Search

KR-20260062762-A - Method, apparatus and system for supporting serverless computing in mobile network

KR20260062762AKR 20260062762 AKR20260062762 AKR 20260062762AKR-20260062762-A

Abstract

The present invention discloses a method, apparatus, and system for supporting serverless computing in a mobile network. According to the present invention, a serverless computing support apparatus in a mobile network is provided, comprising: a processor; and a memory connected to the processor, wherein the memory stores program instructions executed by the processor to receive computing and network metrics collected in advance from each MEC server from a Serverless Computing Aware Function (SCAF) included in a mobile core network function and mobile underlay network metrics collected from a Network Data Analytics Function (NWDAF), and to determine serverless service instance placement information including the number of pre-warmed serverless service instances and one or more MEC servers to which the service instances are to be placed using a serverless orchestration algorithm by referencing the metrics.

Inventors

  • 김영한

Assignees

  • 숭실대학교산학협력단

Dates

Publication Date
20260507
Application Date
20241126
Priority Date
20241029

Claims (12)

  1. As a method of supporting serverless computing in mobile networks, A Serverless Computing Aware Function (SCAF) included in the mobile core network function transmits computing and network metrics collected in advance from each MEC server to the MEC orchestrator; The above SCAF transmits mobile underlay network metrics collected from the Network Data Analytics Function (NWDAF) to the MEC orchestrator; The above MEC orchestrator determines serverless service instance placement information, including the number of pre-warmed serverless service instances and one or more MEC servers on which the service instances will be placed, by using a serverless orchestration algorithm with reference to the metric; A step of one or more MEC servers preparing a serverless service instance according to the above-determined deployment information; and A method for supporting serverless computing in a mobile network, comprising the step of an Edge Application Server Discovery Function (EASDF) included in the functions of the mobile core network exchanging a PDU session and the deployment information with the SCAF in response to a serverless service request from a User Equipment (UE) to establish a data plane routing path for the UE.
  2. In paragraph 1, A method for supporting serverless computing in a mobile network, wherein the above-described serverless orchestration algorithm uses request traffic volume, UE location, and MEC site resource state prediction or mathematical analysis results as input data to determine pre-warmed service instance placement information for preparing future request traffic.
  3. In paragraph 1, The above MEC orchestrator transmits the deployment information to one or more MEC servers through the MEC platform manager, and A method for supporting serverless computing in a mobile network in which a FaaS framework running on one or more MEC servers performs serverless service instance placement operations to prepare a required number of pre-warmed service instances.
  4. In paragraph 1, The Session Management Function (SMF) included in the above mobile core network function receives a PDU session setup request for the serverless service of the UE, and then provides the location of the UE and the service domain name to the SCAF. The above SCAF is a method for supporting serverless computing in a mobile network that determines a destination by considering the computing resource consumption and request provision status of the current serverless service, which are updated by the computing metric agent of each MEC server.
  5. In paragraph 1, The above SCAF is a method for supporting serverless computing in a mobile network that determines one of a plurality of MEC servers, prepared with a pre-warmed serverless service instance capable of providing more available requests, as the target MEC server.
  6. In paragraph 1, A method for supporting serverless computing in a mobile network, wherein when there are no multiple MEC servers prepared with pre-warmed serverless service instances capable of providing more available requests, the above SCAF calculates the expected service latency based on network bandwidth and computing resource consumption to determine the target MEC server to provide serverless services to the UE.
  7. As a serverless computing support system in mobile networks, A MEC orchestrator that determines serverless service instance placement information including the number of pre-warmed serverless service instances and one or more MEC servers on which the service instances are to be placed using a serverless orchestration algorithm; and It includes a Serverless Computing Aware Function (SCAF) that is included in the mobile core network function and transmits computing and network metrics collected in advance from each MEC server for determining the serverless service instance placement information, and mobile underlay network metrics collected from the Network Data Analytics Function (NWDAF), to the MEC orchestrator, wherein The above one or more MEC servers prepare serverless service instances according to the determined deployment information, and A serverless computing support system in a mobile network in which an Edge Application Server Discovery Function (EASDF) included in the functions of the mobile core network exchanges a PDU session and the deployment information with the SCAF in response to a serverless service request from a User Equipment (UE) to establish a data plane routing path for the UE.
  8. In Paragraph 7, A serverless computing support system in a mobile network that uses the above serverless orchestration algorithm as input data to determine pre-warmed service instance placement information to prepare for future request traffic.
  9. In Paragraph 7, The above MEC orchestrator transmits the deployment information to one or more MEC servers through the MEC platform manager, and A serverless computing support system in a mobile network in which a FaaS framework running on one or more of the above MEC servers performs serverless service instance placement operations to prepare a required number of pre-warmed service instances.
  10. In Paragraph 7, The above mobile core network function further includes a Session Management Function (SMF) that, after receiving a PDU session setup request for the serverless service of the UE, provides the location of the UE and the service domain name to the SCAF. The above SCAF is a serverless computing support system in a mobile network that determines a destination by considering the computing resource consumption and request provision status of the current serverless service, which are updated by the computing metric agent of each MEC server.
  11. In Paragraph 7, The above SCAF is a serverless computing support system in a mobile network that determines one of a plurality of MEC servers, prepared with pre-warmed serverless service instances capable of providing more available requests, as the target MEC server.
  12. As a serverless computing support device in a mobile network, processor; and It includes memory connected to the above processor, The above memory is, Receive computing and network metrics pre-collected from each MEC server from the Serverless Computing Aware Function (SCAF) included in the mobile core network function, and mobile underlay network metrics collected from the Network Data Analytics Function (NWDAF), and To determine serverless service instance placement information including the number of pre-warmed serverless service instances and one or more MEC servers on which the service instances will be deployed, using a serverless orchestration algorithm with reference to the above metric. A serverless computing support device in a mobile network that stores program instructions executed by the above processor.

