An industrial production inter-microservice communication path optimization method, medium and system

Abstract

The application provides an industrial production micro-service intercommunication path optimization method, medium and system, belongs to the technical field of electronic digital data processing, and comprises the following steps: collecting basic communication data between micro-service nodes in real time, generating a micro-service node basic data set containing multiple dimensions of indexes; establishing a micro-service node similarity calculation function and a micro-service node similarity threshold, and calculating a node similarity value based on the micro-service node basic data set; constructing a micro-service node communication path optimization double-layer model containing an outer optimization model and an inner optimization model; iteratively optimizing until reaching an upper limit of the number of micro-service node iteration optimization times, and obtaining a final micro-service node communication topology graph containing output results of the main communication path optimization equation set and the local communication path optimization equation set, so as to optimize the industrial production micro-service intercommunication path. The technical problem of insufficient optimization of the micro-service intercommunication path in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital data processing, in particular to an industrial production micro-service intercommunication path optimization method, medium and system. BACKGROUND

[0002] With the rapid development of intelligentization and automation of industrial production, industrial control systems based on micro-service architecture are increasingly valued. This architecture divides the entire production process into multiple independent micro-service modules, and each micro-service node communicates and cooperates with other nodes through a network in a loosely coupled manner, which can improve the flexibility, scalability and maintainability of the system. However, in the industrial production environment, there are numerous micro-service nodes, which are widely distributed and have limited resources. Therefore, how to effectively optimize the communication path between nodes has become a technical problem to be solved.

[0003] Currently, the commonly used micro-service communication path optimization methods in the industry mainly include: dynamic routing strategy based on load balancing, communication link selection based on quality of service, intelligent decision model based on machine learning, etc. These methods have improved the communication efficiency between micro-service nodes to some extent, but still have some problems:

[0004] 1. Load balancing strategy usually only focuses on the resource utilization of the node itself, ignoring the dynamic coupling communication characteristics between nodes, and it is difficult to achieve global optimization;

[0005] 2. The link selection oriented to service quality is too simple, and it is difficult to accurately describe the performance index change rule in complex communication environment;

[0006] 3. Although the intelligent decision model based on machine learning can adapt to dynamic changes, it needs a large amount of historical data to support, and it is difficult to explain the optimization mechanism inside the model.

[0007] Therefore, the existing technology has certain limitations in the optimization of industrial production micro-service communication path, and it is difficult to fully exert the advantages of micro-service architecture, which restricts the further development of intelligent factory.

[0008] Therefore, the existing technology has certain limitations in the optimization of industrial production micro-service communication path, and it is difficult to fully exert the advantages of micro-service architecture, which restricts the further development of intelligent factory. Therefore, there is an urgent need for a new communication path optimization method to solve the technical problem of insufficient optimization of micro-service intercommunication path in the existing technology. SUMMARY

[0009] In view of this, the present application provides an industrial production micro-service intercommunication path optimization method, medium and system, which solves the technical problem of insufficient optimization of micro-service intercommunication path in the prior art. The present application is implemented as follows: the first aspect of the present application provides an industrial production micro-service intercommunication path optimization method, comprising the following steps: collecting basic communication data between micro-service nodes in real time to generate a micro-service node basic data set containing multi-dimensional indicators; performing stable fluctuation decomposition calculation on the micro-service node communication delay data and bandwidth data to obtain corresponding stable components and fluctuation components; establishing a micro-service node similarity calculation function and a micro-service node similarity threshold, and calculating a node similarity value based on the micro-service node basic data set; constructing a micro-service node communication path optimization double-layer model containing an outer optimization model and an inner optimization model; establishing a main communication path optimization equation set and a local communication path optimization equation set; performing iterative optimization of the double-layer model and synchronously optimizing the scheme when the optimization conditions are met; repeatedly performing optimization until the upper limit of the number of micro-service node iterative optimizations is reached, and obtaining a final micro-service node communication topology graph containing the output results of the main communication path optimization equation set and the local communication path optimization equation set.

