Management base, management method, program, and system

Abstract

Provided is a management base provided with: a management unit that manages a plurality of distributed bases disposed in individual regions and each capable of executing a plurality of MEC applications; a metric acquisition unit that acquires, for each combination of the plurality of distributed bases and the plurality of MEC applications, metrics including the number of connected terminals, which indicates the number of user terminals connected to the MEC application; a region-related information acquisition unit that acquires, for each of the plurality of distributed bases, region-related information related to the region in which the distributed base is disposed; and a demand prediction unit that predicts a future demand for each combination of the plurality of distributed bases and the plurality of MEC applications on the basis of the metrics and the region-related information. The distributed bases may have a RAN control function and an AI processing (RAN intelligent controller (RIC) or the like) function.

Description

Management infrastructure, management method, program, and system

[0001] The present invention relates to a management infrastructure, a management method, a program, and a system.

[0002] Patent Document 1 describes that in a management node that manages a distributed file / object storage that manages files used by an application in an accessible manner, the distributed file / object storage is accessible to files managed in storage at other bases, the management node has a processor, and is configured to identify an access status for the file by the application and, based on the access status, control caching by the distributed file / object storage at its own base for files used by the application that are managed in storage at other bases before the application is executed. [Prior Art Documents] [Patent Documents] [Patent Document 1] JP 2023-069701 A

[0003] Conventionally, MEC applications executed on MEC (Multi-access Edge Computing) generally have a static execution region, and MEC application providers select the region in which to deploy the MEC application, taking into account delays and fault tolerance. On the other hand, in the case of distributed platforms such as regionally distributed DCs (Data Centers), each of which has a limited scale, or MEC AI applications that require large-scale computational resources, a method of statically allocating resources for all MEC AI applications to all distributed platforms does not allow for effective use of computational resources. Therefore, it is important to predict in advance how much MEC application demand will occur in each region.

[0004] In the system according to the present embodiment, for example, future demand prediction for each region of each MEC application is performed using past RAN (Radio Access Network) metrics and region-related information related to the region where the past RAN was provided. The system may perform the demand prediction using a machine learning model. The RAN metrics may include the number of connected UEs (User Equipment), throughput, delay, and handover history for each distributed infrastructure / MEC application. The region-related information may include event information and weather. Predicted demand items may include the number of connected UEs in the future and the throughput required in the future. For example, a time-series machine learning model may be trained to predict demand items for each distributed infrastructure / MEC application from various RAN metrics / region-related information for each distributed infrastructure / MEC application, and the model may be served on a management infrastructure that manages multiple distributed infrastructures.

[0005] As a specific example, the system 10 according to the present embodiment includes a distributed infrastructure having a RAN control function for controlling the RAN and an AI processing function for performing AI processing, and a management infrastructure for managing the multiple distributed infrastructures. The management infrastructure performs demand forecasting for each combination of the multiple distributed infrastructures and the multiple MEC applications. Based on the forecast results, the management infrastructure may adjust the deployment status of the MEC applications or perform communication control.

[0006] Types of AI processing include AI processing related to RAN control (sometimes referred to as RAN-controlled AI processing) and AI processing not related to RAN control (sometimes referred to as non-RAN-controlled AI processing).

[0007] An example of RAN control AI processing is RIC (RAN Intelligent Controller). RIC is a technology that uses AI to optimize RAN radio resources and automate RAN operations. RIC includes Non-RT RIC (Non-Real Time RIC) and Near-RT RIC (Near-Real Time RIC). Non-RT RIC is sometimes called Centralized RIC. Non-RT RIC is located inside SMO (Service Management and Orchestration), which manages and orchestrates the RAN. Non-RT RIC generates and notifies policies related to RAN control and sends information to Near-RT RIC. For example, the Non-RT RIC performs machine learning using data collected from the RAN to generate a trained model for RAN control and transmits it to the Near-RT RIC. The Near-RT RIC is sometimes called a Distributed RIC. Compared to the Non-RT RIC, the Near-RT RIC is located closer to the RAN nodes (RU (Radio Unit), DU (Distributed Unit), CU (Central Unit)) and controls the RAN nodes, resources, etc. The Near-RT RIC performs processing with higher real-time performance than the Non-RT RIC. The Near-RT RIC performs inference processing related to RAN control using, for example, a trained model acquired from the Non-RT RIC. RAN control AI processing is not limited to the RIC.

[0008] Non-RAN controlled AI processes may be so-called MEC AI applications, which include any AI learning and inference processes that are not related to RAN control.

[0009] According to one embodiment of the present invention, there is provided a management platform. The management platform may include a management unit that manages multiple distributed platforms deployed in various locations, each capable of executing multiple MEC applications. The management platform may include a metrics acquisition unit that acquires, for each combination of the multiple distributed platforms and the multiple MEC applications, metrics including a number of connected terminals indicating the number of user terminals connected to an MEC application. The management platform may include a region-related information acquisition unit that acquires, for each of the multiple distributed platforms, region-related information related to the region in which the distributed platform is deployed. The management platform may include a demand forecasting unit that predicts future demand for each combination of the multiple distributed platforms and the multiple MEC applications based on the metrics and the region-related information.

