WO2023124309A1

WO2023124309A1 - Cloud native upf signaling plane load balancing selection method and system

Info

Publication number: WO2023124309A1
Application number: PCT/CN2022/121299
Authority: WO
Inventors: 石凯; 刘辉; 赵臻
Original assignee: 浪潮通信技术有限公司
Priority date: 2021-12-27
Filing date: 2022-09-26
Publication date: 2023-07-06
Also published as: CN114363963A

Abstract

The present application provides a cloud native user plane function (UPF) signaling plane load balancing selection method and system. The method comprises: when a packet forwarding control protocol (PFCP) session is established, determining a corresponding UPF service instance in a UPF cluster on the basis of a UPF signaling plane load balancing selection algorithm; and processing, on the basis of the UPF service instance, a PFCP message between a session management function (SMF) and the UPF. According to the present application, a cloud native UPF signaling plane load balancing selection algorithm is used in a cloud native core network to process signaling communication between an external SMF and a UPF cluster, so that a containerized UPF cluster can meet the requirements of elasticity, reliability, and expansibility.

Description

A cloud-native UPF signaling plane load balancing selection method and system

Cross References to Related Applications

This application claims the priority of the Chinese patent application with application number 2021116149961 and titled "A cloud-native UPF signaling plane load balancing selection method and system" filed on December 27, 2021, which is incorporated by reference in its entirety This article.

technical field

The present application relates to the technical field of mobile communication, and in particular to a cloud-native UPF signaling plane load balancing selection method and system.

Background technique

In 5G (5th Generation Mobile Communication Technology, fifth-generation mobile communication technology), the 5G core network (5GC) is split into two parts: CPF (Control Plane Function, control plane function) and UPF (User Plane Function, user plane function). class, as shown in Figure 1.

The control plane SMF (Session Management function, session management function) configures UPF through PFCP (Packet Forwarding Control Protocol, message forwarding control protocol). (Data Network, data network) to establish a tunnel, and according to the forwarding rules configured by the tunnel information and PFCP, the UE (User Equipment, user equipment) data is forwarded, discarded, cached, and QoS operations are processed. In the cloud-native core network, when the UPF is decomposed into multiple running service instances, it is necessary to solve the problem of how to route and select the UPF service instance for the PFCP message sent from the control plane to the UPF.

Therefore, it is necessary to propose a new method for load balancing the cloud-native UPF signaling plane to solve the problem of how to route and select UPF service instances for PFCP messages sent from the control plane to UPF.

Contents of the invention

The present application provides a cloud-native UPF signaling plane load balancing selection method and system to solve the defect in the prior art of how to route and select UPF service instances for PFCP messages sent to UPF from the control plane in the cloud-native core network.

In the first aspect, the present application provides a method for selecting cloud-native UPF signaling plane load balancing, including:

When the message forwarding control protocol PFCP session is established, the corresponding UPF service instance in the UPF cluster is determined based on the user plane function UPF signaling plane load balancing selection algorithm;

Based on the UPF service instance, the PFCP message between the session management function SMF and the UPF is processed.

According to a cloud-native UPF signaling plane load balancing selection method provided by the present application, when the PFCP session of the message forwarding control protocol is established, the corresponding UPF in the UPF cluster is determined based on the user plane function UPF signaling plane load balancing selection algorithm Service instances, including:

Obtaining a signaling message, if it is judged that the signaling message is a Request type, then further confirming whether the signaling message is a session-level message;

If it is judged that the signaling message is a session-level message, then further confirm whether the signaling message is PFCPSessionEstablishmentRequest, if so, call the metrics interface, calculate the load load of each UPF service instance, determine the minimum load service instance, otherwise obtain The UP-SEID carried in the signaling message is compared with the SEID-UPF table to confirm the UPF service instance;

If it is determined that the signaling message is not a session-level message, randomly select a service instance as the UPF service instance.

