Micro-service deployment method and device, equipment and storage medium

Abstract

The invention discloses a micro-service deployment method, device and equipment and a storage medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: determining a plurality of to-be-deployed micro-services, edge servers and to-be-deployed applications corresponding to a government affair large model, micro-service information corresponding to the micro-service to be deployed, server information corresponding to the edge server and topological information corresponding to the application to be deployed are determined respectively; constructing a micro-service deployment model corresponding to the government affair large model based on the micro-service information, the server information and the topological information; solving the micro-service deployment model by using a preset model solving algorithm to obtain a micro-service deployment strategy corresponding to the government affair large model, and deploying the to-be-deployed micro-service to a preset government affair large model platform based on the micro-service deployment strategy; the preset model solving algorithm is an algorithm constructed based on a genetic algorithm and a tabu search algorithm. In this way, the optimal micro-service deployment scheme can be quickly found out in the complex edge network environment, and the resource utilization efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and particularly relates to a microservice deployment method, device, equipment, and storage medium. Background Art

[0002] At present, with the rapid development of artificial intelligence and big data technologies, data intelligence platforms centered around large models, such as intelligent question-and-answer platforms, etc., have extremely high requirements for network response latency, data privacy, and computing efficiency. However, there are many drawbacks in the microservice deployment of current cloud computing and large model platforms. For example, the response latency remains high, the resource utilization rate is low, and the cost is high. Currently, as an emerging architecture, the edge computing network architecture deploys servers with computing capabilities on the edge side close to the terminal device, which can fully integrate the scattered computing resources on the edge side and achieve deep coordination of cloud, network, edge, and terminal. This architecture greatly shortens the data transmission path, effectively reduces network latency, and improves the real-time performance and privacy of data processing. The microservice architecture can disassemble an application program into multiple independent microservice units, and each unit is responsible for a specific business function, achieving independent deployment, flexible combination, and efficient expansion of services. However, the integration of the edge computing network and the microservice architecture also brings challenges. Reasonably deploying microservices in the edge computing network to minimize the response latency, maximize the resource utilization rate, and meet various resource constraint conditions has become a key problem. Traditional deployment strategies are extremely likely to fall into local optimal solutions in a complex edge network environment, especially when facing large-scale microservice deployments, resulting in low deployment efficiency and being unable to find the global optimal solution. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a microservice deployment method, device, equipment, and storage medium, which can quickly find the optimal microservice deployment plan in a complex edge network environment, effectively reduce the service response latency, and improve the resource utilization efficiency. The specific solutions are as follows:

[0004] In a first aspect, the present application provides a microservice deployment method, including:

[0005] Determine a number of microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model, and respectively determine the microservice information corresponding to each of the microservices to be deployed, the server information corresponding to the edge servers, and the topology information corresponding to the applications to be deployed;

[0006] Construct a microservice deployment model corresponding to the government affairs large model based on the microservice information, the server information, and the topology information;

[0007] Solve the microservice deployment model using a preset model solving algorithm to obtain the microservice deployment strategy corresponding to the government affairs large model, and deploy several of the to-be-deployed microservices corresponding to the government affairs large model to a preset government affairs large model platform based on the microservice deployment strategy; the preset model solving algorithm is an algorithm constructed based on a genetic algorithm and a tabu search algorithm.

[0008] Optionally, determining several to-be-deployed microservices corresponding to the government affairs large model includes:

[0009] Determine a target parameter slicing rule based on the business type of the government affairs large model, and slice the parameters of the government affairs large model based on a preset parameter slicing technique and the target parameter slicing rule;

[0010] Slice the functions of the government affairs large model based on a preset function decoupling technique, and combine the sliced parameters and the sliced functions to obtain the inference service units corresponding to each function of the government affairs large model;

[0011] Determine the to-be-deployed microservices according to the inference service units, and organize and coordinate the to-be-deployed microservices based on a preset communication protocol so that the to-be-deployed microservices can perform data exchange through the preset communication protocol.

[0012] Optionally, respectively determining the microservice information corresponding to each of the to-be-deployed microservices and the server information corresponding to the edge server includes:

[0013] Determine the target quantity of the to-be-deployed microservices, and determine the core quantity of the edge server based on the target quantity, where the target quantity and the core quantity correspond one by one;

[0014] Determine the computing resources of the edge server, and determine the storage resources of the edge server based on the memory capacity and memory bandwidth of the edge server;

[0015] Determine the communication capabilities of the edge server based on the inter-node bandwidth and cross-domain transmission delay of the edge server, and determine the real-time status of the edge server based on the current load rate and available resource capacity of the edge server.

[0016] Optionally, determining the topology information corresponding to the to-be-deployed application includes:

[0017] Determine the target to-be-deployed microservices traversed during a single run of the to-be-deployed application, and determine the topology structure corresponding to the to-be-deployed application based on a preset directed acyclic graph structure and the target to-be-deployed microservices;

[0018] Determine the computing amount corresponding to the target microservice to be deployed during a single run of the application to be deployed, and determine the data transmission amount between the target microservices to be deployed;

[0019] Determine the average arrival rate of the target requests corresponding to the application to be deployed, and execute the target requests based on the preset first-come-first-served principle.

[0020] Optionally, the constructing the microservice deployment model corresponding to the government affairs large model based on the microservice information, the server information, and the topology information includes:

[0021] Determine the constraint conditions corresponding to the microservice deployment model based on the microservice information and the server information, and construct the objective function corresponding to the microservice deployment model based on the topology information;

[0022] Set the decision variables corresponding to the microservice deployment model, and construct the microservice deployment model corresponding to the government affairs large model based on the constraint conditions, the objective function, and the decision variables; the decision variables are used to determine the deployment status of the microservices to be deployed and the edge servers.

