Method for supporting edge load analysis at application data analysis enabler

Abstract

The invention provides functionality for providing edge load analysis at an edge analysis producer, as well as functionality for optimizing edge service performance using the edge load analysis. The edge analysis producer is configured to collect data and perform edge load analysis in consideration of data producers from different domains to allow edge analysis implementation. Further, based on the edge load analysis derived by the edge analysis producer, the analysis consumer is configured to generate a trigger event indicative of the predicted overload and the particular action to be performed.

Description

[0001] Related Applications

[0002] This application claims priority to U.S. Application Serial No. 18 / 189,795, filed on March 24, 2023, entitled "Method for Supporting Edge Load Analytics at Application Data Analytics Enabler", the disclosure of which is incorporated herein by reference in its entirety. U.S. Application Serial No. 18 / 189,795 claims priority to Greek Application Serial No. 20230100229, filed on March 21, 2023, entitled "Method for Supporting Edge Load Analytics at Application Data Analytics Enabler", the disclosure of which is incorporated herein by reference in its entirety. Background Art

[0003] In 3GPP, data analytics services are provided by the Network Data Analytics Function (NWDAF) and are intended to support network data analytics services in the 5G Core (5GC) network. Such analytics can collect data from other network functions (NFs) or analytics functions (AFs), or from Operations, Administration, and Maintenance (OAM), and can be opened to third parties and / or AFs to provide statistics and predictions related to slice load levels, observed service experiences, NF loads, network performance, analysis related to user equipment (UE) such as mobility or communication, user data congestion, Quality of Service (QoS) sustainability, data network (DN) performance, etc.

[0004] In a vertical scenario, additional data analysis on top of the 5G System (5GS) may be required to provide useful outputs to the application-specific layer for end-to-end application services, including application server-related and application session-related statistics and / or predictions. The statistics and / or predictions can be supported and / or enhanced by collecting data from different domains based on consumer needs. Such collection can be from the 5GS (such as NWDAF or Management Domain Analytics Service (MDAS)) via a northbound Application Programming Interface (API), or from the application-specific layer in the Data Network (DN). For example, the data can be related to: collecting HD maps, camera feeds, sensor data, data related to edge and / or cloud resources, data related to the application server state (such as the load of the Edge Application Server (EAS) or the load of the Application Server (AS)), or data from the UE side including UE routing and / or trajectory. Thus, application data collection can be provided by different sources, different sources including for example vertical-specific servers, applications of the UE, EAS, third-party servers, or the Service Enablement Architecture Layer (SEAL). Therefore, it is necessary to identify how to collect this data to allow statistics and / or predictions by the analytics enablement layer.

[0005] The Application Data Analytics Enabler Server (ADAES) is a new enabling service that can be part of the SEAL and discusses new potential application data analytics services related to obtaining statistics and / or predictions to optimize application service operations by notifying the application-specific layer and potentially the 5GS about expected and / or predicted changes in application service parameters (such as QoS parameters considering both connected and non-connected deployments).

[0006] Edge deployment is extremely important for applications that require performance levels not met by existing cloud deployments. Edge data analysis can involve statistics and / or predictions of computing resources and the expected and / or predicted load of the platform hosting the edge applications. It may be necessary to open these edge data analytics as a service to the EAS. Edge applications can be edge-native applications or edge-enhanced applications at the centralized cloud. In particular, for edge-native applications that require lightweight design and high portability, using edge analytics at the edge platform can help improve application service operations.

[0007] Supporting edge analytics at the enablement layer that can be related to edge performance, faults, and service availability would be useful for edge applications to allow dynamically deciding to scale down, scale out, migrate from the edge to the cloud under heavy load, or migrate from the cloud to the edge to improve the quality of experience for end-users.

[0008] Therefore, it is expected to provide edge analysis enabling related to edge performance, faults, loads, service availability, etc. by collecting data from data producers in different domains and performing edge load analysis based on the collected data. In a further step, it is expected to utilize these edge load analyses to optimize edge service performance.

[0009] These problems have not been solved in 3GPP so far. Prior art solutions provide mechanisms for application layer analysis, which can be related to application servers or sessions regarding performance. However, such prior art solutions do not discuss the analyzed subscribed edge load types, nor provide analysis for data network name (DNN), for data network access identifier (DNAI), for edge data network (EDN), or for edge enabler server (EES), etc.

[0010] Similarly, the prior art does not solve the problem of collecting certain types of data from some sources such as N6 endpoints, multi-access edge computing (MEC) platform services such as radio network information service (RNIS), operation and maintenance (OAM) functions for computing loads, etc. Since the prior art does not publicly provide analysis for each DNN, for each DNAI, for each EDN, or for each EES, etc., the prior art also does not mention providing output data for each DNN, for each DNAI, for each EDN, or for each EES, etc. In addition, the problem of using the analysis output to recommend certain actions for edge nodes and trigger overload has not been discussed so far. Summary of the Invention

[0011] In one aspect, the present invention provides a computer-implemented method for edge load analysis at an ADAES. In another aspect, the present invention is directed to an apparatus (such as an ADAES) configured to execute the computer-implemented method.