[0010] In the management infrastructure, the metrics acquisition unit may acquire the metrics further including a communication status between an MEC application and a user terminal for each combination of the plurality of distributed infrastructures and the plurality of MEC applications. The metrics acquisition unit may acquire the metrics further including at least one of a throughput and a communication delay between an MEC application and a user terminal for each combination of the plurality of distributed infrastructures and the plurality of MEC applications.

[0011] In any of the management platforms, the metrics acquisition unit may acquire, for each combination of the multiple distributed platforms and the multiple MEC applications, the metrics further including a HO history indicating the handover history of the user terminal while connected to the MEC application.

[0012] In any of the management platforms, the regional-related information acquisition unit may acquire, for each of the plurality of distributed platforms, the regional-related information including weather information and event information for the region in which the distributed platform is located.

[0013] Any of the management platforms may include a memory unit that stores the metrics and region-related information for each combination of the multiple distributed platforms and the multiple MEC applications acquired by the metrics acquisition unit and the region-related information acquisition unit as learning data, and a learning execution unit that generates a machine learning model that predicts demand for each combination of the multiple distributed platforms and the multiple MEC applications from the metrics and region-related information for each combination of the multiple distributed platforms and the multiple MEC applications by performing machine learning using the learning data stored in the memory unit, and the demand prediction unit may input the metrics and region-related information for each combination of the multiple distributed platforms and the multiple MEC applications for each of a plurality of past periods into the machine learning model to predict demand for each combination of the multiple distributed platforms and the multiple MEC applications for a future period. The machine learning model may input metrics and regional information for each combination of the multiple distributed platforms and the multiple MEC applications, and output the number of connected terminals for each combination of the multiple distributed platforms and the multiple MEC applications, and the demand forecasting unit may input the metrics and regional information for each combination of the multiple distributed platforms and the multiple MEC applications for each of multiple past periods into the machine learning model, and obtain the number of connected terminals output from the machine learning model as demand for each combination of the multiple distributed platforms and the multiple MEC applications for a future period.

[0014] Any one of the management platforms may further include an adjustment unit that adjusts the deployment status of each MEC application on the multiple distributed platforms based on the prediction results by the demand prediction unit.

[0015] According to one embodiment of the present invention, there is provided a management method executed by a management platform that manages multiple distributed platforms deployed in various locations, each capable of executing multiple MEC applications. The management method may include a metrics acquisition step of acquiring, for each combination of the multiple distributed platforms and the multiple MEC applications, metrics including a number of connected terminals indicating the number of user terminals connected to an MEC application. The management method may also include a region-related information acquisition step of acquiring, for each of the multiple distributed platforms, region-related information related to the region in which the distributed platform is deployed. The management method may also include a demand forecasting step of forecasting future demand for each combination of the multiple distributed platforms and the multiple MEC applications based on the metrics and the region-related information.

[0016] According to one embodiment of the present invention, there is provided a program for causing a computer to execute the management method.

[0017] According to one embodiment of the present invention, there is provided a system comprising the management infrastructure and the plurality of distribution infrastructures.

[0018] The above summary of the invention does not list all of the necessary features of the present invention, and subcombinations of these features may also constitute inventions.

[0019] FIG. 1 is a schematic diagram illustrating an example of a system 10. FIG. 2 is an explanatory diagram illustrating processing content by the management infrastructure 100. FIG. 3 is an explanatory diagram illustrating learning by the management infrastructure 100. FIG. 4 is a schematic diagram illustrating an example of the functional configuration of the management infrastructure 100. FIG. 5 is a schematic diagram illustrating an example of the hardware configuration of a computer 1200 that functions as the management infrastructure 100 or the distribution infrastructure 200.

[0020] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention as claimed. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0021] 1 schematically illustrates an example of a system 10. The system 10 includes a management infrastructure 100 and a plurality of distributed infrastructures 200. In the system 10, the management infrastructure 100 and the plurality of distributed infrastructures 200 may cooperate to control a RAN 310 and perform AI processing. The RAN 310 provides mobile communication services to a UE 30. The UE 30 may be an example of a user terminal.

[0022] The RAN 310 may be a virtualized vRAN (Virtual RAN), and the system 10 may control the vRAN. The RAN 310 may be a physical RAN, and the system 10 may control the physical RAN. In this embodiment, a case where the RAN 310 is a vRAN will be mainly described as an example.

[0023] The AI ​​processing performed by the system 10 may include RAN-controlled AI processing (sometimes referred to as RAN_AI). The AI ​​processing performed by the system 10 may include non-RAN-controlled AI processing (sometimes referred to as non-RAN_AI).