[0023] Optionally, the solving the microservice deployment model by using a preset model solving algorithm includes:

[0024] Initialize the first target parameters corresponding to the preset hybrid genetic tabu search algorithm, encode the individuals in the population based on the first target parameters, and generate an initial population based on the preset greedy algorithm and the encoded individuals; the individuals are used to represent the microservice deployment strategies;

[0025] Determine the fitness value of the individuals based on the objective function corresponding to the microservice deployment model, and update the initial population based on the preset selection operation, preset crossover operation, preset mutation operation, and the fitness value;

[0026] Adjust the updated initial population based on the preset tabu search algorithm to obtain the adjusted initial population, and determine the target population based on the preset iteration number threshold, the adjusted initial population, and the fitness value to obtain the microservice deployment strategy corresponding to the government affairs large model.

[0027] Optionally, the adjusting the updated initial population based on the preset tabu search algorithm to obtain the adjusted initial population includes:

[0028] Initialize the second target parameters corresponding to the preset tabu search algorithm, where the second target parameters include the length of the tabu list, the maximum number of algorithm iterations, the neighborhood function, the fitness function, and the aspiration rule;

[0029] Determine the neighborhood solution corresponding to the target individual in the updated initial population based on the neighborhood function, determine the target fitness value corresponding to the target individual based on the fitness function, and determine the candidate solution from the neighborhood solutions based on the target fitness value;

[0030] Adjust the updated initial population based on the maximum number of algorithm iterations, the aspiration rule, and the candidate solution to obtain the adjusted initial population.

[0031] In a second aspect, the present application provides a microservice deployment device, including:

[0032] An information determination module, configured to determine a number of microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model, and respectively determine the microservice information corresponding to each microservice to be deployed, the server information corresponding to the edge server, and the topology information corresponding to the application to be deployed;

[0033] A model construction module, configured to construct a microservice deployment model corresponding to the government affairs large model based on the microservice information, the server information, and the topology information;

[0034] A microservice deployment module, configured to solve the microservice deployment model by using a preset model solving algorithm to obtain a microservice deployment strategy corresponding to the government affairs large model, so as to deploy a number of microservices to be deployed corresponding to the government affairs large model to a preset government affairs large model platform based on the microservice deployment strategy; the preset model solving algorithm is an algorithm constructed based on a genetic algorithm and a tabu search algorithm.

[0035] In a third aspect, the present application provides an electronic device, including:

[0036] A memory, configured to store a computer program;

[0037] A processor, configured to execute the computer program to implement the foregoing microservice deployment method.

[0038] In a fourth aspect, the present application provides a computer-readable storage medium, configured to store a computer program; wherein, when the computer program is executed by a processor, the foregoing microservice deployment method is implemented.

[0039] In this application, first, a number of microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model are determined, and microservice information corresponding to each of the microservices to be deployed, server information corresponding to the edge servers, and topology information corresponding to the applications to be deployed are respectively determined; then, a microservice deployment model corresponding to the government affairs large model is constructed based on the microservice information, the server information, and the topology information; finally, a preset model solving algorithm is used to solve the microservice deployment model to obtain a microservice deployment strategy corresponding to the government affairs large model, so as to deploy the number of microservices to be deployed corresponding to the government affairs large model to a preset government affairs large model platform based on the microservice deployment strategy; the preset model solving algorithm is an algorithm constructed based on a genetic algorithm and a tabu search algorithm. As can be seen from the above, in this application, first, the microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model, as well as the corresponding microservice information, server information, and topology information, are determined. Then, a microservice deployment model is constructed according to the microservice information, server information, and topology information. Finally, the microservice deployment model is solved using a hybrid genetic tabu search algorithm to obtain a microservice deployment strategy. In this way, in this application, the microservice computing resource requirements, the computing capabilities of edge servers, and the application topology structure dependency relationships are considered, and the optimal microservice deployment plan is quickly found in a complex edge network environment through a genetic algorithm and a tabu search algorithm, effectively reducing the service response delay and improving the resource utilization efficiency and real-time decision-making ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0041] Figure 1 It is a flowchart of a microservice deployment method provided by this application;

[0042] Figure 2 It is a schematic diagram of a specific deployment plan of microservices on edge servers provided by this application;

[0043] Figure 3 It is a schematic diagram of a specific application topology structure provided by this application;

[0044] Figure 4 It is a specific flowchart of a specific hybrid genetic tabu search algorithm provided by this application;

[0045] Figure 5 It is a specific flowchart of a specific tabu search algorithm provided by this application;

[0046] Figure 6 A specific flowchart of a microservice deployment method provided for this application;

[0047] Figure 7 A schematic diagram of a specific application topology structure provided for this application;

[0048] Figure 8 A specific flowchart of a microservice deployment method provided for this application;

[0049] Figure 9 A schematic diagram of the structure of a microservice deployment device provided for this application;

[0050] Figure 10 A structural diagram of an electronic device provided for this application. Detailed implementation manners

[0051] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0052] Currently, as an emerging architecture, the edge computing network architecture deploys servers with computing capabilities on the edge side close to the terminal device, which can fully integrate the scattered computing resources on the edge side and achieve deep coordination of cloud, network, edge, and terminal. This architecture greatly shortens the data transmission path, effectively reduces network latency, and improves the real-time performance and privacy of data processing at the same time. The microservice architecture can disassemble an application into multiple independent microservice units, and each unit is responsible for a specific business function, achieving independent deployment, flexible combination, and efficient expansion of services. However, the integration of the edge computing network and the microservice architecture also brings challenges. Reasonably deploying microservices in the edge computing network to minimize the response latency, maximize the resource utilization rate, and meet various resource constraint conditions has become a key problem. Traditional deployment strategies are extremely likely to fall into local optimal solutions in a complex edge network environment, especially when facing large-scale microservice deployments, resulting in low deployment efficiency and being unable to find the global optimal solution. Therefore, this application provides a microservice deployment solution that can quickly find the optimal microservice deployment solution in a complex edge network environment, effectively reduce the service response latency, and improve the resource utilization efficiency.

[0053] See Figure 1 As shown, an embodiment of the present invention discloses a microservice deployment method, which may include:

[0054] Step S11: Determine a number of microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model, and respectively determine the microservice information corresponding to each of the microservices to be deployed, the server information corresponding to the edge servers, and the topology information corresponding to the applications to be deployed.