[0012] Edge load analysis provides an understanding of the operation and performance of an EDN. In particular, edge load analysis can provide statistics or predictions of parameters related to EAS or EES loads for one or more EASs or EESs respectively, and edge platform load parameters, which can include aggregated loads for each EDN or for each DNAI due to edge support services and the load level of edge computing resources, for example.

[0013] Accordingly, a first aspect of the present invention relates to an apparatus including one or more processors configured to execute computer-readable instructions for supporting edge load analysis. The instructions cause the one or more processors to receive a subscription request for edge load analysis of an edge node from an analysis consumer. Thereby, the subscription request indicates at least an analysis event identifier. Further, the apparatus is configured to determine a mapping of the analysis event identifier to at least one of: a list of data collection event identifiers, and a list of data producer identifiers. Subsequently, computer-readable instructions executed by the one or more processors of the apparatus cause the one or more processors to send a data collection subscription request to data producers identified by the list of data producer identifiers, where the data collection subscription request includes at least one of: the analysis event identifier, and a corresponding data collection event identifier. In a next step, the apparatus is configured to receive data from the data producers, and the received data thereby corresponds to the analysis event identifier or the corresponding data collection event identifier. Then, computer-readable instructions executed by the one or more processors cause the one or more processors to derive an edge load analysis for the edge node from the received data corresponding to the subscription request. The edge load analysis may indicate at least one of: statistics and predictions of the load for the edge node. Finally, the apparatus according to this first aspect of the present invention is configured to send the derived edge load analysis to the analysis consumer.

[0014] Such derived edge load analysis can improve edge support services by allowing proactive edge service operations to change to handle possible edge overload situations. For example, these analyses can trigger the migration of EAS to a different EDN or a central data network, or alternatively trigger proactive EAS reselection for a target UE or a group of UEs.

[0015] Accordingly, a second embodiment of the present invention relates to an apparatus including one or more processors configured to execute computer-readable instructions for optimizing edge service performance using edge load analysis. The instructions cause the one or more processors to send a subscription request for edge load analysis to an edge analysis producer and subsequently receive the derived edge load analysis from the edge analysis producer. Subsequently, the computer-readable instructions of the apparatus according to the second embodiment of the present invention cause the one or more processors to generate a trigger event indicating predicted overload and actions based on the derived edge load analysis.

[0016] As described above, in a further aspect, the present invention also relates to a computer-implemented method for providing edge load analysis. In a first step of the method, a subscription request for load analysis of an edge node is received by the ADAES. Thereby, the subscription request indicates an analysis event identifier. In addition, the method includes a step of determining, by the ADAES, a mapping of the analysis event identifier to at least one of the following: a list of data collection event identifiers, and a list of data producer identifiers. Subsequently, the method includes a step of sending, by the ADAES, a data collection subscription request to the data producers identified by the list of data producer identifiers. The data collection subscription request includes at least one of the following: the analysis event identifier, and the corresponding data collection event identifier. In addition, the method according to a further aspect of the present invention includes receiving data at the ADAES from the data producers. Thereby, the received data corresponds to at least one of the following: the analysis event identifier, and the corresponding data collection event identifier. Then, a further method step includes deriving an edge load analysis for the edge node from the received data corresponding to the subscription request, wherein the edge load analysis indicates at least one of the following: statistics and prediction of the load for the edge node. Subsequently, the method includes sending the derived edge load analysis to an analysis consumer. Finally, according to another aspect of the method based on the present invention, a trigger event indicating a predicted overload and an action can be generated.

[0017] The present invention content is provided to introduce a selection of concepts in a simplified form, which will be further described in the following detailed description. The present invention content is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter. Brief Description of the Drawings

[0018] Figure 1 is an exemplary flowchart of a process for supporting edge load analysis according to a first embodiment of the present invention.

[0019] Figure 2 is an exemplary flowchart of a process for triggering an action using edge load analysis according to a second embodiment of the present invention.

[0020] Figure 3 shows an example of an apparatus configured to execute the processes according to both the first and second embodiments of the present invention. Detailed Description of the Embodiments

[0021] Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to like elements unless otherwise specified.

[0022] Figure 1Process 100 for supporting edge load analysis at edge analytics producer 102 according to a first embodiment of the present invention is shown. In particular, Figure 1 ADAES is shown as an example of edge analytics producer 102.