[0024] The distributed infrastructure 200 may be a data center located in various locations. The distributed infrastructure 200 may be configured with multiple devices. The distributed infrastructure 200 may be realized on a virtualization infrastructure consisting of multiple devices. The distributed infrastructure 200 may be realized by a single device. That is, the distributed infrastructure 200 may be a distributed device. The distributed infrastructure 200 may function as a BBU (BaseBand Unit), and the radio base station 300 may function as an RRU (Remote Radio Unit). The distributed infrastructure 200 may implement a CU. The distributed infrastructure 200 may implement a DU. The distributed infrastructure 200 may implement a UPF (User Plane Function).

[0025] The management infrastructure 100 may be a data center that manages multiple distributed infrastructures 200. The management infrastructure 100 may be configured with multiple devices. The management infrastructure 100 may be realized on a virtualization infrastructure made up of multiple devices. The management infrastructure 100 may also be realized by a single device. In other words, the management infrastructure 100 may be a management device.

[0026] The management infrastructure 100 may be referred to as a Core Brain, and the distributed infrastructure 200 may be referred to as a Regional Brain. While FIG. 1 illustrates an example in which a single-level management infrastructure 100 is arranged below the management infrastructure 100, this is not limiting. The distributed infrastructure 200 may have multiple levels. For example, when a two-level distributed infrastructure 200 is arranged below the management infrastructure 100, the management infrastructure 100 may be referred to as a Core Brain, the distributed infrastructure 200 at the level below that may be referred to as a Regional Brain, and the distributed infrastructure 200 at the level further below that may be referred to as a Sub-Regional Brain.

[0027] In the case where a Core Brain, a Regional Brain, and a Sub Regional Brain exist, the demand forecasting process according to this embodiment may be executed by the Sub Regional Brain or by the Regional Brain.

[0028] The distributed infrastructure 200 may be arranged with one or more central processing units (CPUs). The distributed infrastructure 200 may be arranged with one or more graphics processing units (GPUs). The distributed infrastructure 200 may be arranged with multiple super chips, each of which has a CPU and a GPU connected via an interconnect. The interconnect may have memory consistency and may be capable of achieving high bandwidth and low latency. In this way, the distributed infrastructure 200 may have CPU resources and GPU resources as computational resources.

[0029] 2 and 3 are explanatory diagrams for explaining the processing content by the management infrastructure 100. In the example shown in FIGS. 2 and 3, each of multiple distributed infrastructures 200 executes an MEC application, and multiple UEs 30 are connected to the MEC application. The MEC application may be an application executed on the MEC. The MEC application may not be an application for RAN control such as RIC, but may be an application to which the UEs 30 are connected and which provides any service to the UEs 30 and the owners of the UEs 30. FIG. 2 shows an example of a daytime situation, and FIG. 3 shows an example of a nighttime situation.

[0030] The demand for each of the multiple MEC applications in each of the multiple distributed infrastructures 200 may vary depending on the location of the distributed infrastructure 200. For example, in a factory area, there may be a high demand for image recognition (CV: Computer Vision) MEC applications. In an office district, there may be a high demand for natural language processing (NLP: Natural Language Processing) MEC applications. In a financial district, there may be a high demand for structured data analysis MEC applications. In a residential area, there may be a high demand for natural language MEC applications and cloud gaming (CGaming) MEC applications. This balance may fluctuate daily or hourly, and may also fluctuate depending on the weather, events, etc. in the area where the distributed infrastructure 200 is located.

[0031] As a specific example, as shown in Figures 2 and 3, during the daytime, there is high demand for image recognition MEC applications with high latency requirements in factory areas, and high demand for natural language MEC applications with low latency requirements in residential areas; at night, compared to the daytime, there is lower demand for MEC applications in factory areas, and higher demand for cloud gaming MEC applications with high latency requirements in residential areas.

[0032] Considering such a situation, it is conceivable to control the deployment such that, for example, during the day, a financial AIMEC application is deployed on a distributed infrastructure 200 close to a financial district, and a natural language MEC application is deployed on a distributed infrastructure 200 close to a residential area, and conversely, at night, a cloud gaming MEC application is deployed on a distributed infrastructure 200 close to a residential area, and a recommendation system and learning instances are deployed on a distributed infrastructure 200 close to an office district. This can contribute to optimizing the deployment status of MEC applications on multiple distributed infrastructures 200.

[0033] However, it is difficult and unrealistic for a person to predict where and when demand for what kind of MEC application will increase in order to adjust the deployment of the MEC application, etc. The management infrastructure 100 according to this embodiment collects various information and uses the collected information to predict future demand for each combination of multiple distributed infrastructures 200 and multiple MEC applications.

[0034] 2 and 3 , the management infrastructure 100 includes an SMO 102, which executes demand forecasting for each combination of multiple distributed infrastructures 200 and multiple MEC applications using a demand forecasting model 104. Note that the management infrastructure 100 may execute demand forecasting without including the SMO 102.

[0035] The SMO 102 may collect metrics 250 including the number of connected terminals indicating the number of UEs 30 connected to the MEC application for each of the multiple MEC applications running on the multiple distributed infrastructures 200.