According to a cloud-native UPF signaling plane load balancing selection method provided by the present application, the calling of the metrics interface calculates the load load of each UPF service instance and determines the minimum load service instance, including:

Through the metrics interface, obtain the CPU occupancy rate, memory occupancy rate, service instance capacity and the current session number of the service instance of each UPF service instance;

If it is determined that the CPU occupancy rate is greater than the first threshold, or that the memory occupancy rate is greater than the second threshold, or that the ratio of the number of current sessions of the service instance to the capacity of the service instance is greater than the third threshold, then it is determined that the load is positive infinity;

If it is determined that the CPU occupancy rate is not greater than the first threshold, the memory occupancy rate is not greater than the second threshold, and the ratio of the current session number of the service instance to the service instance capacity is not greater than the third threshold, then determine the load is a weighted summation of the CPU occupancy rate, the memory occupancy rate, and the ratio of the number of current sessions of the service instance to the capacity of the service instance.

According to a cloud native UPF signaling plane load balancing selection method provided by this application, the UP-SEID carried in the signaling message is obtained, the UP-SEID is compared with the SEID-UPF table, and the UPF service is confirmed Examples include:

If there is a service instance corresponding to the UP-SEID in the SEID-UPF table, then determining that the service instance corresponding to the UP-SEID is the UPF service instance;

If there is no service instance corresponding to the UP-SEID in the SEID-UPF table, randomly select a service instance as the UPF service instance.

According to a cloud-native UPF signaling plane load balancing selection method provided by the present application, when the PFCP session of the message forwarding control protocol is established, the corresponding UPF in the UPF cluster is determined based on the user plane function UPF signaling plane load balancing selection algorithm Service instance, also includes:

Obtain a signaling message, and if it is judged that the signaling message is not a Request type, then further confirm whether the signaling message is PFCPSessionEstablishmentResponse;

If it is judged that the signaling message is PFCPSessionEstablishmentResponse, then obtain the session endpoint identifier in the message header and process the SEID field, otherwise send the signaling message to the outside and end the process;

Confirming whether there is a service instance corresponding to the SEID field in the SEID-UPF table based on the SEID field, if it exists, updating the SEID of the corresponding service instance, sending the signaling message to the outside and ending the process, Otherwise, the UPF service instance is determined based on the capacity of the SEID-UPF table.

According to a cloud-native UPF signaling plane load balancing selection method provided by the present application, the determination of the UPF service instance based on the capacity of the SEID-UPF table includes:

If it is judged that the SEID-UPF table is full, then replace the SEID-UPF table based on the least recently used page replacement algorithm LRU, send the signaling message to the outside and end the process;

Otherwise, insert the SEID field and service instance record into the SEID-UPF table, send the signaling message to the outside and end the process.

In the second aspect, the present application also provides a cloud-native UPF signaling plane load balancing selection system, including:

The selection module is used to determine the corresponding UPF service instance in the UPF cluster based on the user plane function UPF signaling plane load balancing selection algorithm when the PFCP session is to be established;

The processing module is configured to process the PFCP message between the session management function SMF and the UPF based on the UPF service instance.

In a third aspect, the present application also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor. Describe the steps of the cloud-native UPF signaling plane load balancing selection method.

In the fourth aspect, the present application also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, load balancing of the cloud-native UPF signaling plane as described in any one of the above is realized. Steps to choose a method.

In the fifth aspect, the present application further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of any one of the cloud-native UPF signaling plane load balancing selection methods described above are implemented.

The cloud-native UPF signaling plane load balancing selection method and system provided by this application process the signaling communication between the external SMF and the UPF cluster through the cloud-native UPF signaling plane load balancing selection algorithm in the cloud-native core network, Make the containerized UPF cluster meet the requirements of elasticity, reliability and scalability.

Description of drawings

In order to more clearly illustrate the technical solutions in this application or the prior art, the accompanying drawings that need to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings in the following description are the present For some embodiments of the application, those skilled in the art can also obtain other drawings based on these drawings without creative work.

FIG. 1 is a structural diagram of core network user plane functions and control plane functions provided by the prior art;

Fig. 2 is one of the flow diagrams of the cloud native UPF signaling plane load balancing selection method provided by the present application;

Fig. 3 is the architecture diagram of the cloud-native UPF signaling plane load balancer provided by this application;

Fig. 4 is a schematic diagram of the format of the PFCP message header provided by the present application;

Fig. 5 is the second schematic flow diagram of the cloud-native UPF signaling plane load balancing selection method provided by the present application;

FIG. 6 is a schematic structural diagram of a cloud-native UPF signaling plane load balancing selection system provided by the present application;

FIG. 7 is a schematic structural diagram of an electronic device provided by the present application.