[0023] In Figure 1 the first embodiment of the present invention shown, data for performing edge load analysis can be collected from EDN 104 and / or from one or more of a plurality of EAS 104A and EES 104B included in EDN 104. Alternatively or additionally, data for performing edge load analysis can be collected from analytics data repository (ADR) 106. ADR 106 is an entity that stores historical data and / or analytics (i.e., data and / or analytics that have been acquired by ADAES (such as ADAES 102) related to a past time period). After acquiring such data and / or analytics, ADAES (such as ADAES 102) can store the historical data and / or analytics in ADR 106.

[0024] ADR 106 can be an application layer - analytics and data repository function (A - ADRF) defined in TS23.436, an analytics and data repository function (ADRF) defined in TS23.288, a common application programming interface (API) framework (CAPIF) core function defined in TS23.222, or any other repository that stores offline data in an edge or cloud platform.

[0025] ADAES 102 can easily meet the pre - conditions that a plurality of edge services have found corresponding application programming interfaces to access EDN 104, and that OAM and NWDAF of 5GS 108 have been subscribed to for receiving management and data network (DN) performance analysis respectively.

[0026] Figure 1 An analytics consumer 110 of the ADAES 102 analysis service is also shown. The analytics consumer 110 can send a subscription request 120 for load analysis of an edge node to ADAES 102. Hereby, the edge node can be at least one of EDN 104, EAS 104A, EES 104B, etc.

[0027] The subscription request 120 can include an information element indicating an analysis event identifier and can include additional information elements. These additional information elements include at least one of the following: analytics consumer identifier, analytics filter information, analysis type, destination EAS identifier, destination EES identifier, DNN, DNAI, region of interest, and time validity.

[0028] The analysis event identifier can be an identifier of an analysis event. An example of such an event can be edge performance analysis. The analysis consumer identifier can be an identifier of an analysis consumer 110 (such as a vertical application layer (VAL) server, EAS, etc.). The analysis filter information includes filter information for the analysis event, and the analysis type can describe whether the analysis event can involve prediction or statistics. The destination EAS identifier can identify the destination EAS 104A associated with the subscription request 120, and the destination EES identifier can identify the destination EES 104B associated with the subscription request 120. The DNN element can describe the DNN information associated with the subscription request 120, and the DNAI element can indicate the DNAI information associated with the subscription request 120. The preferred confidence level can indicate the accuracy level of the analysis service that can be achieved in the case of prediction. The area of interest can describe the geographical area or service area associated with the subscription request 120. The time validity information element can describe the time validity of the subscription request 120.

[0029] In addition, as Figure 1 shown, the ADAES 102 can send a subscription response 125 to the analysis consumer 110 as an acknowledgement.

[0030] The ADAES 102 can also determine a mapping 130 of the analysis event identifier included in the subscription request 120 to at least one of the following: a list of data collection event identifiers, and a list of data producer identifiers. Alternatively, such a mapping 130 can have been pre-configured by the OAM. The data producer can be at least one of the EAS 104A, the EES 104B onboarded to the EDN 104, the ADR 106, components of the 5GS 108 (such as the OAM and the 5GC), and Figure 1 at least one of the MEC platform services not shown in the figure.

[0031] As a next step, the ADAES 102 can send a data collection subscription request 140 to the data producers identified by the list of data producer identifiers. The data collection subscription request 140 can include at least one of the following: the analysis event identifier, and the corresponding data collection event identifier.

[0032] In addition, the data collection subscription request 140 can include at least one of the following: the ADAES identifier, the data collection requirements, the list of data producer identifiers, the destination EAS identifier, the destination EES identifier, the DNN, the DNAI, the area of interest, and the time validity.

[0033] Accordingly, the ADAES identifier can be the identifier of ADAES102, the destination EAS identifier can identify the destination EAS associated with the subscription request 120, the destination EES identifier can identify the destination EES associated with the subscription request 120, the DNN element can indicate the DNN information associated with the subscription request 120, and the DNAI element can indicate the DNAI information associated with the subscription request 120. Additionally, the area of interest can indicate at least one geographical area and service area associated with the subscription request 120, while the time validity information element can describe the time validity indication of the subscription request 120.

[0034] When the data collection subscription request 140 is executed via the Application Layer Data Collection and Coordination Function (A-DCCF), the above list of data producer identifiers may need to be included in the data collection subscription request 140.

[0035] Furthermore, the data collection requirements included in the data collection subscription request 140 can include at least one of the following items: data format, reporting frequency, level of data abstraction, and level of data precision. For example, the reporting frequency can indicate that the collected data can be provided once or periodically based on a threshold of, for example, a load greater than a certain percentage.

[0036] Subsequently, in the process 100 for supporting edge load analysis according to the first embodiment of the present invention, the data producer can send a data collection subscription response 145 to ADAES102. The data collection subscription response 145 received at ADAES102 can be an affirmative or negative confirmation of the data collection subscription request 140.