[0036] 2 and 3 , the SMO 102 collects metrics 250 including a connected terminal count indicating the number of UEs 30 connected to the MEC application for each of the multiple MEC applications executed by the multiple distributed infrastructures 200. The metrics 250 may include the throughput between the UE 30 and the MEC application. The metrics 250 may include the communication delay between the UE 30 and the MEC application. The metrics 250 may include a handover history of the UE 30 while using the MEC application. The metrics 250 may also include other information related to the demand for the MEC application.

[0037] The SMO 102 also collects region-related information 450 from the system 400. The region-related information 450 is information related to the region in which the distributed infrastructure 200 is located. The region-related information 450 may include weather information indicating the weather in the region in which the distributed infrastructure 200 is located. The system 400 may include a server that provides weather information for various locations, and the SMO 102 may collect the weather information from the server. The region-related information 450 may include event information related to events being held in the region in which the distributed infrastructure 200 is located. The system 400 may include a server that provides event information for various locations, and the SMO 102 may collect the event information from the server.

[0038] 2 , MEC_APP_A 202 is executed in the distributed infrastructure 200 corresponding to the factory area, and MEC_APP_B 204 is executed in the distributed infrastructure 200 corresponding to the residential area, and SMO 102 collects metrics 250 including the number of connected terminals indicating the number of UEs 30 connected to MEC_APP_A 202, and metrics 250 including connected terminal information indicating the number of UEs 30 connected to MEC_APP_B 204. SMO 102 also collects area-related information 450 from system 400.

[0039] By using the collected metrics 250 and area-related information 450 to generate and update the demand forecasting model 104, the SMO 102 can predict that during the day, demand for MEC_APP_A 202 will increase in the factory area and demand for MEC_APP_B 204 will increase in the residential area.

[0040] 3, MEC_APP_C 206 is executed in the distributed infrastructure 200 corresponding to the factory area, and MEC_APP_D 208 is executed in the distributed infrastructure 200 corresponding to the residential area, and SMO 102 collects metrics 250 including the number of connected terminals indicating the number of UEs 30 connected to MEC_APP_C 206, and metrics 250 including connected terminal information indicating the number of UEs 30 connected to MEC_APP_D 208. SMO 102 also collects area-related information 450 from system 400.

[0041] By using the collected metrics 250 and area-related information 450 to generate and update the demand forecasting model 104, the SMO 102 can predict that demand for MEC_APP_C 206 will increase in industrial areas at night, and that demand for MEC_APP_D 208 will increase in residential areas.

[0042] Fig. 4 is an explanatory diagram for explaining learning by the management infrastructure 100. In the example shown in Fig. 4, the management infrastructure 100 receives as input metrics 250 for each combination of a plurality of distributed infrastructures 200 and a plurality of MEC applications for each predetermined period in the past, and region-related information 450, and generates a demand forecasting model 104 that outputs the number of connected UEs, required throughput, etc. for each future combination of a plurality of distributed infrastructures 200 and a plurality of MEC applications.

[0043] The predetermined period may be set to, for example, 15 minutes, and may be changeable after being set.

[0044] The demand forecasting model 104 may be any model as long as it is capable of receiving the metrics 250 and the region-related information 450 as input and outputting part of the information of the metrics 250, such as the number of connected UEs and throughput. The demand forecasting model 104 may be, for example, a neural network that uses the past metrics 250 and the region-related information 450 as an input layer, and part of the future metrics 250 as an output layer, and includes one or more intermediate layers.

[0045] 5 shows an example of the functional configuration of the management platform 100. The management platform 100 includes a storage unit 110, a management unit 112, a metrics 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. Note that it is not essential for the management platform 100 to include all of these units.

[0046] The management unit 112 manages the multiple distributed infrastructures 200. The management unit 112 may manage the status of the multiple distributed infrastructures 200. The management unit 112 may transmit various instructions to the multiple distributed infrastructures 200, for example.

[0047] The metrics acquisition unit 114 acquires metrics 250 for each combination of multiple distributed infrastructures 200 and multiple MEC applications. The metrics acquisition unit 114 stores the acquired metrics 250 in the storage unit 110.

[0048] The metrics 250 may include a number of connected terminals indicating the number of UEs 30 connected to the MEC application. The metrics acquisition unit 114 acquires the number of connected terminals, for example, from the MEC application. The metrics acquisition unit 114 acquires the number of connected terminals, for example, by collecting information on the MEC applications to which the UEs 30 are connected from a plurality of UEs 30. The metrics acquisition unit 114 acquires the number of connected terminals, for example, from the core network of the mobile communication system.

[0049] The metrics 250 may include the communication status between the MEC application and the UE 30.

[0050] For example, the metrics 250 include the throughput between the UE 30 and the MEC application. The metrics acquisition unit 114 may acquire the throughput between the UE 30 and the MEC application from, for example, the distributed infrastructure 200 that executes the MEC application. The metrics acquisition unit 114 may acquire the throughput between the UE 30 and the MEC application from the UE 30.

