a machine learning model that predicts the demand for each combination of the multiple distributed platforms and the multiple MEC applications based on metrics and regional association information for each combination of the multiple distributed platforms and the multiple MEC applications. The demand forecasting unit inputs the metrics and regional association information of each combination of the multiple distributed platforms and the multiple MEC applications for each of the past multiple periods into the machine learning model to predict the demand for each combination of the multiple distributed platforms and the multiple MEC applications for future periods.

7. The management platform according to claim 6, wherein, The machine learning model takes as input the metrics and regional association information of each combination of the multiple distributed platforms and the multiple MEC applications, and outputs the number of connected terminals for each combination of the multiple distributed platforms and the multiple MEC applications. The demand forecasting unit inputs the metrics and regional association information of each combination of the multiple distributed platforms and the multiple MEC applications for each of the multiple past periods into the machine learning model, and obtains the number of connected terminals output from the machine learning model as the demand for each combination of the multiple distributed platforms and the multiple MEC applications for future periods.

8. The management platform according to any one of claims 1 to 5 further comprises: The adjustment unit adjusts the deployment status of the respective MEC applications on the multiple distributed platforms based on the forecast results of the demand forecasting unit.

9. A management method, wherein the management method is executed by a management platform, the management platform being configured in various locations to manage multiple distributed platforms capable of executing multiple MEC applications, the management method comprising: During the metric acquisition phase, for each combination of the plurality of distributed platforms and the plurality of MEC applications, a metric including the number of connected terminals representing the number of user terminals connected to the MEC application is acquired; In the regional association information acquisition stage, for each of the multiple distributed platforms, regional association information with the regions configured with the distributed platforms is acquired; as well as In the demand forecasting phase, future demand is predicted based on the metrics and the regional association information, according to each combination of the multiple distributed platforms and the multiple MEC applications.

10. A program, Used to enable a computer to perform the management method as described in claim 9.

11. A system having: The management platform according to any one of claims 1 to 5; and The multiple distributed platforms.

Images