[0055] In this embodiment, the government affairs large model platform can split the government affairs large model into inference service units that can be independently deployed through parameter slicing and function decoupling technologies. Specifically, first, determine the target parameter slicing rule based on the business type of the government affairs large model, and slice the parameters of the government affairs large model based on the preset parameter slicing technology and the target parameter slicing rule; then slice the functions of the government affairs large model based on the preset function decoupling technology, and combine the sliced parameters and the sliced functions to obtain the inference service units corresponding to the functions of the government affairs large model; finally, determine the microservices to be deployed according to the inference service units, and organize and coordinate the microservices to be deployed based on the preset communication protocol so that the microservices to be deployed can exchange data through the preset communication protocol. It should be noted that each microservice corresponds to a specific function module of the large model, such as the data lineage tracking engine function module and the cross-domain semantic alignment function module. The key features of the microservice include: computing features, such as parameter scale and single-inference computing volume; memory features, such as parameter loading memory requirements and runtime peak memory; hardware dependencies, such as accelerator type requirements, namely GPU (Graphics Processing Unit), NPU (Neural Processing Unit), FPGA (Field-Programmable Gate Array); service level, such as latency sensitivity. Different microservices can be organized and coordinated with each other through lightweight communication protocols to jointly complete complex business logics. In a specific implementation, determine the first set I composed of the microservices to be deployed, and obtain the basic information of the microservices. The microservice information corresponding to the microservices to be deployed includes: microservice , required computing speed , required memory resources , required video memory resources .

[0056] It can be understood that in this embodiment, a core of an edge server is allocated to each microservice to be deployed, that is, the maximum number of microservices that can be deployed on an edge server is equal to the number of cores of the edge server. In addition, each microservice is only deployed on one edge server and only one instance is deployed. An example of the deployment scheme of the microservice in the edge server is as Figure 2As shown in the figure. The key features of the heterogeneous edge server set include: computing resources, such as accelerator computing power and the number of CPU cores; storage resources, such as memory capacity and memory bandwidth; communication capabilities, such as inter-node bandwidth and cross-domain transmission delay; real-time status, such as the current load rate and available resource capacity. Specifically, first determine the target number of microservices to be deployed, and determine the number of cores of the edge server based on the target number, with the target number corresponding one-to-one to the number of cores; then determine the computing resources of the edge server, and determine the storage resources of the edge server based on the memory capacity and memory bandwidth of the edge server; finally, determine the communication capabilities of the edge server based on the inter-node bandwidth and cross-domain transmission delay of the edge server, and determine the real-time status of the edge server based on the current load rate and available resource capacity of the edge server. In a specific implementation, determine the second set J composed of available edge servers, and obtain the basic information of the servers. The server information corresponding to the edge server includes: the number of cores of the server , the total memory resources of the server , the server has a processing speed of , the data transmission speed between edge server j and edge server k is , where and , the required video memory resources .

[0057] It should be noted that each application is composed of different microservices combined with each other. Each application can provide more complete and comprehensive services compared to a single microservice to support data cleaning services, metadata annotation services, data lineage tracking services, data quality assessment services, etc. of the government affairs large model, so as to build a data asset map application, open up a cross-department data sharing channel application, and build a real-time data governance workflow. In addition, the requests of all applications are executed in sequence according to the rule of first come, first served. In a specific implementation, determine the first set L composed of applications to be deployed, and obtain the basic information of the applications to be deployed. Specifically, first determine the target microservices to be deployed traversed during a single run of the application to be deployed, and determine the topological structure corresponding to the application to be deployed based on the preset directed acyclic graph structure and the target microservices to be deployed. An example of the topological structure of the application is shown as Figure 3 ; then determine the computing amount corresponding to the target microservices to be deployed during a single run of the application to be deployed, and determine the data transmission amount between the target microservices to be deployed; finally, determine the average arrival rate of the target requests corresponding to the application to be deployed, and execute the target requests based on the preset first come, first served principle. In this embodiment, the application basic information includes the topological structure of the application , , , is a directed acyclic graph and 。 is a vertex set, and which represents the set of microservices traversed in a single run of the application In each microservice is a vertex of a directed acyclic graph; is an arc set, which represents the set of dependencies between microservices required for a single run of the application where represents that the microservice corresponding to vertex n can only run after the microservice corresponding to vertex m is completed, where . The basic information of the application to be deployed also includes the amount of computation of the microservice corresponding to vertex m during a single run of the application , the amount of data transmitted from the microservice corresponding to vertex m to the microservice corresponding to vertex n after the microservice corresponding to vertex m is executed during a single run of the application , and the average arrival rate of requests for the application .

[0058] Step S12: Construct a microservice deployment model corresponding to the government affairs large model based on the microservice information, the server information, and the topology information.

[0059] In this embodiment, first, determine the constraint conditions corresponding to the microservice deployment model based on the microservice information and the server information, and construct the objective function corresponding to the microservice deployment model based on the topology information; then set the decision variables corresponding to the microservice deployment model, and construct the microservice deployment model corresponding to the government affairs large model based on the constraint conditions, the objective function, and the decision variables; where the decision variables are used to determine the deployment status of the microservices to be deployed and the edge servers. Specifically, the specific meanings of the decision variables are as follows:

[0060] ;

[0061] where . The objective function of the microservice deployment model is as follows:

[0062] ;

[0063] where is the single-run delay of the application , and is the average arrival rate of requests for application h, where . The constraint conditions of the microservice deployment model are as follows:

[0064] (1) ;

[0065] (2) ;

[0066] (3) ;

[0067] (4) ;

[0068] (5) ;

[0069] (6) ;

[0070] Among them, constraint (1) means that the computing speed required by any microservice i does not exceed the computing speed of its corresponding deployed edge server. Constraint (2) means that the video memory resource of any edge server j is not lower than the sum of the video memory resources required by the microservices deployed on it. Constraint (3) means that the memory resource of any edge server j is not lower than the sum of the memory resources required by the microservices deployed on it. Constraint (4) means that the number of microservices deployed on any edge server j does not exceed its maximum deployable number of microservices. Constraint (5) means that each microservice is deployed on only one edge server. Constraint (6) means the value range of the decision variable. It should be noted that the communication time between microservices on the same edge server is not counted.