[0051] For example, the metrics 250 include a communication delay between the UE 30 and the MEC application. The metrics acquisition unit 114 may acquire the communication delay between the UE 30 and the MEC application from the distributed infrastructure 200 that executes the MEC application. The metrics acquisition unit 114 may acquire the communication delay between the UE 30 and the MEC application from the UE 30.

[0052] The metrics 250 may include a handover history of the UE 30 while it is connected to the MEC application. The metrics acquisition unit 114 may acquire the handover history of the UE 30 while it is connected to the MEC application by acquiring handover information of the UE 30 and information about the MEC application to which the UE 30 is connected. The metrics acquisition unit 114 may acquire the handover information of the UE 30 from the distribution infrastructure 200. The metrics acquisition unit 114 may acquire the handover information of the UE 30 from a core network of the mobile communication system. The metrics acquisition unit 114 may acquire the handover information of the UE 30 from the UE 30. The metrics acquisition unit 114 may acquire information about the MEC application to which the UE 30 is connected from the distribution infrastructure 200. The metrics acquisition unit 114 may acquire information about the MEC application to which the UE 30 is connected from the UE 30.

[0053] The region-related information acquisition unit 116 acquires, for each of the multiple distributed infrastructures 200, region-related information 450 related to the region in which the distributed infrastructure 200 is located. The region-related information acquisition unit 116 stores the acquired region-related information in the storage unit 110. The storage unit 110 may store, as learning data, the metrics 250 and the region-related information 450 for each combination of the multiple distributed infrastructures 200 and the multiple MEC applications.

[0054] The region-related information 450 may include weather information indicating the weather in the region where the distributed infrastructure 200 is located. The region-related information acquisition unit 116 may acquire weather information from a server that provides weather information for various regions. The region-related information acquisition unit 116 may also acquire weather information from a weather management server that collects and manages weather information for various regions.

[0055] The region-related information 450 may include event information about events held in the region where the distributed infrastructure 200 is located. The region-related information acquisition unit 116 may acquire the event information from a server that provides event information for each region. The region-related information acquisition unit 116 may also acquire the event information from an event management server that collects and manages event information for each region.

[0056] The learning execution unit 118 executes machine learning using the learning data stored in the storage unit 110 to generate a machine learning model that predicts demand for each combination of multiple distributed infrastructures 200 and multiple MEC applications from the metrics 250 and region-related information 450 for each combination of multiple distributed infrastructures 200 and multiple MEC applications. The learning execution unit 118 stores the generated machine learning model in the storage unit 110.

[0057] The learning execution unit 118 uses the number of connected terminals for learning, thereby generating a machine learning model capable of predicting the number of connected terminals in the future. The learning execution unit 118 uses throughput for learning, thereby generating a machine learning model capable of predicting future throughput. The learning execution unit 118 uses communication delay for learning, thereby generating a machine learning model capable of predicting future communication delay. The learning execution unit 118 uses handover history for learning, thereby generating a machine learning model capable of realizing predictions that take into account the mobility trends of users of user terminals connected to a certain MEC application. For example, if there is a history of many handovers in which user terminals that are located in the range of a radio base station 300 under the first distributed infrastructure 200 and connected to MEC application A during Monday morning hours hand over to a radio base station 300 under the second distributed infrastructure 200 while remaining connected to MEC application A, then by using this handover history for learning, it is possible to generate a machine learning model that can predict that demand for MEC application A in the second distributed infrastructure 200 will increase when there are a large number of user terminals that are located in the range of a radio base station under the first distributed infrastructure 200 and connected to MEC application A during Monday morning hours.

[0058] By using weather information for learning, the learning execution unit 118 can generate a machine learning model that can realize predictions that take into account differences in the MEC applications used due to differences in weather. For example, if in a certain region, many user terminals connect to MEC application A on sunny days and many user terminals connect to MEC application B on rainy days, taking this tendency into account, a machine learning model can be generated that can predict that, for the distributed infrastructure 200 in that region, demand for MEC application A will be high on sunny days and demand for MEC application B will be high on rainy days.

[0059] By using the event information for learning, the learning execution unit 118 can generate a machine learning model that can realize predictions that take into account differences in the MEC applications used depending on whether or not an event is held and the type of event that is held. For example, if in a certain region, on days when there are no events, many user terminals connect to MEC application A, on days when a music-related event is held, many user terminals connect to MEC application B, and on days when a sports-related event is held, a machine learning model can be generated that takes this trend into account and predicts that, for the distributed infrastructure 200 in that region, demand for MEC application A will be high on days when there are no events, that demand for MEC application B will be high on days when a music event is held, and that demand for MEC application C will be high on days when a sports-related event is held.

[0060] The learning execution unit 118 uses multiple or all of the following information for learning: date and time, number of connected terminals, throughput, communication delay, handover history, weather information, and event information, and by taking these pieces of information into consideration in a comprehensive manner, can generate a machine learning model that can predict demand for each combination of multiple distributed platforms 200 and multiple MEC applications.