Detailed ways

In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the accompanying drawings in this application. Obviously, the described embodiments are part of the embodiments of this application , but not all examples. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.

Fig. 2 is one of the schematic flow diagrams of the cloud-native UPF signaling plane load balancing selection method provided by this application, as shown in Fig. 2, including:

Step S1, when the message forwarding control protocol PFCP session is to be established, determine the corresponding UPF service instance in the UPF cluster based on the user plane function UPF signaling plane load balancing selection algorithm;

Step S2, based on the UPF service instance, process the PFCP message between the session management function SMF and the UPF.

It should be noted that in the process of core network cloud-native evolution, the key is to split network functions into several fine-grained micro-services according to functional dimensions, and combine virtualization and micro-service governance to achieve cloud-native elasticity, reliability and Scalability requirements.

This application proposes a cloud-native UPF signaling plane load balancing algorithm, which is deployed on the UPF signaling plane load balancer device. The device selects the service of the UPF cluster when the PFCP session is established according to the cloud-native UPF signaling plane load balancing selection algorithm. instance, and keep the session consistent through the sticky policy. Through this algorithm, the requirements of cloud-native UPF cluster elasticity, reliability and scalability can be achieved. Figure 3 shows the architecture diagram of the cloud-native UPF signaling plane load balancer.

For the N4 interface between SMF and UPF, it is the interface between the control plane and the forwarding plane of the 5G core network, and SEID (Session Endpoint Identifier Handling, session endpoint identifier handling) is used as the PFCP between the control plane network element entities (that is, SMF and UPF) The unique identifier of the session context, F-SEID (Fully Qualified SEID), includes the IP address of the PFCP entity and the assigned SEID. It is independently allocated by the CPF and UPF entities and sent to the other party in the PFCP message. The other PFCP entity uses this ID as identification The unique identifier of the PFCP session. A PFCP endpoint MUST use the locally assigned SEID value of the peer PFCP endpoint when sending messages. SEID values are exchanged between PFCP endpoints using PFCP messages. A PFCP entity sends a SEID value to a peer PFCP entity and expects to receive all subsequent control plane messages related to this PFCP session via the "F-SEID" IE. Messages related to a PFCP session shall share the same F-SEID as the PFCP session. The F-SEID shall be released after the PFCP session is released.

The optional SEID of the PFCP message header occupies 8 bytes. For node-related messages, the PFCP message header does not contain the SEID field; for session-related messages, the SEID must be included. In the session establishment request, SEID is set to all 0s. The general format of the PFCP message header is shown in Figure 4.

This application handles the signaling communication between the external SMF and the UPF cluster according to the cloud-native UPF signaling plane load balancing selection algorithm in the cloud-native core network, so that the containerized UPF cluster can meet the requirements of elasticity, reliability and scalability Require.

Based on the foregoing embodiments, step S1 includes:

Exemplarily, the flow chart shown in Figure 5 first monitors the signaling message, after obtaining the signaling message, checks whether the message is a Request type, and if so, further judges whether it is a session-level message, here for the node For related messages or session-related messages, the message header in PFCP needs to distinguish whether it contains SEID.

If it is further judged to be a session-level message, check whether the message is PFCPSessionEstablishmentRequest, if so, call the metrics interface, calculate the load load of each UPF service instance, and determine the minimum load service instance, otherwise obtain the UP-SEID carried in the signaling message , compared with the SEID-UPF table to obtain the UPF service instance.

If it is further judged that it is not a session-level message, the service instance is randomly selected as the UPF service instance.

This application judges that the type of the signaling message is the Request type, and then obtains the corresponding UPF service instance according to the corresponding routing strategy, so as to meet the requirements of elasticity, reliability and scalability of the UPF cluster.

Based on any of the above-mentioned embodiments, the calling of the metrics interface calculates the load load of each UPF service instance, and determines the minimum load service instance, including:

For example, call the metrics interface to calculate the load of each UPF service instance, using the following algorithm:

Obtain the CPU usage cpu, memory usage memory, service instance capacity and the current session number seNumber of the UPF service instance, and calculate the load of the UPF service instance, as follows:

According to the above calculation principle, the load of each UPF service instance is obtained, and the service instance with the smallest load is selected as the UPF service instance.