[0071] It should be pointed out that define as the time required from the start of running the application to the completion of the execution of the microservice corresponding to the vertex n. The specific meaning of the is as follows:

[0072] ;

[0073] Among them, is the set composed of vertices with in-degree zero in ; is the running delay of the microservice corresponding to the vertex n; is the data transmission delay from vertex m to vertex n; is the time required for all predecessor microservices from the start of running the application to transmit data to the microservice corresponding to the vertex n; is the computing amount of the microservice corresponding to the vertex n during a single run of the application ;

[0074] ;

[0075] ;

[0076] .

[0077] In this embodiment, Perform a topological sort on all vertices in , and apply The single - run delay of

[0078] ;

[0079] Among them, is a set composed of vertices with an out - degree of zero in

[0080] In this embodiment, the transmission delay, running delay of microservices, and the topological structure relationship in the application are comprehensively considered. The transmission delay, running delay, and topological dependence relationship between microservices are accurately quantified. Combining the heterogeneous resources and communication capabilities between edge servers, a microservice deployment model is established. Based on the microservice deployment model, the matching relationship between microservices, applications, and servers can be accurately described, with high adaptability. It is especially suitable for microservice deployment scenarios such as large - model platforms and large - scale, high - dynamic government large - model platforms. It can flexibly adjust the deployment plan according to the dynamically changing business requirements and resource status to ensure the stable and efficient operation of the system in high - load and high - concurrency environments.

[0081] Step S13: Use a preset model - solving algorithm to solve the microservice deployment model to obtain the microservice deployment strategy corresponding to the government large - model, so as to deploy several to - be - deployed microservices corresponding to the government large - model to a preset government large - model platform based on the microservice deployment strategy; the preset model - solving algorithm is an algorithm constructed based on a genetic algorithm and a tabu search algorithm.

[0082] In this embodiment, the above - mentioned use of the preset model - solving algorithm to solve the microservice deployment model may include: First, initialize the first target parameters corresponding to the preset hybrid genetic tabu search algorithm, encode the individuals in the population based on the first target parameters, and generate an initial population based on a preset greedy algorithm and the encoded individuals; the individuals are used to represent microservice deployment strategies; then determine the fitness value of the individuals based on the objective function corresponding to the microservice deployment model, and update the initial population based on a preset selection operation, a preset crossover operation, a preset mutation operation, and the fitness value; finally, adjust the updated initial population based on a preset tabu search algorithm to obtain the adjusted initial population, and determine the target population based on a preset iteration - number threshold, the adjusted initial population, and the fitness value to obtain the microservice deployment strategy corresponding to the government large - model. Specifically, as shown in Figure 4 The process of the hybrid genetic tabu search algorithm is as follows:

[0083] S1: Initialize the parameters of the hybrid genetic tabu search algorithm. The specific parameters include the population size K, the crossover probability , the mutation probability , and the maximum number of algorithm iterations G.

[0084] S2: Initialize the population. Encode the individuals in the population and initialize the population.

[0085] S3: Calculate the fitness value fit of each individual in the initial population.

[0086] S4: Selection operation. Adopt the roulette wheel selection method with the elite retention strategy.

[0087] S5: Crossover operation. Adopt the multi-point crossover method.

[0088] S6: Mutation operation. Select the microservice with the longest current execution time in each application, and randomly select the server location where the microservice is deployed.

[0089] S7: Perform local tabu search to adjust the population and obtain the latest generation of population.

[0090] S8: Accumulate the number of iterations, and judge whether the maximum number of iterations is reached. If not, go to S3.

[0091] S9: Judge the fitness value of the individuals in the current population, and output the individual with the largest fitness value.

[0092] It should be noted that in this embodiment, a tabu search algorithm is provided. Through the tabu search algorithm rules, all individuals in the population after selection, crossover, and mutation are corrected in different cases to obtain a new generation of population. If the fitness function value of the i-th individual in the new generation of population is less than the fitness value of the i-th individual in the previous generation of population, then save this individual; otherwise, do not accept this individual. The above-mentioned initial population adjusted and updated based on the preset tabu search algorithm to obtain the adjusted initial population may include: first, initialize the second target parameters corresponding to the preset tabu search algorithm, and the second target parameters include the length of the tabu list, the maximum number of algorithm iterations, the neighborhood function, the fitness function, and the aspiration rule; then, based on the neighborhood function, determine the neighborhood solution corresponding to the target individual in the updated initial population, and based on the fitness function, determine the target fitness value corresponding to the target individual, and based on the target fitness value, determine the candidate solution from the neighborhood solutions; finally, adjust the updated initial population based on the maximum number of algorithm iterations, the aspiration rule, and the candidate solution to obtain the adjusted initial population. Specifically, as shown in Figure 5 shown, the specific process of the tabu search algorithm is as follows:

[0093] T1: Select an individual x from the population after crossover and mutation according to the genetic algorithm, and use it as the initial current solution of the tabu search algorithm. Initialize the parameters of the tabu search algorithm, including the length L of the tabu list, the maximum number of algorithm iterations T, the neighborhood function, the fitness function, and the aspiration criterion.

[0094] T2: Determine whether the algorithm meets the termination condition. If so, output the optimized result; if not, continue with the following steps.

[0095] T3: Generate several neighborhood solutions according to the neighborhood function of the current solution, and retain the better neighborhood solutions as candidate solutions according to the fitness values of the solutions.

[0096] T4: Determine whether the candidate solution meets the aspiration criterion. If it does, lift the solution from the tabu restriction and use it as the new current optimal solution, update the current optimal state, and determine whether the termination condition is met. If it is met, end; if not, continue with step S3.

[0097] T5: If the candidate solution does not meet the aspiration criterion, select the optimal state among the candidate solutions as the new current solution, replace the tabu object that entered the tabu list earliest with the corresponding tabu object, and then determine whether the termination condition is met.