[0061] The learning execution unit 118 receives the metrics 250 and the region-related information 450 for each combination of the multiple distributed infrastructures 200 and the multiple MEC applications for multiple past time periods as input, and generates a machine learning model that outputs at least a portion of the metrics 250 for a future time period. The time period may be in minutes, such as 15 minutes, hours, days, or any other unit.

[0062] As a specific example, the learning execution unit 118 takes as input the number of connected terminals, throughput, communication delay, and handover history for each combination of multiple distributed infrastructures 200 and multiple MEC applications for multiple past periods, as well as regional-related information 450, and generates a machine learning model that outputs the number of connected terminals and throughput for a future period.

[0063] The demand forecasting unit 120 forecasts future demand for each combination of multiple distribution infrastructures 200 and multiple MEC applications based on the metrics 250 and the region-related information 450 .

[0064] For example, the demand forecasting unit 120 inputs the metrics 250 and the region-related information 450 for each combination of the multiple distributed infrastructures 200 and the multiple MEC applications for each of multiple past periods into the machine learning model generated by the learning execution unit 118, and forecasts the demand for each combination of the multiple distributed infrastructures 200 and the multiple MEC applications for a future period. For example, the demand forecasting unit 120 inputs the metrics 250 and the region-related information 450 for each combination of the multiple distributed infrastructures 200 and the multiple MEC applications for each of multiple past periods into the machine learning model, and acquires the number of connected terminals output from the machine learning model as the demand for each combination of the multiple distributed infrastructures 200 and the multiple MEC applications for a future period.

[0065] The demand forecasting unit 120 may input metrics 250 and regional-related information 450 for each combination of multiple distributed infrastructures 200 and multiple MEC applications for each of multiple past periods into a machine learning model, and obtain the number of connected terminals and throughput output from the machine learning model as the predicted number of connected terminals and required throughput for each combination of multiple distributed infrastructures 200 and multiple MEC applications for a future period.

[0066] The prediction result output unit 122 outputs the prediction result by the demand prediction unit 120. The prediction result output unit 122 may display and output the prediction result by the demand prediction unit 120. For example, the prediction result output unit 122 displays the prediction result by the demand prediction unit 120 on a display provided in the management infrastructure 100. The prediction result output unit 122 may transmit and output the prediction result by the demand prediction unit 120. For example, the prediction result output unit 122 transmits the prediction result by the demand prediction unit 120 to an external device.

[0067] The adjustment unit 124 adjusts the deployment status of the MEC applications on each of the multiple distributed infrastructures 200 based on the prediction result by the demand prediction unit 120. For example, based on the future demand for each combination of the multiple distributed infrastructures 200 and the multiple MEC applications predicted by the demand prediction unit 120, the adjustment unit 124 deploys the MEC applications on each of the multiple distributed infrastructures 200, changes the applications to be deployed among the multiple distributed infrastructures 200, or changes the amount of resources allocated to the MEC applications on each of the multiple distributed infrastructures 200.

[0068] 6 schematically illustrates an example of the hardware configuration of a computer 1200 that functions as the management infrastructure 100 or the distribution infrastructure 200. A program installed on the computer 1200 can cause the computer 1200 to function as one or more "parts" of an apparatus according to the present embodiment, or can cause the computer 1200 to execute operations associated with the apparatus according to the present embodiment or one or more "parts," and / or can cause the computer 1200 to execute a process according to the present embodiment or steps of the process. Such a program can be executed by the CPU 1212 to cause the computer 1200 to perform specific operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

[0069] The computer 1200 according to this embodiment includes a CPU 1212, a GPU 1213, a RAM 1214, and a graphics controller 1216, which are interconnected by a host controller 1210. The computer 1200 also includes input / output units such as a communications interface 1222, a storage device 1224, a DVD drive 1226, and an IC card drive, which are connected to the host controller 1210 via an input / output controller 1220. The DVD drive 1226 may be a DVD-ROM drive, a DVD-RAM drive, or the like. The storage device 1224 may be a hard disk drive, a solid-state drive, or the like. The computer 1200 also includes a ROM 1230 and legacy input / output units such as a keyboard, which are connected to the input / output controller 1220 via an input / output chip 1240.

[0070] The CPU 1212 operates according to programs stored in the ROM 1230 and the RAM 1214, thereby controlling each unit. The graphics controller 1216 acquires image data generated by the CPU 1212 into a frame buffer or the like provided in the RAM 1214 or into the graphics controller 1216 itself, and causes the image data to be displayed on the display device 1218.

[0071] The communication interface 1222 communicates with other electronic devices via a network. The storage device 1224 stores programs and data used by the CPU 1212 in the computer 1200. The DVD drive 1226 reads programs or data from a DVD-ROM 1227 or the like and provides them to the storage device 1224. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0072] The ROM 1230 stores therein a boot program or the like that is executed by the computer 1200 upon activation, and / or programs that depend on the hardware of the computer 1200. The input / output chip 1240 may also connect various input / output units to the input / output controller 1220 via a USB port, a parallel port, a serial port, a keyboard port, a mouse port, etc.