[0037] In addition, according to Figure 1 the first embodiment of the present invention as shown, ADAES102 can receive data from the data producer. The received data can correspond to an analysis event identifier or a corresponding data collection event identifier and can be received at ADAES102 as a data notification message. The data notification message can include one or more of the following items: the data notification message 150A received by ADAES102 from ADR 106, and the data notification message 150B received from a data producer such as at least one of EDN 104, EAS104A, and EES104B.

[0038] The data received by ADAES102 can include offline data from ADR 106. As previously mentioned, ADR 106 can be an A-ADRF, ADRF, CAPIF core function, or any other repository that stores offline data in an edge or cloud platform.

[0039] The offline data received from ADR 106 can include statistics and can include at least one of the following: historical and non-real-time measurements, and data and analytics generated offline based on historical measurements. In some implementations, such offline data can be stored in a database or repository (such as ADR 106).

[0040] The offline data received from ADR 106 at ADAES102 can include at least one of the following: load statistics for multiple EAS104A connections or EES 104B connections for a given region or time window, statistics regarding average edge computing resource usage, resource ratios of overall resource availability based on EDN104, EDN 104 overload indication, high load indication events, and probabilities of unavailability of EAS104A and EES104B due to high load.

[0041] The data notification message 150A received by ADAES102 from ADR 106 can include at least one of the following: data collection event identifier, data producer identifier, destination EAS identifier, destination EES identifier, DNN, DNAI, analytics identifier, data type, and data output.

[0042] All other mentioned information elements of the data notification message 150A, except for the data type and data output information elements, have been described in detail above in the discussion of the subscription request for edge analytics sent by the analytics consumer 110 and the data collection subscription request sent by ADAES102. Therefore, the description of these information elements will not be repeated here.

[0043] The data type of the data received from ADR 106 and included in the data notification message 150A can include the type of reported data samples, which can be at least one of the following: network data, application data, edge data, and different granularities and abstractions of data.

[0044] The data output of the data received from ADR 106 and included in the data notification message 150A can include subscription-based, offline reported data or historical data regarding the requested parameters. The data output can be data for each EDN 104, for each DNAI, or for each EAS104A or for each EES104B, and can include load statistics and edge computing resource utilization statistics for at least one of a given time and region of interest.

[0045] Optionally or additionally to the offline data, the data received at the ADAES 102 may also include real-time data collected from data producers. The data producers may initiate the collection 160 of real-time measurements of at least one of the load and resource utilization for the EDN 104, EES 104A, and EAS 104B at the requested time, as well as the analysis from the 5GS 108.

[0046] More specifically, the real-time data collected at the data producers may include at least one of the following items: load statistics for multiple EAS 104A connections or EES 104B connections for a given area or time window, statistics on the average edge computing resource usage, resource ratios of the overall resource availability based on the EDN 104, EDN 104 overload indication, high-load indication events, and the probability of unavailability of the EAS 104A and EES 104B due to high load.

[0047] The real-time data collected from the data producers is received at the ADAES 102 as a data notification message 150B. The content of the data notification message 150B may include the same information elements as those already described for the data notification message 150A received from the ADR 106 at the ADAES 102. Therefore, the content of the data notification message 150B is considered to be similar to the content of the data notification message 150A. Therefore, its detailed description will be omitted here.

[0048] In addition, it should be noted that the data notification message 150A and the data notification message 150B are not necessarily meant to be sent in sequence. Instead, the two data notification messages 150A and 150B may also be sent in parallel or in a different order.

[0049] The data producers that may provide the real-time data collected may include at least one of the EAS 104A, EES 104B, N6 endpoints, OAM functions, Service Enabler Architecture Layer Data Delivery Server (SEALDD), 5GC and NWDAF, MDAS, and MEC such as RNIS.

[0050] Accordingly, EAS104A can provide at least one of the following: the computing resource load for each EAS 104A and the number of connections of EAS104A, while EAS104B can provide at least one of the following: the computing resource load for each EAS104B and the number of connections of EES104B. In addition, the N6 endpoint can provide the N6 load, and the OAM function can provide at least one of the computing resource load for each EAS104A and the number of connections of EAS 104A, and at least one of the computing resource load for each EES104B and the number of connections of EES104B. In addition, SEALDD can provide N6 load measurement and SEALDD computing resource load, and at least one 5GC and NWDAF can provide data network performance analysis. In addition, at least one of OAM and MDAS can provide user plane function (UPF) load analysis for each DNAI, and MEC platform services such as RNIS can provide, for all cells within EDN104, at least one of the average radio conditions for each cell, and load and resource utilization.

[0051] Returning to the overall method according to the first embodiment of the present invention as Figure 1 shown, the data received at ADAES102 corresponding to the subscription request 120 can be used to derive 170 the edge load analysis for the edge node. The edge load analysis derived in this way can indicate at least one of the following: the statistics and prediction of the load for the edge node. As described above, the edge node can be at least one of EDN 104, EAS104A, EES104B, etc.