[0098] As can be seen from the above, in this embodiment, several microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model are first determined, and the microservice information corresponding to each of the microservices to be deployed, the server information corresponding to the edge servers, and the topology information corresponding to the applications to be deployed are respectively determined; then, a microservice deployment model corresponding to the government affairs large model is constructed based on the microservice information, the server information, and the topology information; finally, a preset model solving algorithm is used to solve the microservice deployment model to obtain a microservice deployment strategy corresponding to the government affairs large model, so as to deploy the several microservices to be deployed corresponding to the government affairs large model to a preset government affairs large model platform based on the microservice deployment strategy; the preset model solving algorithm is an algorithm constructed based on the genetic algorithm and the tabu search algorithm. As can be seen from the above, in this embodiment, the microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model, as well as the corresponding microservice information, server information, and topology information, are first determined, and then a microservice deployment model is constructed according to the microservice information, server information, and topology information, and finally the microservice deployment model is solved using a hybrid genetic tabu search algorithm to obtain a microservice deployment strategy. In this way, in this embodiment, the microservice computing resource requirements, the computing capabilities of edge servers, and the application topology structure dependencies are considered, and the optimal microservice deployment plan can be quickly found in a complex edge network environment through the genetic algorithm and the tabu search algorithm, effectively reducing the service response delay and improving the resource utilization efficiency and real-time decision-making ability.

[0099] SeeFigure 6 As shown in Figure 6 , in order to dynamically adjust the microservice deployment strategy and improve resource utilization, an embodiment of the present invention further discloses a microservice deployment method, which may include:

[0100] Step S21: Determine a number of microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model, and respectively determine the microservice information corresponding to each of the microservices to be deployed, the server information corresponding to the edge servers, and the topology information corresponding to the applications to be deployed.

[0101] Step S22: Construct a microservice deployment model corresponding to the government affairs large model based on the microservice information, the server information, and the topology information.

[0102] Step S23: Use a preset model solving algorithm to solve the microservice deployment model, obtain a microservice deployment strategy corresponding to the government affairs large model, and deploy a number of the microservices to be deployed corresponding to the government affairs large model to a preset government affairs large model platform based on the microservice deployment strategy; the preset model solving algorithm is an algorithm constructed based on a genetic algorithm and a tabu search algorithm.

[0103] In this embodiment, a hybrid genetic tabu search algorithm is provided, and the specific process is as follows:

[0104] S1: Initialize the genetic algorithm parameters: Set the population size to 100, the crossover probability to 0.8, the mutation probability to 0.05, the initial temperature to 100, the termination temperature to 0.1, and the maximum number of algorithm iterations to 200.

[0105] S2: Population initialization: Encode the deployment of all microservices using natural number encoding. The encoding length is the number of microservices, each position represents a microservice, and the element at each position represents the serial number of the edge server to which the corresponding microservice is deployed. In this embodiment, a specific encoding example is as follows:

[0106] When there are 5 microservices and 4 edge servers, the deployment situation of the microservices can be encoded as:

[0107] ;

[0108] Among them, The corresponding meaning is: Microservices 1 and 3 are deployed on edge server 1, microservice 2 is deployed on edge server 2, microservice 4 is deployed on edge server 3, and microservice 5 is deployed on edge server 4.

[0109] During the population initialization process, solve the greedy solution. Given that the set of all microservices is I and the set of edge servers is J, the generation scheme of the greedy solution is as follows:

[0110] Step 1: Calculate the weighted priority of microservices according to the computing speed requirements and computing resource requirements of microservices, and sort all microservices in descending order according to their priorities.

[0111] Step 2: Sort the edge servers according to the computing speed they possess.

[0112] Step 3: Assign the microservice with the highest priority to the server with the strongest computing power, and update I and J.

[0113] Step 4: If the set I is empty, the initial solution generation algorithm ends; otherwise, go to Step 3.

[0114] During the population initialization process for the remaining chromosomes, in this embodiment, among the allocable server nodes, any available server is randomly selected to generate the initial solution. The specific algorithm steps are as follows:

[0115] Step 1: Determine the set of server nodes that all microservices satisfy the resource constraints according to the computing speed, video memory resources, and memory constraints.

[0116] Step 2: Randomly select an available server node for the microservice with the smallest number in set I, and update set I.

[0117] Step 3: If the set I is empty, output the allocation result and terminate the algorithm; otherwise, go to Step 2.

[0118] S3: Fitness value calculation: Calculate the fitness value of an individual according to the objective function of the model.

[0119] S4: Selection operation: Adopt roulette wheel selection with elitist retention strategy.

[0120] S5: Crossover operation: Select any two microservices deployed on different edge servers and exchange their deployment positions.

[0121] S6: Mutation operation: Select the microservice with the longest execution time among all current applications and redeploy it randomly on a server.

[0122] S7: Perform the tabu search algorithm operation: Conduct local search according to the tabu search algorithm to find an individual with better fitness. If the fitness value of the individual is better than that of the parent individual before crossover and mutation, accept this individual. If the fitness value of the individual is worse than that of the parent individual before crossover and mutation, do not accept this individual.

[0123] S8: Accumulate the number of iterations and determine whether the maximum number of iterations is reached. If not, go to S3.

[0124] S9: Judge the fitness values of individuals in the current population and output the individual with the maximum fitness value.

[0125] It should be noted that the specific operation steps of the tabu search algorithm in step S7 are as follows:

[0126] T1: Select an individual x in the population after crossover and mutation according to the genetic algorithm, and use it as the initial current solution of the tabu search algorithm. Initialize the parameters of the tabu search algorithm, set the length of the tabu list to 3, and the maximum number of algorithm iterations to 100. The generation rule of neighborhood solutions is as follows: randomly select two microservices, and if their deployment positions in the current solution are different, swap their positions to obtain a neighborhood solution. Repeat the above steps until 10 neighborhood solutions are generated. The fitness function is the objective function of the model. The aspiration criterion is that the fitness value of the candidate solution is better than the currently found optimal solution.