This application calculates the CPU occupancy rate, memory occupancy rate, service instance capacity, and the number of current sessions of the service instance respectively, and sets a certain threshold to obtain the load under high load and low load status, so as to select the service with the smallest load instance.

Based on any of the above embodiments, the acquiring the UP-SEID carried in the signaling message, comparing the UP-SEID with the SEID-UPF table, and confirming the UPF service instance includes:

Exemplarily, if it is further judged that it is not a session-level message, then obtain the UP-SEID carried in the signaling message, use the UP-SEID as the query condition, query the SEID-UPF table, and see if there is a service instance recorded by the SEID in the table , if the service instance with the SEID record is found in the table, select the instance as the UPF service instance, otherwise, randomly select the UPF service instance, send the signaling message to the UPF service instance, and end the process.

This application selects a specific UPF service instance by comparing the SEID records in the SEID-UPF table, which has the characteristics of high execution efficiency and high reliability.

Based on any of the above-mentioned embodiments, step S1 also includes:

Exemplarily, as shown in FIG. 5 , for another case of a signaling message, if it is determined that it is not a Request type, it is necessary to further confirm whether the signaling message is a PFCPSessionEstablishmentResponse, and if so, obtain the SEID field in the message header to The SEID in the message header is used as a query condition, and the SEID-UPF table is queried to see if there is a service instance of the SEID record in the table. If there is a service instance of the SEID record in the table, the SEID of the service instance is updated, and the information Send the message to the outside and end the process. If the service instance recorded by the SEID does not exist in the table, determine the UPF service instance based on the capacity of the SEID-UPF table.

In addition, if it is confirmed that the signaling message is not PFCPSessionEstablishmentResponse, the signaling message is directly sent to the outside and the process ends.

This application judges that the type of the signaling message is non-Request type, and then obtains the corresponding UPF service instance according to the corresponding routing strategy, so as to meet the requirements of elasticity, reliability and scalability of the UPF cluster.

Based on any of the above embodiments, the determining the UPF service instance based on the capacity of the SEID-UPF table includes:

Exemplarily, in this application, the determination of the UPF service instance based on the capacity of the SEID-UPF table includes:

Check whether the capacity of the SEID-UPF table is full. If the capacity is full, replace the SEID-UPF table according to the LRU (Least Recently Used, least recently used page replacement algorithm), and then send the signaling message to the outside and end process;

If the capacity of the above SEID-UPF table is not full, insert the SEID and service instance records into the table, then send the signaling message to the outside and end the process.

By judging the capacity of the SEID-UPF table, the present application further judges whether it is necessary to update the SEID information in the table, so as to obtain the service instance with the smallest load among the UPF service instances in time to ensure smoothness of business.

The following describes the complete process of the cloud-native UPF signaling plane load balancing selection method in Figure 5. As shown in Figure 5, the steps include:

(1) Algorithm starts;

(2) Monitor the signaling message, and check whether the message is a Request type after the signaling message is obtained. If the message type is Request type, go to step (3); otherwise, go to step (8);

(3) Check whether it is a session-level message. If it is a session-level message, go to step (4); otherwise, go to step (20);

(4) Check whether the message is PFCPSessionEstablishmentRequest, if the message is of this type, go to step (5); otherwise, go to step (17);

(5) Call the metrics interface to obtain the CPU occupancy rate cpu of the UPF service instance, the memory occupancy rate memory, the service instance capacity and the current session number seNumber of the service instance;

(6) Set the load of the UPF service instance as load:

(7) Calculate the load of each UPF service instance, and select the service instance with the smallest load;

(8) If it is not a Request type message, then check whether the message is PFCPSessionEstablishmentResponse, if it is this type of message, go to step (9); otherwise, go to step (15);

(9) obtain the SEID field in the message header;

(10) Use the SEID as a query condition to query the SEID-UPF table to see if there is a service instance with a record of the SEID in the table. If there is a service instance of the SEID record in the table, go to step (11); otherwise, go to step (12);

(11) update the SEID of the service instance, go to step (16);