[0052] More specifically, based on the analysis identifier and the request type, ADEAS102 can derive an edge analysis of at least one of the following: EDN 104 load, DNAI load, and the load for each EAS104A and the load for each EAS104B. ADEAS 102 can derive these analyses based on at least one of the following: the performance analysis received for each data network, and the load analysis for each DNAI or UPF. In addition, when deriving these analyses, ADEAS102 can also consider measurements of at least one of the following: computing and radio access network (RAN) resource load, and the number of connections of EAS104A and EES104B active at EDN 104.

[0053] Finally, according to the present invention Figure 1In the process 100 shown, the ADAES 102 can send the derived edge load analysis 170 to the analytics consumer 110. As described above, these sent edge load analyses 170 can indicate predictions considering the EDN 104 load input from both the 5GS 108 and from the edge platform services.

[0054] Importantly, due to the expected high load on the edge resources, these predictions can also be in the form of recommendations for triggering the repositioning of the EAS 104A to a different platform.

[0055] The ADAES 102 can send the derived edge load analysis 170 to the analytics consumer 110 in the form of an edge analysis notification 180. This edge analysis notification 180 can include at least one of the following items: an analysis identifier, an analysis type, an analysis output, and a confidence level. While the analysis identifier can be an identifier of the analysis event, the analysis type can indicate the type of analysis based on the analysis event. Such an analysis type can include at least one of the following items: offline or online analysis, machine learning enabled analysis, statistical and predictive analysis. The analysis output can indicate at least one of the following items: prediction parameters, and statistical parameters, which can be related to and based on the statistics or predictions of the following items: the edge performance or load of the edge platform and at least one of the EAS 104A and EES 104B for at least one of a given region and time. Finally, the achieved confidence level can be provided in the case of predictive analysis.

[0056] Figure 2 A second embodiment of the present invention is shown. The second embodiment of the present invention is directed to a process 200 for optimizing edge service performance using edge load analysis.

[0057] A method 100 for an edge load analysis service is described according to a first embodiment of the present invention to provide an understanding of the operation and performance of the EDN 104 for at least one or more of the EAS 104A and EES 104B, and in particular an understanding of at least one or more statistics and predictions of parameters related to the load of the EAS 104A or EES 104B.

[0058] Such analysis can improve the edge support service by allowing the edge service operation to proactively handle possible edge overload situations.

[0059] Some edge support services can benefit from using ADAES102 analysis related to the service load of at least one of EDN 104, or EAS104A and EES104B. In standard TS23.558, one of the conditions for service continuity is the overload situation of at least one of EAS104A and EDN. Therefore, edge load analysis including at least one of prediction and statistics can help proactively trigger actions to prevent service loss due to expected overload.

[0060] As Figure 2 shown, a prerequisite for the second embodiment of the present invention is that an edge analysis producer 102 such as ADAES is available at EDN 104, and the edge analysis producer 102 is accessible to EES104B.

[0061] In the first step of method 200 according to the second embodiment of the present invention, for example Figure 2 the device of EES104B as shown can send a subscription request 210 for edge load analysis to the edge analysis producer 102.

[0062] Thus, as described above, the edge analysis producer 102 can be ADEAS, and the device 104B can include at least one of the following: EES, EAS, and analysis consumer.

[0063] The subscription request 210 can also correspond to the subscription request 120 of the first embodiment of the present invention.

[0064] In the next step of the method according to the second embodiment of the present invention, for example, the device of EES104B can receive the derived edge load analysis 220 from the edge analysis producer 102. The derived edge load analysis can be derived by the edge analysis producer 102 according to the first embodiment of the present invention.

[0065] Optionally, EES104B can also provide the received edge load analysis to EAS104A.

[0066] Finally, based on the edge load analysis derived in step 220, the device can generate a trigger event 240, which indicates a prediction or expected overload and actions for at least one of EDN 104, EAS104A, and EES 104B.

[0067] Such possible actions can include application context relocation (ACR), which includes at least one of the following: migration of edge nodes such as EAS104A and EES 104B to different EDN 104, and proactive EAS reselection for a target user equipment (UE) or a group of UEs.

[0068] Depending on the service continuity scenario according to standard TS23.558, the last step of method 200 according to the second embodiment of the present invention is performed at EAS104A or at EES104B. In this context, it should be noted that the entity in EAS104A and EES104B that indicates an expected overload of the EAS or EES based on the received load analysis may be responsible for triggering the corresponding actions.

[0069] Figure 3 Illustrated is an ADAES102 and a device 300 for analyzing a consumer 110.

[0070] The device 300 may include a memory 320, one or more processors 310A, 310B, etc., and a transceiver 340. The memory 320 may be a volatile memory (e.g., DRAM or SRAM) or a non-volatile memory (e.g., SDD or HDD memory). The memory 320 stores computer-readable instructions 330 that one or more processors 310A, 310B, etc. are configured to execute. When these computer-readable instructions 330 are executed, one or more processors 310A, 310B, etc. may respectively implement the methods of the first and second embodiments as described above with reference to Figure 1 and Figure 2 the first and second embodiments described.