[0127] T2: Judge whether the algorithm meets the termination condition. If the maximum number of iterations of the tabu search algorithm is reached, output the currently found optimal solution as the optimized result and end the algorithm. If the termination condition is not met, continue with the following steps.

[0128] T3: Generate 10 neighborhood solutions according to the neighborhood rule of the current solution, and retain 3 better neighborhood solutions as candidate solutions according to the fitness values of the solutions.

[0129] T4: Traverse the set of candidate solutions, check whether each candidate solution meets the aspiration criterion. If it meets, release the solution from the tabu restriction and use it as the new current optimal solution to update the current optimal state. Judge whether the termination condition is met. If it is met, end; if not, continue with step T3.

[0130] T5: If the candidate solution does not meet the aspiration criterion, select the optimal state among the candidate solutions as the new current solution, replace the earliest tabu object in the tabu list with the corresponding tabu object, and then judge whether the termination condition is met.

[0131] Among them, for the more specific processing procedures of the above steps S21 and S22, reference can be made to the corresponding content disclosed in the foregoing embodiments, and details will not be repeated here.

[0132] As can be seen from the above, in this embodiment, a hybrid genetic tabu search algorithm based on the greedy algorithm to generate the initial population and combined with the elite retention strategy is adopted, effectively breaking through the limitation that the traditional genetic algorithm is prone to falling into local optimum. During the iteration process, the algorithm dynamically optimizes the matching relationship between microservices and servers, and between servers through adaptive crossover and mutation operations. In this way, the present application can quickly generate a better microservice deployment decision under complex constraint conditions, significantly reduce the end-to-end delay of the service link, and improve the system robustness. At the same time, the hybrid genetic tabu search algorithm in this embodiment deeply integrates the microservice resource requirements and the dynamic supply capacity of edge servers, gives full play to the global search ability of the genetic algorithm and the local optimization ability of the tabu search, and realizes global and local collaborative optimization. When allocating resources, the hybrid genetic tabu search algorithm can dynamically balance the computing load, effectively avoid the occurrence of resource overload or idleness, and then significantly improve the resource utilization rate and load balancing degree of server nodes. In this way, the real-time performance of microservice response is significantly improved, and the overall service efficiency of the edge network is also enhanced.

[0133] In a specific implementation manner, a microservice deployment example of a government affairs large model platform is taken as an example for illustration. The government affairs large model platform consists of model training and optimization services, intelligent question answering services, text generation services, data analysis and prediction services, security authentication and encryption services, user management and permission services, interface adaptation and integration services, and model deployment and management services. The information of each microservice is shown in Table 1. The number of edge servers is 5, and the computing speed, storage capacity, and number of cores of the edge servers are shown in Table 2. The data transmission rate between any two edge servers is shown in Table 3. In this embodiment, as shown in Figure 7 shown, the average arrival rates of applications (a)-(c) are set to 1, 2, and 1 respectively. The traffic on any arc in the topology structure of each application is randomly generated between 5GB and 50GB, and the computing volume of any microservice is expressed in trillions of floating-point operations per second and randomly generated between 0.1 and 1.

[0134] Table 1

[0135]

[0136] Table 2

[0137]

[0138] Table 3

[0139]

[0140] As shown in Figure 8 shown, in a specific implementation manner, the microservice deployment process is as follows:

[0141] Step 1: Determine all the deployed microservices in this scheduling to form a set I, and obtain the detailed information of each microservice in the microservice set I;

[0142] Step 2: Determine all available edge servers in this scheduling to form a set J, and obtain the detailed information of each server in the set J;

[0143] Step 3: Confirm all the applications participating in this scheduling to form a set L, and obtain the topological information of all the application structures in the set L;

[0144] Step 4: Set decision variables according to the basic information of microservices, servers and applications, and construct a microservice deployment model.

[0145] Step 5: Solve the established microservice deployment model by using a hybrid genetic tabu search algorithm. Solve the microservice deployment model according to the topological structure information of the microservice set I, the edge server set J and the application set L to determine the microservice deployment result.

[0146] Correspondingly, as shown in Figure 9 This embodiment of the present application also provides a microservice deployment device, which may include:

[0147] An information determination module 11, configured to determine a number of microservices to be deployed, edge servers and applications to be deployed corresponding to the government affairs large model, and respectively determine the microservice information corresponding to each of the microservices to be deployed, the server information corresponding to the edge servers, and the topological information corresponding to the applications to be deployed;

[0148] A model construction module 12, configured to construct a microservice deployment model corresponding to the government affairs large model based on the microservice information, the server information and the topological information;

[0149] A microservice deployment module 13, configured to solve the microservice deployment model by using a preset model solving algorithm to obtain a microservice deployment strategy corresponding to the government affairs large model, so as to deploy a number of the microservices to be deployed corresponding to the government affairs large model to a preset government affairs large model platform based on the microservice deployment strategy; the preset model solving algorithm is an algorithm constructed based on a genetic algorithm and a tabu search algorithm.

[0150] As can be seen from the above, in this application, several microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model are first determined, and the microservice information corresponding to each of the microservices to be deployed, the server information corresponding to the edge servers, and the topology information corresponding to the applications to be deployed are respectively determined; then, a microservice deployment model corresponding to the government affairs large model is constructed based on the microservice information, the server information, and the topology information; finally, a preset model solving algorithm is used to solve the microservice deployment model to obtain a microservice deployment strategy corresponding to the government affairs large model, so as to deploy several microservices to be deployed corresponding to the government affairs large model to a preset government affairs large model platform based on the microservice deployment strategy; the preset model solving algorithm is an algorithm constructed based on a genetic algorithm and a tabu search algorithm. As can be seen from the above, in this application, the microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model, as well as the corresponding microservice information, server information, and topology information, are first determined, and then a microservice deployment model is constructed according to the microservice information, server information, and topology information, and finally the hybrid genetic tabu search algorithm is used to solve the microservice deployment model to obtain a microservice deployment strategy. In this way, in this application, the computing resource requirements of microservices, the computing capabilities of edge servers, and the dependencies of application topologies are considered, and the optimal microservice deployment plan can be quickly found in a complex edge network environment through a genetic algorithm and a tabu search algorithm, effectively reducing the service response delay and improving the resource utilization efficiency and real-time decision-making ability.