(12) Check whether the SEID-UPF table is full. If the SEID-UPF table is full, proceed to step (13); otherwise, proceed to step (14);

(13) replace the SEID-UPF table according to the LRU algorithm, and proceed to step (16);

(14) insert the record of SEID and service instance in the table, go to step (16);

(15) If the message type is not PFCPSessionEstablishmentRespone, go to step (16);

(16) Send the signaling message to the outside, go to step (22);

(17) Obtain the UP-SEID carried in the signaling message;

(18) Use the UP-SEID as a query condition to query the SEID-UPF table to see if there is a service instance of the SEID record in the table; if there is a service instance of the SEID record in the table, go to step (19); otherwise, Go to step (20);

(19) Select the UPF service instance;

(20) Randomly select a UPF service instance;

(21) Send a signaling message to the UPF service instance;

(22) End the algorithm flow.

The following describes the cloud-native UPF signaling plane load balancing selection system provided by this application. The cloud-native UPF signaling plane load balancing selection system described below and the cloud-native UPF signaling plane load balancing selection method described above can refer to each other .

Fig. 6 is a schematic structural diagram of the cloud-native UPF signaling plane load balancing selection system provided by the present application, as shown in Fig. 6, including: a selection module 61 and a processing module 62, wherein:

The selection module 61 is used to determine the corresponding UPF service instance in the UPF cluster based on the user plane function UPF signaling plane load balancing selection algorithm when the PFCP session is to be established; the processing module 62 is used to based on the UPF service instance, Process the PFCP message between the session management function SMF and UPF.

This application handles the signaling communication between the external SMF and the UPF cluster through the cloud-native UPF signaling plane load balancing selection algorithm in the cloud-native core network, so that the containerized UPF cluster can meet the requirements of elasticity, reliability and scalability Require.

FIG. 7 illustrates a schematic diagram of the physical structure of an electronic device. As shown in FIG. 7, the electronic device may include: a processor (processor) 710, a communication interface (Communications Interface) 720, a memory (memory) 730 and a communication bus 740, Wherein, the processor 710 , the communication interface 720 , and the memory 730 communicate with each other through the communication bus 740 . The processor 710 can call the logic instructions in the memory 730 to execute the cloud-native UPF signaling plane load balancing selection method, which includes: when the packet forwarding control protocol PFCP session is established, based on the user plane function UPF signaling plane load balancing The selection algorithm determines the corresponding UPF service instance in the UPF cluster; based on the UPF service instance, the PFCP message between the session management function SMF and the UPF is processed.

In addition, the above-mentioned logic instructions in the memory 730 may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc., which can store program codes. .

On the other hand, the present application also provides a computer program product, the computer program product includes a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can Execute the cloud-native UPF signaling plane load balancing selection method provided by the above methods. The method includes: when the packet forwarding control protocol PFCP session is established, determine the corresponding load balancing selection algorithm in the UPF cluster based on the user plane function UPF signaling plane load balancing selection algorithm. The UPF service instance; based on the UPF service instance, the PFCP message between the session management function SMF and the UPF is processed.

In yet another aspect, the present application also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the cloud-native UPF signaling plane load balancing provided by the above-mentioned methods is implemented. The selection method includes: when the PFCP session of the message forwarding control protocol is to be established, determine the corresponding UPF service instance in the UPF cluster based on the user plane function UPF signaling plane load balancing selection algorithm; based on the UPF service instance, session management PFCP messages between functional SMF and UPF are processed.

The device embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network elements. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. It can be understood and implemented by those skilled in the art without any creative efforts.

Through the above description of the implementations, those skilled in the art can clearly understand that each implementation can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic discs, optical discs, etc., including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) execute the methods described in various embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, rather than limiting them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.