[0071] The embodiments presented herein should not be construed as being limited to the specific combinations of features described as being performed by hardware and / or software entities. In particular, other possible embodiments may include any combination of features from the described embodiments. Additionally, features described in the context of a particular embodiment may also be included in other embodiments without being presented so explicitly. An embodiment may include more or fewer features than described. Moreover, software and hardware entities may perform more or fewer features than described in a particular embodiment. A software or hardware entity may also perform features described in the context of other software or hardware entities. Additionally, steps described in a particular order in the context of a method may be performed in any other reasonable order. It should be understood that this specification encompasses all embodiments resulting from these alternative combinations of features and entities.

[0072] Although the present invention has been described with respect to physical embodiments constructed in accordance with the present invention, it will be apparent to those skilled in the art that various modifications, variations, and improvements can be made to the present invention in accordance with the above teachings and within the scope of the appended claims without departing from the spirit and intended scope of the present disclosure. In addition, those fields that are familiar to those of ordinary skill in the art are not described herein so as not to obscure the invention described herein. Accordingly, it should be understood that the present invention is not limited by the specific illustrative embodiments, but only by the scope of the appended claims. Additional aspects of the techniques, features, and / or methods discussed herein relate to one or more of the following items:

[0073] An apparatus for wireless communication, comprising: at least one memory and at least one processor, the at least one processor being coupled to the at least one memory and configured to cause the apparatus: receive, from an analytics consumer, a subscription request for edge load analysis of an edge node, the subscription request indicating an analysis event identifier; determine a mapping of the analysis event identifier to at least one of: a list of data collection event identifiers, or a list of data producer identifiers; send a data collection subscription request to the data producers identified by the list of data producer identifiers, the data collection subscription request including at least one of: the analysis event identifier, or the corresponding data collection event identifier; receive data from the data producers, the received data corresponding to the analysis event identifier or the corresponding data collection event identifier; derive, from the received data corresponding to the subscription request, an edge load analysis for the edge node, the edge load analysis indicating at least one of: a statistic or prediction of the load for the edge node; and send the derived edge load analysis to the analytics consumer.

[0074] Alternatively, or in addition to the above device, any one or combination of the following: the edge node is an Edge Data Network (EDN), an Edge Enabler Server (EES), or an Edge Application Server (EAS). The subscription request also includes at least one of the following: an analytics consumer identifier, filter information for an analytics event, the type of analytics for the analytics event, a destination EAS identifier identifying the destination EAS associated with the subscription request, a destination EES identifier identifying the destination EES associated with the subscription request, data network name (DNN) information associated with the subscription request, data network access identifier (DNAI) information associated with the subscription request, a preferred confidence level for prediction, a geographic region associated with the subscription request, a service region associated with the subscription request, or a time validity indication for the subscription request. The type of analytics for the analytics event indicates whether the analytics event involves prediction or statistics. At least one processor is configured to cause the device to send a subscription response to the analytics consumer as an acknowledgement. The mapping is preconfigured by an Operations, Administration, and Maintenance (OAM) function. The data collection subscription request also includes at least one of the following: a device server identifier, data collection requirements, a list of data producer identifiers, a destination EAS identifier identifying the destination EAS associated with the subscription request, a destination EES identifier identifying the destination EES associated with the subscription request, DNN information associated with the subscription request, DNAI information associated with the subscription request, a preferred confidence level for prediction, a geographic region associated with the subscription request, a service region associated with the subscription request, or a time validity indication for the subscription request.

[0075] Alternatively or additionally, the data collection requirements include at least one of the following: data format, reporting frequency, level of abstraction of the data, or level of precision of the data. At least one processor is configured to cause the device to receive a data collection subscription response from a data producer, the data collection subscription response being an affirmative or negative acknowledgement. At least one processor is configured to cause the device to receive offline data from an analytics data repository. The received data includes at least one of the following: load statistics for multiple EAS connections or EES connections for a given area or time window, statistics regarding average edge computing resource usage, resource ratios of overall resource availability based on EDN, EDN overload indication, high load indication events, or probability of EAS and EES unavailability due to high load. The received data pertains to a given time or area of interest. At least one processor is configured to cause the device to receive real-time collected data from a data producer. The real-time collected data includes at least one of the following: load statistics for multiple EAS connections or EES connections for a given area or time window, statistics regarding average edge computing resource usage, resource ratios of overall resource availability based on EDN, EDN overload indication, high load indication events, or probability of EAS and EES unavailability due to high load. The data producer includes at least one of the following: an EAS that provides at least one of the computational resource load for each EAS or the number of connections of the EAS; an EES that provides at least one of the computational resource load for each EES or the number of connections of the EES; an N6 endpoint that provides an N6 load; an OAM function that provides at least one of the computational resource load for each EAS or the number of connections of the EAS; an OAM function that provides at least one of the computational resource load for each EES or the number of connections of the EES; a service enabling architecture layer data transfer server (SEALDD) that provides N6 load measurement and SEALDD computational resource load; at least one of a 5G core (5GC) or a network data analytics function (NWDAF) that provides data network performance analysis; a management domain analysis service (MDAS) that provides load analysis for each DNAI; or a multi-access edge computing (MEC) platform service that includes a radio network information service (RNIS), the radio network information service (RNIS) providing average radio conditions and load for each cell for all cells within the EDN.