[0151] In some specific embodiments, the information determination module 11 may include:

[0152] A parameter slicing unit, configured to determine a target parameter slicing rule based on the service type of the government affairs large model, and slice the parameters of the government affairs large model based on a preset parameter slicing technique and the target parameter slicing rule;

[0153] An inference service unit determination unit, configured to slice the functions of the government affairs large model based on a preset function decoupling technique, and combine the sliced parameters and the sliced functions to obtain inference service units corresponding to the respective functions of the government affairs large model;

[0154] A microservice to be deployed determination unit, configured to determine the microservices to be deployed according to the inference service units, and organize and coordinate the microservices to be deployed based on a preset communication protocol, so that the microservices to be deployed can perform data exchange through the preset communication protocol.

[0155] In some specific embodiments, the information determination module 11 may include:

[0156] A core quantity determination unit, configured to determine the target quantity of the microservices to be deployed, and determine the core quantity of the edge server based on the target quantity, where the target quantity and the core quantity are in one-to-one correspondence;

[0157] A resource determination unit, configured to determine the computing resources of the edge server, and determine the storage resources of the edge server based on the memory capacity and memory bandwidth of the edge server;

[0158] A real-time status determination unit, configured to determine the communication capability of the edge server based on the inter-node bandwidth and cross-domain transmission delay of the edge server, and determine the real-time status of the edge server based on the current load rate and available resource capacity of the edge server.

[0159] In some specific embodiments, the information determination module 11 may include:

[0160] A topology structure determination unit, configured to determine the target microservices to be deployed traversed during a single run of the application to be deployed, and determine the topology structure corresponding to the application to be deployed based on a preset directed acyclic graph structure and the target microservices to be deployed;

[0161] A data transmission volume determination unit, configured to determine the computing volume corresponding to the target microservices to be deployed during a single run of the application to be deployed, and determine the data transmission volume between the target microservices to be deployed;

[0162] An average arrival rate determination unit, configured to determine the average arrival rate of the target requests corresponding to the application to be deployed, and execute the target requests based on a preset first-come-first-served principle.

[0163] In some specific embodiments, the model construction module 12 may include:

[0164] A target function determination unit, configured to determine the constraint conditions corresponding to the microservice deployment model based on the microservice information and the server information, and construct the target function corresponding to the microservice deployment model based on the topology information;

[0165] A model construction unit, configured to set the decision variables corresponding to the microservice deployment model, and construct the microservice deployment model corresponding to the e-government large model based on the constraint conditions, the target function, and the decision variables; the decision variables are used to determine the deployment status of the microservices to be deployed and the edge server.

[0166] In some specific embodiments, the microservice deployment module 13 may include:

[0167] An initial population generation sub-module, configured to initialize first target parameters corresponding to a preset hybrid genetic tabu search algorithm, encode individuals in the population based on the first target parameters, and generate an initial population based on a preset greedy algorithm and the encoded individuals; the individuals are used to represent microservice deployment strategies.

[0168] An initial population update sub-module, configured to determine fitness values of the individuals based on an objective function corresponding to the microservice deployment model, and update the initial population based on a preset selection operation, a preset crossover operation, a preset mutation operation, and the fitness values.

[0169] A population adjustment sub-module, configured to adjust the updated initial population based on a preset tabu search algorithm to obtain the adjusted initial population, and determine a target population based on a preset iteration number threshold, the adjusted initial population, and the fitness values, so as to obtain the microservice deployment strategy corresponding to the government affairs large model.

[0170] In some specific embodiments, the population adjustment sub-module may include:

[0171] A parameter initialization unit, configured to initialize second target parameters corresponding to the preset tabu search algorithm, where the second target parameters include the length of the tabu list, the maximum number of algorithm iterations, a neighborhood function, a fitness function, and a aspiration rule.

[0172] A candidate solution determination unit, configured to determine neighborhood solutions corresponding to a target individual in the updated initial population based on the neighborhood function, determine a target fitness value corresponding to the target individual based on the fitness function, and determine candidate solutions from the neighborhood solutions based on the target fitness value.

[0173] A population adjustment unit, configured to adjust the updated initial population based on the maximum number of algorithm iterations, the aspiration rule, and the candidate solutions to obtain the adjusted initial population.

[0174] Furthermore, an embodiment of the present application also discloses an electronic device. Figure 10 It is a structural diagram of an electronic device 20 shown according to an exemplary embodiment, and the content in the figure should not be considered as any limitation to the scope of use of the present application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. Among them, the memory 22 is used to store a computer program, and the computer program is loaded and executed by the processor 21 to implement the relevant steps in the microservice deployment method disclosed in any of the foregoing embodiments. Additionally, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0175] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and no specific limitation is imposed on it here; the input / output interface 25 is used to obtain external input data or output data to the outside, and its specific interface type can be selected according to specific application requirements, and no specific limitation is made here.

[0176] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, a random access memory, a magnetic disk, an optical disk, etc. The resources stored thereon can include an operating system 221, a computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0177] Among them, the operating system 221 is used to manage and control each hardware device and the computer program 222 on the electronic device 20, and it can be Windows Server, Netware, Unix, Linux, etc. In addition to the computer program that can be used to complete the microservice deployment method executed by the electronic device 20 disclosed in any of the foregoing embodiments, the computer program 222 can further include computer programs that can be used to complete other specific tasks.

[0178] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the microservice deployment method disclosed above is implemented. For the specific steps of this method, reference can be made to the corresponding content disclosed in the foregoing embodiments, and details will not be repeated here.

[0179] In this specification, each embodiment is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0180] Those skilled in the art can further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0181] The steps of the methods or algorithms described in connection with the embodiments disclosed herein may be implemented directly in hardware, in software modules executed by a processor, or in a combination thereof. The software modules may be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the art.