[0073] The programs are provided by a computer-readable storage medium such as a DVD-ROM 1227 or an IC card. The programs are read from the computer-readable storage medium, installed in the storage device 1224, RAM 1214, or ROM 1230, which are also examples of computer-readable storage media, and executed by the CPU 1212. Information processing described in these programs is read by the computer 1200, and brings about cooperation between the programs and the various types of hardware resources described above. An apparatus or a method may be configured by implementing operations or processing of information in accordance with the use of the computer 1200.

[0074] For example, when communication is performed between computer 1200 and an external device, CPU 1212 may execute a communication program loaded into RAM 1214 and instruct communication interface 1222 to perform communication processing based on the processing described in the communication program. Under the control of CPU 1212, communication interface 1222 reads transmission data stored in a transmission buffer area provided in RAM 1214, storage device 1224, DVD-ROM 1227, or a recording medium such as an IC card, and transmits the read transmission data to a network, or writes received data received from the network to a reception buffer area or the like provided on the recording medium.

[0075] Furthermore, the CPU 1212 may cause all or a necessary portion of a file or database stored in an external recording medium such as the storage device 1224, the DVD drive 1226 (DVD-ROM 1227), an IC card, etc. to be read into the RAM 1214, and may perform various types of processing on the data on the RAM 1214. The CPU 1212 may then write back the processed data to the external recording medium.

[0076] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and may undergo information processing. The CPU 1212 may perform various types of processing on data read from the RAM 1214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search / replacement, etc., as described throughout this disclosure and specified by the instruction sequences of the programs, and write the results back to the RAM 1214. The CPU 1212 may also search for information in a file, database, etc. on the recording medium. For example, if multiple entries, each having an attribute value of a first attribute associated with an attribute value of a second attribute, are stored on the recording medium, the CPU 1212 may search for an entry whose attribute value of the first attribute matches a specified condition from among the multiple entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0077] The above-described programs or software modules may be stored in a computer-readable storage medium on or near the computer 1200. 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 also be used as a computer-readable storage medium, thereby providing the programs to the computer 1200 via the network.

[0078] The blocks in the flowcharts and block diagrams in the present embodiments may represent stages of a process in which an operation is performed or "parts" of a device responsible for performing the operation. Particular stages and "parts" may be implemented by dedicated circuitry, programmable circuitry provided with computer-readable instructions stored on a computer-readable storage medium, and / or a processor provided with computer-readable instructions stored on a computer-readable storage medium. The dedicated circuitry may include digital and / or analog hardware circuits, and may include integrated circuits (ICs) and / or discrete circuits. The programmable circuitry may include reconfigurable hardware circuits, such as field programmable gate arrays (FPGAs) and programmable logic arrays (PLAs), including AND, OR, XOR, NAND, NOR, and other logical operations, flip-flops, registers, and memory elements.

[0079] A computer-readable storage medium may include any tangible device capable of storing instructions that are executed by an appropriate device, such that a computer-readable storage medium having instructions stored thereon comprises an article of manufacture, including instructions that can be executed to create means for performing the operations specified in the flowcharts or block diagrams. Examples of computer-readable storage media may include electronic, magnetic, optical, electromagnetic, and semiconductor storage media. More specific examples of computer-readable storage media may include floppy disks, diskettes, hard disks, random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), electrically erasable programmable read-only memories (EEPROMs), static random access memories (SRAMs), compact disc read-only memories (CD-ROMs), digital versatile discs (DVDs), Blu-ray discs, memory sticks, integrated circuit cards, and the like.

[0080] The computer readable instructions may include either assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, JAVA®, C++, etc., and conventional procedural programming languages ​​such as the “C” programming language or similar programming languages.

[0081] The computer-readable instructions may be provided to a general-purpose computer, a special-purpose computer, or another programmable data processing device processor or programmable circuit, either locally or via a local area network (LAN), a wide area network (WAN) such as the Internet, so that the processor or programmable circuit of the programmable data processing device, such as a computer, executes the computer-readable instructions to generate means for performing the operations specified in the flowcharts or block diagrams. Here, the computer may be a personal computer (PC), a tablet computer, a smartphone, a workstation, a server computer, a general-purpose computer, a special-purpose computer, or the like, or may be a computer system in which multiple computers are connected. Such a computer system in which multiple computers are connected is also called a distributed computing system, and is a broad definition of computer. In a distributed computing system, multiple computers collectively execute a program by each executing a portion of the program and passing data between the computers as needed during program execution.

[0082] Examples of processors include computer processors, central processing units (CPUs), processing units, microprocessors, digital signal processors, controllers, microcontrollers, etc. A computer may have one processor or multiple processors. In a multiprocessor system with multiple processors, each processor executes a portion of a program and passes data between processors as needed during program execution, allowing the multiple processors to collectively execute the program. For example, in multitasking, each of the multiple processors may execute a portion of each task in small chunks by switching tasks at each time slice. In this case, which portion of a program each processor executes changes dynamically. Which portion of a program each of the multiple processors executes may also be statically determined by multiprocessor-aware programming.