[0076] An apparatus for wireless communication, comprising: at least one memory and at least one processor, the at least one processor being coupled to the at least one memory and configured to cause the apparatus to: send a subscription request for edge load analysis to an edge analysis producer; receive the resulting edge load analysis from the edge analysis producer; and generate a trigger event indicating predicted overload and an action, at least in part based on the resulting edge load analysis. Alternatively, or in addition to the above apparatus, the action comprises at least one of the following: migration of an edge node to a different EDN, or proactive EAS reselection for a target user equipment (UE) or a group of UEs. The apparatus comprises at least one of the following: an EES, an EAS, or an analysis consumer.

[0077] A method for wireless communication, comprising: receiving, by an ADAES, from an analysis consumer, a subscription request for edge load analysis of an edge node, the subscription request indicating an analysis event identifier; determining, by the ADAES, a mapping of the analysis event identifier to at least one of the following: a list of data collection event identifiers, or a list of data producer identifiers; sending, by the ADAES, a data collection subscription request to data producers identified by the list of data producer identifiers, the data collection subscription request comprising at least one of the following: the analysis event identifier, or a corresponding data collection event identifier; receiving, by the ADAES, data from the data producers, the received data corresponding to at least one of the following: the analysis event identifier, or a corresponding data collection event identifier; deriving, from the received data corresponding to the subscription request, an edge load analysis for the edge node, the edge load analysis indicating at least one of the following: a statistic or a prediction of the load for the edge node; and sending the resulting edge load analysis to the analysis consumer. Alternatively, or in addition to the above method, the method further comprises: generating a trigger event indicating predicted overload and an action, wherein the action comprises at least one of the following: migration of an edge node to a different EDN, or proactive EAS reselection for a target UE or a group of UEs.

Claims

1. A device for wireless communication, comprising: at least one memory; and at least one processor, the at least one processor being coupled to the at least one memory and configured to cause the device to: receive, from an analytics consumer, a subscription request for edge load analysis of an edge node, the subscription request indicating an analytics event identifier; determine a mapping of the analytics event identifier to at least one of: a list of data collection event identifiers, or a list of data producer identifiers; send a data collection subscription request to data producers identified by the list of data producer identifiers, the data collection subscription request including at least one of: the analytics event identifier, or a corresponding data collection event identifier; receive data from the data producers, the received data corresponding to the analytics event identifier or the corresponding data collection event identifier; derive, from the received data corresponding to the subscription request, the edge load analysis for the edge node, the edge load analysis indicating at least one of: statistics or prediction of the load for the edge node; and send the derived edge load analysis to the analytics consumer.

2. The device according to claim 1, wherein the edge node is an edge data network (EDN), an edge enabler server (EES), or an edge application server (EAS).

3. The device according to claim 1, wherein the subscription request further includes at least one of: an analytics consumer identifier, filter information for the analytics event, an analytics type of the analytics event, a destination edge application server (EAS) identifier identifying a destination EAS associated with the subscription request, a destination EES identifier identifying a destination EES associated with the subscription request, data network name (DNN) information associated with the subscription request, data network access identifier (DNAI) information associated with the subscription request, a preferred confidence level for the prediction, a geographical region associated with the subscription request, a service area associated with the subscription request, or a time validity indication of the subscription request.

4. The device according to claim 3, wherein the analytics type of the analytics event indicates whether the analytics event relates to the prediction or the statistics.

5. The device according to claim 1, wherein the at least one processor is configured to cause the device to send a subscription response to the analytics consumer as an acknowledgement.

6. The device according to claim 1, wherein the mapping is preconfigured by an operation, administration, and maintenance (OAM) function.

7. The apparatus according to claim 1, wherein the data collection subscription request further includes at least one of the following: a device server identifier, data collection requirements, a list of the data producer identifiers, a destination edge application server (EAS) identifier identifying a destination EAS associated with the subscription request, a destination EES identifier identifying a destination edge enabler server (EES) associated with the subscription request, data network name (DNN) information associated with the subscription request, data network access identifier (DNAI) information associated with the subscription request, a preferred confidence level for the prediction, a geographical region associated with the subscription request, a service region associated with the subscription request, or a time validity indication of the subscription request.