[0182] Finally, it should also be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0183] The technical solutions provided in this application have been introduced in detail above. Specific examples are used in this document to illustrate the principles and implementation manners of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A microservice deployment method, characterized in that, Including: Determine a number of microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model, and respectively determine the microservice information corresponding to each of the microservices to be deployed, the server information corresponding to the edge servers, and the topology information corresponding to the applications to be deployed; Construct a microservice deployment model corresponding to the government affairs large model based on the microservice information, the server information, and the topology information; Use a preset model solving algorithm to solve the microservice deployment model, and obtain a microservice deployment strategy corresponding to the government affairs large model, so as to deploy a number of the microservices to be deployed corresponding to the government affairs large model to a preset government affairs large model platform based on the microservice deployment strategy; The preset model solving algorithm is an algorithm constructed based on a genetic algorithm and a tabu search algorithm.

2. The microservice deployment method according to claim 1, wherein The determining a number of microservices to be deployed corresponding to the government affairs large model includes: Determine a target parameter slicing rule based on the business type of the government affairs large model, and slice the parameters of the government affairs large model based on a preset parameter slicing technique and the target parameter slicing rule; Slice the functions of the government affairs large model based on a preset function decoupling technique, and combine the sliced parameters and the sliced functions to obtain inference service units corresponding to each function of the government affairs large model; Determine the microservices to be deployed according to the inference service units, and organize and coordinate the microservices to be deployed based on a preset communication protocol, so that the microservices to be deployed can perform data exchange through the preset communication protocol.

3. The microservice deployment method according to claim 1, wherein The respectively determining the microservice information corresponding to each of the microservices to be deployed and the server information corresponding to the edge servers includes: Determine the target quantity of the microservices to be deployed, and determine the core quantity of the edge servers based on the target quantity, where the target quantity and the core quantity are in one-to-one correspondence; Determine the computing resources of the edge servers, and determine the storage resources of the edge servers based on the memory capacity and memory bandwidth of the edge servers; Determine the communication capabilities of the edge servers based on the inter-node bandwidth and cross-domain transmission delay of the edge servers, and determine the real-time status of the edge servers based on the current load rate and available resource capacity of the edge servers.

4. The microservice deployment method according to claim 1, wherein Determine the topology information corresponding to the applications to be deployed, including: Determine the target microservices to be deployed traversed during a single run of the application to be deployed, and determine the topology structure corresponding to the application to be deployed based on a preset directed acyclic graph structure and the target microservices to be deployed; Determine the amount of computation corresponding to the target microservices to be deployed during a single run of the application to be deployed, and determine the amount of data transmission between the target microservices to be deployed; Determine the average arrival rate of the target requests corresponding to the application to be deployed, and execute the target requests based on a preset first-come-first-served principle.

5. The microservice deployment method according to claim 1, wherein The constructing a microservice deployment model corresponding to the government affairs large model based on the microservice information, the server information, and the topology information includes: Determine the constraint conditions corresponding to the microservice deployment model based on the microservice information and the server information, and construct the objective function corresponding to the microservice deployment model based on the topology information; Set the decision variables corresponding to the microservice deployment model, and construct the microservice deployment model corresponding to the government affairs large model based on the constraint conditions, the objective function, and the decision variables; the decision variables are used to determine the deployment status of the microservices to be deployed and the edge servers.

6. The microservice deployment method according to any one of claims 1 to 5, characterized in that, The solving the microservice deployment model by using a preset model solving algorithm includes: Initialize the first target parameters corresponding to the preset hybrid genetic tabu search algorithm, encode the individuals in the population based on the first target parameters, and generate an initial population based on the preset greedy algorithm and the encoded individuals; the individuals are used to represent microservice deployment strategies. Determine the fitness value of the individuals based on the objective function corresponding to the microservice deployment model, and update the initial population based on the preset selection operation, preset crossover operation, preset mutation operation, and the fitness value. Adjust the updated initial population based on the preset tabu search algorithm to obtain the adjusted initial population, and determine the target population based on the preset iteration number threshold, the adjusted initial population, and the fitness value to obtain the microservice deployment strategy corresponding to the government affairs large model.

7. The microservice deployment method according to claim 6, characterized in that The adjusting the updated initial population based on the preset tabu search algorithm to obtain the adjusted initial population includes: Initialize the second target parameters corresponding to the preset tabu search algorithm, where the second target parameters include the length of the tabu list, the maximum number of algorithm iterations, the neighborhood function, the fitness function, and the aspiration rule. Determine the neighborhood solutions corresponding to the target individuals in the updated initial population based on the neighborhood function, determine the target fitness value corresponding to the target individuals based on the fitness function, and determine the candidate solutions from the neighborhood solutions based on the target fitness value. Adjust the updated initial population based on the maximum number of algorithm iterations, the aspiration rule, and the candidate solutions to obtain the adjusted initial population.

8. A microservice deployment device, characterized in that, including: An information determination module, configured to determine several microservices to be deployed, edge servers, and applications to be deployed corresponding to the government affairs large model, and respectively determine the microservice information corresponding to each microservice to be deployed, the server information corresponding to the edge servers, and the topology information corresponding to the applications to be deployed. A model construction module, configured to construct a microservice deployment model corresponding to the government affairs large model based on the microservice information, the server information, and the topology information. A microservice deployment module, configured to solve the microservice deployment model by using a preset model solving algorithm to obtain a microservice deployment strategy corresponding to the government affairs large model, and deploy several microservices to be deployed corresponding to the government affairs large model to a preset government affairs large model platform based on the microservice deployment strategy. The preset model solving algorithm is an algorithm constructed based on a genetic algorithm and a tabu search algorithm.

9. An electronic device, characterized in that, The electronic device includes a processor and a memory; wherein, the memory is used to store a computer program, and the computer program is loaded and executed by the processor to implement the microservice deployment method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, For storing a computer program, which, when executed by a processor, implements the microservice deployment method according to any one of claims 1 to 7.

Images