[0083] Although the present invention has been described above using the embodiments, the technical 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 and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention.

[0084] It should be noted that the execution order of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before," "prior to," etc., and that the processes can be performed in any order unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using "first," "next," etc. for convenience, this does not mean that the processes must be performed in this order.

[0085] 10 System, 30 UE, 100 Management base station, 102 SMO, 104 Demand forecasting model, 110 Memory unit, 112 Management unit, 114 Metrics acquisition unit, 116 Area-related information acquisition unit, 118 Learning execution unit, 120 Demand forecasting unit, 122 Forecast result output unit, 124 Adjustment unit, 200 Distribution base station, 202 MEC_APP_A, 204 MEC_APP_B, 206 MEC_APP_C, 208 MEC_APP_D, 250 Metrics, 300 Radio base station, 310 RAN, 400 System, 450 Area-related information, 1200 Computer, 1210 Host controller, 1212 CPU, 1213 GPU, 1214 RAM, 1216 Graphics controller, 1218 Display device, 1220 input / output controller, 1222 communication interface, 1224 storage device, 1226 DVD drive, 1227 DVD-ROM, 1230 ROM, 1240 input / output chip

Claims

1. A management platform comprising: a management unit that manages multiple distributed platforms that are deployed in various locations and each of which is capable of running multiple MEC applications; a metrics acquisition unit that acquires, for each combination of the multiple distributed platforms and the multiple MEC applications, metrics including the number of connected terminals that indicate the number of user terminals connected to the MEC application; a region-related information acquisition unit that acquires, for each of the multiple distributed platforms, region-related information related to the region in which the distributed platform is deployed; and a demand forecasting unit that predicts future demand for each combination of the multiple distributed platforms and the multiple MEC applications based on the metrics and the region-related information.

2. The management infrastructure described in claim 1, wherein the metrics acquisition unit acquires the metrics further including the communication status between the MEC application and the user terminal for each combination of the plurality of distributed infrastructures and the plurality of MEC applications.

3. The management infrastructure described in claim 2, wherein the metrics acquisition unit acquires the metrics further including at least one of the throughput and communication delay between the MEC application and the user terminal for each combination of the multiple distributed infrastructures and the multiple MEC applications.

4. The management infrastructure described in claim 2, wherein the metrics acquisition unit acquires the metrics further including a HO history indicating the handover history of the user terminal while connected to the MEC application for each combination of the multiple distributed infrastructures and the multiple MEC applications.

5. The management infrastructure described in claim 1, wherein the regional-related information acquisition unit acquires regional-related information for each of the plurality of distributed infrastructures, the regional-related information including weather information and event information for the region in which the distributed infrastructure is located.

6. A management platform as claimed in any one of claims 1 to 5, comprising: a memory unit that stores the metrics and the region-related information for each combination of the multiple distributed platforms and the multiple MEC applications acquired by the metrics acquisition unit and the region-related information acquisition unit as learning data; and a learning execution unit that generates a machine learning model that predicts demand for each combination of the multiple distributed platforms and the multiple MEC applications from the metrics and region-related information for each combination of the multiple distributed platforms and the multiple MEC applications by performing machine learning using the learning data stored in the memory unit, wherein the demand forecasting unit inputs the metrics and the region-related information for each combination of the multiple distributed platforms and the multiple MEC applications for each of multiple past periods into the machine learning model to predict demand for each combination of the multiple distributed platforms and the multiple MEC applications for a future period.

7. The management infrastructure described in claim 6, wherein the machine learning model receives metrics and regional information for each combination of the multiple distributed platforms and the multiple MEC applications as input and outputs the number of connected terminals for each combination of the multiple distributed platforms and the multiple MEC applications, and the demand forecasting unit receives the metrics and regional information for each combination of the multiple distributed platforms and the multiple MEC applications for each of multiple past periods as input to the machine learning model and obtains the number of connected terminals output from the machine learning model as demand for each combination of the multiple distributed platforms and the multiple MEC applications for a future period.

8. A management platform described in any one of claims 1 to 5, further comprising an adjustment unit that adjusts the deployment status of each MEC application on the multiple distributed platforms based on the prediction results by the demand prediction unit.

9. A management method executed by a management platform that manages multiple distributed platforms that are deployed in various locations and each of which is capable of running multiple MEC applications, comprising: a metrics acquisition step that acquires metrics including the number of connected terminals indicating the number of user terminals connected to an MEC application for each combination of the multiple distributed platforms and the multiple MEC applications; a region-related information acquisition step that acquires, for each of the multiple distributed platforms, region-related information related to the region in which the distributed platform is deployed; and a demand forecasting step that predicts future demand for each combination of the multiple distributed platforms and the multiple MEC applications based on the metrics and the region-related information.

10. A program for causing a computer to execute the management method according to claim 9.

11. A system comprising: a management infrastructure according to any one of claims 1 to 5; and a plurality of said distributed infrastructures.

Images