8. The apparatus according to claim 7, wherein the data collection requirements include at least one of the following: data format, reporting frequency, level of abstraction of the data, or level of precision of the data.

9. The apparatus according to claim 1, wherein the at least one processor is configured to cause the apparatus to: receive a data collection subscription response from the data producer, the data collection subscription response being an affirmative or negative confirmation.

10. The apparatus according to claim 1, wherein the at least one processor is configured to cause the apparatus to receive offline data from an analysis data repository.

11. The apparatus according to claim 10, wherein the received data includes at least one of the following: load statistics of multiple edge application server (EAS) connections or edge enabler server (EES) connections for a given region or time window, statistics on average edge computing resource usage, resource ratios of overall resource availability based on an edge data network (EDN), an EDN overload indication, a high load indication event, or a probability of unavailability of EASs and EESs due to high load.

12. The apparatus according to claim 10, wherein the received data relates to a given time or region of interest.

13. The apparatus according to claim 1, wherein the at least one processor is configured to cause the apparatus to receive real-time collected data from the data producer.

14. The apparatus according to claim 13, wherein the real-time collected data includes at least one of the following: load statistics of multiple edge application server (EAS) connections or edge enabler server (EES) connections for a given region or time window, statistics on average edge computing resource usage, resource ratios of overall resource availability based on an edge data network (EDN), an EDN overload indication, a high load indication event, or a probability of unavailability of EASs and EESs due to high load.

15. The apparatus according to claim 1, wherein the data producer includes at least one of the following: Edge Application Server (EAS), where the Edge Application Server (EAS) provides at least one of the following: the computing resource load for each EAS, or the number of connections of the EAS; Edge Enabler Server (EES), where the Edge Enabler Server (EES) provides at least one of the following: the computing resource load for each EES, or the number of connections of the EES; N6 endpoint, where the N6 endpoint provides an N6 load; Operation Administration and Maintenance (OAM) function, where the Operation Administration and Maintenance (OAM) function provides at least one of the following: the computing resource load for each EAS, or the number of connections of the EAS; OAM function, where the OAM function provides at least one of the following: the computing resource load for each EES, or the number of connections of the EES; Service Enabler Architecture Layer Data Delivery Server (SEALDD), where the Service Enabler Architecture Layer Data Delivery Server (SEALDD) provides N6 load measurement and SEALDD computing resource load; At least one of 5G Core (5GC) or Network Data Analytics Function (NWDAF) that provides data network performance analysis; Management Domain Analysis Service (MDAS), where the Management Domain Analysis Service (MDAS) provides load analysis for each Data Network Access Identifier (DNAI); or Multi-Access Edge Computing (MEC) platform service, where the Multi-Access Edge Computing (MEC) platform service includes Radio Network Information Service (RNIS), and the Radio Network Information Service (RNIS) provides the average radio condition and load for each cell for all cells within an Edge Data Network (EDN).

16. An apparatus for wireless communication, comprising: At least one memory; And At least one processor, the at least one processor being coupled to the at least one memory and configured to cause the apparatus to: Send a subscription request for edge load analysis to an edge analysis producer; Receive the resulting edge load analysis from the edge analysis producer; and Generate a trigger event indicating predicted overload and an action, at least in part based on the resulting edge load analysis.

17. The apparatus according to claim 16, wherein the action includes at least one of the following: migration of an edge node to a different Edge Data Network (EDN), or proactive re-selection of an Edge Application Server (EAS) for a target User Equipment (UE) or a group of UEs.

18. The apparatus according to claim 16, wherein the apparatus includes at least one of the following: an Edge Enabler Server (EES), an Edge Application Server (EAS), or an analysis consumer.

19. A method for wireless communication, the method comprising: Receiving, by an Application Data Analytics Enablement Server (ADAES), from an analysis consumer, a subscription request for edge load analysis of an edge node, the subscription request indicating an analysis event identifier; The ADAES determines a mapping of the analysis event identifier to at least one of the following: a list of data collection event identifiers, or a list of data producer identifiers; The ADAES sends a data collection subscription request to the data producers identified by the list of data producer identifiers, the data collection subscription request including at least one of the following: the analysis event identifier, or the corresponding data collection event identifier; The ADAES receives data from the data producers, the received data corresponding to at least one of the following: the analysis event identifier, or the corresponding data collection event identifier; An edge load analysis for the edge node is derived from the received data corresponding to the subscription request, the edge load analysis indicating at least one of the following: a statistic or prediction of the load for the edge node; The derived edge load analysis is sent to the analysis consumer.

20. The method according to claim 19, further comprising: Generating a trigger event indicating a predicted overload and an action, wherein the action includes at least one of the following: migration of the edge node to a different edge data network (EDN), or proactive edge application server (EAS) reselection for a target user equipment (UE) or a group of UEs.

Images