Description

Method, apparatus and system for supporting serverless computing in mobile network The present invention relates to a method, device, and system for supporting serverless computing in a mobile network, and more specifically, to a method and device for dynamically establishing a Protocol Data Unit session for a User Equipment (UE) request targeting a serverless computing service running at a Mobile Edge Computing (MEC) site in relation to the field of mobile network traffic steering. Serverless computing is a service deployment paradigm that can provide dynamic automatic scaling capabilities for resources allocated to services based on actual service demand. In serverless computing, if there are no service requests, the number of service instances can be reduced to zero, which can minimize the deployment resource costs for the service provider. However, as service demand increases, initializing new service instances takes a long time, which can cause delays; this is known as the cold start problem in serverless computing. A common method applied to various serverless computing orchestration and placement algorithms to address the cold start problem is to pre-warm serverless service instances before actual request traffic arrives. Based on the expected upcoming request demand for serverless services and the current resource usage conditions of the serverless computing sites, serverless orchestration algorithms pre-initialize an appropriate number of serverless service instances and determine their proper placement across various computing sites. When actual requests arrive, the cold start period is avoided, and the serverless service instances are already ready to process user requests. To apply pre-warmed serverless orchestration methods in a mobile edge network, the mobile network control plane must know the deployment and request serving capability of pre-warmed serverless service instances. Without this information, UE requests may be sent to MEC sites that lack pre-warmed serverless service instances or to MEC sites that have already reached their maximum request serving capacity. In both cases, new service instances are automatically scaled up, leading to cold start issues. Current mobile edge computing capabilities standardized by ETSI 3GPP lack support for steering UE requests to MEC sites with pre-warmed serverless service instances. ETSI 3GPP Technical Specification (TS) 23.501 defines Application Function (AF) traffic impacts to enable mobile networks to steering UE requests to optimal MEC sites. Figure 1 is a diagram illustrating all ETSI 3GPP defined MEC-related information that can be provided in an AF traffic impact request. Referring to FIG. 1, the target application of the AF traffic impact request can be provided in terms of an application identifier (ID), a Domain Name System (DNS) Fully Qualified Domain Name (FQDN), a Data Network Name (DNN), optional Network Slice Selection Assistance Information (NSSAI), and a Public Land Mobile Network (PLMN) identifier. The optimal MEC site for steering UE requests can be provided with the Data Network Access Identifier (DNAI), the Edge Application Server (EAS) IP address, and the corresponding N6 traffic information. The UEs affected by AF traffic impact requests can be determined to be all UEs, UE groups, or individual UEs. There are two disadvantages to using this information in a mobile network serverless edge computing orchestration environment. The first drawback is the lack of information regarding the placement of pre-warmed serverless service instances and request serving capabilities. The EAS information for the defined AF traffic impact request information can only specify MEC sites hosting pre-warmed serverless service instances. This allows the mobile network to prevent UE requests from being routed to MEC sites that do not have pre-warmed serverless service instances. Additional information required is the maximum request serving capacity of the pre-warmed serverless instances running at the MEC sites. With this information, the mobile network control plane can properly distribute the load of UE requests between MEC sites that have pre-warmed serverless instances. The second drawback lies in the method of defining the UEs affected by AF traffic impact requests. If the target UE identifier is "all UEs," steering all UE requests targeting the same serverless service to the same pre-warmed serverless service instance can exceed maximum serving capacity, potentially leading to new instance initialization and cold start issues. Meanwhile, if the target UE identifier is an individual UE or a group of UEs, the AF requires UE information to specify it in the target UE identifier information field. Since the AF is located at the edge, it can only obtain this information after the UE request reaches the MEC site. Cold start issues can occur when the request arrives and during the AF traffic impact process. Figure 1 is a diagram illustrating MEC-rel