Claims

A cloud-native UPF signaling plane load balancing selection method, comprising:

When the message forwarding control protocol PFCP session is established, the corresponding UPF service instance in the UPF cluster is determined based on the user plane function UPF signaling plane load balancing selection algorithm;

Based on the UPF service instance, the PFCP message between the session management function SMF and the UPF is processed.
The cloud-native UPF signaling plane load balancing selection method according to claim 1, wherein, when the PFCP session of the message forwarding control protocol is established, the UPF signaling plane load balancing selection algorithm based on the user plane function determines the corresponding UPF cluster. UPF service instances, including:

Obtaining a signaling message, if it is judged that the signaling message is a Request type, then further confirming whether the signaling message is a session-level message;

If it is judged that the signaling message is a session-level message, then further confirm whether the signaling message is PFCPSessionEstablishmentRequest, if so, call the metrics interface, calculate the load load of each UPF service instance, determine the minimum load service instance, otherwise obtain The UP-SEID carried in the signaling message is compared with the SEID-UPF table to confirm the UPF service instance;

If it is determined that the signaling message is not a session-level message, randomly select a service instance as the UPF service instance.
According to the cloud-native UPF signaling plane load balancing selection method according to claim 2, wherein the calling metrics interface calculates the load load of each UPF service instance, and determines the minimum load service instance, including:

Through the metrics interface, obtain the CPU occupancy rate, memory occupancy rate, service instance capacity and the current session number of the service instance of each UPF service instance;

If it is determined that the CPU occupancy rate is greater than the first threshold, or that the memory occupancy rate is greater than the second threshold, or that the ratio of the number of current sessions of the service instance to the capacity of the service instance is greater than the third threshold, then it is determined that the load is positive infinity;

If it is determined that the CPU occupancy rate is not greater than the first threshold, the memory occupancy rate is not greater than the second threshold, and the ratio of the current session number of the service instance to the service instance capacity is not greater than the third threshold, then determine the load is a weighted summation of the CPU occupancy rate, the memory occupancy rate, and the ratio of the number of current sessions of the service instance to the capacity of the service instance.
The cloud-native UPF signaling plane load balancing selection method according to claim 2, wherein said obtaining the UP-SEID carried in the signaling message, comparing the UP-SEID with the SEID-UPF table, and confirming UPF service instances, including:

If there is a service instance corresponding to the UP-SEID in the SEID-UPF table, then determining that the service instance corresponding to the UP-SEID is the UPF service instance;

If there is no service instance corresponding to the UP-SEID in the SEID-UPF table, randomly select a service instance as the UPF service instance.
The cloud-native UPF signaling plane load balancing selection method according to claim 1, wherein, when the PFCP session of the message forwarding control protocol is established, the UPF signaling plane load balancing selection algorithm based on the user plane function determines the corresponding UPF cluster. The UPF service instance also includes:

Obtain a signaling message, and if it is judged that the signaling message is not a Request type, then further confirm whether the signaling message is PFCPSessionEstablishmentResponse;

If it is judged that the signaling message is PFCPSessionEstablishmentResponse, then obtain the session endpoint identifier in the message header and process the SEID field, otherwise send the signaling message to the outside and end the process;

Confirming whether there is a service instance corresponding to the SEID field in the SEID-UPF table based on the SEID field, if it exists, updating the SEID of the corresponding service instance, sending the signaling message to the outside and ending the process, Otherwise, the UPF service instance is determined based on the capacity of the SEID-UPF table.
The cloud-native UPF signaling plane load balancing selection method according to claim 5, wherein the determining the UPF service instance based on the capacity of the SEID-UPF table includes:

If it is judged that the SEID-UPF table is full, then replace the SEID-UPF table based on the least recently used page replacement algorithm LRU, send the signaling message to the outside and end the process;

Otherwise, insert the SEID field and service instance record into the SEID-UPF table, send the signaling message to the outside and end the process.
A cloud-native UPF signaling plane load balancing selection system, comprising:

The selection module is used to determine the corresponding UPF service instance in the UPF cluster based on the user plane function UPF signaling plane load balancing selection algorithm when the PFCP session is to be established;

The processing module is configured to process the PFCP message between the session management function SMF and the UPF based on the UPF service instance.
An electronic device, comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, when the processor executes the program, the computer program described in any one of claims 1 to 6 is realized. Describe the steps of the cloud-native UPF signaling plane load balancing selection method.
A non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the cloud-native UPF signaling plane load balancing selection method as described in any one of claims 1 to 6 is realized step.
A computer program product, including a computer program, when the computer program is executed by a processor, the steps of the cloud-native UPF signaling plane load balancing selection method according to any one of claims 1 to 6 are realized.