[0010] The micro-service node basic data set includes micro-service node communication delay data, micro-service node bandwidth data, micro-service node processing capability data, micro-service node load rate data, micro-service node data transmission volume data, micro-service node historical failure rate data, micro-service node concurrent processing capability data, micro-service node network congestion degree data, micro-service node hop weight data, micro-service node geographic distance data, micro-service node communication protocol compatibility data, micro-service node service quality level data, micro-service node fault tolerance capability data, micro-service node load balancing factor data, micro-service node resource utilization rate data, micro-service node energy consumption coefficient data, and micro-service node maintenance cost data.

[0011] The stable fluctuation decomposition calculation on the micro-service node communication delay data obtains micro-service node communication delay stable components and micro-service node communication delay fluctuation components, and the stable fluctuation decomposition calculation on the micro-service node bandwidth data obtains micro-service node bandwidth stable components and micro-service node bandwidth fluctuation components.

[0012] The micro-service node communication path outer optimization model is used for global communication path optimization, and the micro-service node communication path inner optimization model is used for local communication path optimization.

[0013] The main communication path optimization equation set includes a micro-service node type judgment equation, a micro-service node path distance equation, a micro-service node efficiency constraint equation, and a micro-service node total cost equation.

[0014] The micro-service node type determination equation is used for determining whether the micro-service node belongs to a first-level transmission micro-service node or a second-level transmission micro-service node; the micro-service node path distance equation is used for calculating the shortest distance of a communication path between the micro-service nodes; the micro-service node efficiency constraint equation is used for limiting the efficiency threshold of the communication path between the micro-service nodes; and the micro-service node total cost equation is used for calculating the overall resource consumption of the communication path of the micro-service node.

[0015] The local communication path optimization equation set includes a micro-service node power balance equation, a micro-service node bandwidth balance equation and a micro-service node load balance equation.

[0016] The micro-service node power balance equation is used for balancing the calculation processing power distribution of the micro-service node; the micro-service node bandwidth balance equation is used for balancing the bandwidth resource distribution of the micro-service node; and the micro-service node load balance equation is used for balancing the task load distribution of the micro-service node.

[0017] The second aspect of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores program instructions, and the program instructions are used for executing the above-mentioned industrial production micro-service intercommunication path optimization method when running in a computer.

[0018] The third aspect of the present application provides an industrial production micro-service intercommunication path optimization method, which includes the above-mentioned computer readable storage medium.

[0019] The technical effect of the present application is that the present application realizes comprehensive description and dynamic optimization of micro-service node communication characteristics by using real-time monitoring, stable fluctuation decomposition, similarity calculation, double-layer optimization modeling and other technical means.

[0020] Firstly, key communication data between micro-service nodes are collected in real time, including communication delay, bandwidth, processing capacity, load rate and the like, and a basic data set of the micro-service node is constructed. Then, the Hilbert-Huang transform algorithm is used for stable fluctuation decomposition of the communication delay and bandwidth data, and the stable components and fluctuation components thereof are obtained. This lays a foundation for subsequent similarity calculation and optimization modeling.

[0021] Secondly, a micro-service node similarity calculation function is established, which can quantitatively describe the similarity degree between nodes. This provides a basis for subsequent local optimization. Subsequently, a double-layer micro-service node communication path optimization model is constructed. The outer model is responsible for global communication path optimization, including node type determination, shortest path distance calculation, communication efficiency constraint and overall resource consumption and the like; and the inner model is responsible for local communication path optimization, including power balance, bandwidth balance and load balance and the like.

[0022] Finally, through multiple rounds of iterative optimization, the optimized micro-service node communication topology graph is finally obtained. The topology graph can provide decision support for industrial production systems, significantly improving the overall communication performance and resource utilization efficiency.

[0023] Compared with the prior art, the scheme has the following advantages:

[0024] 1. Real-time monitoring and stable fluctuation decomposition are used to comprehensively characterize the dynamic communication characteristics of micro-service nodes, providing more accurate data support for subsequent optimization decisions;

[0025] 2. Similarity calculation is introduced to identify the correlation between nodes, enabling the propagation and sharing of local optimization results among similar nodes, improving the overall optimization effect;

[0026] 3. A double-layer optimization model is constructed, considering both the overall communication efficiency and the balance of resources between nodes, reflecting the coordinated optimization of the global and local;

[0027] 4. Using interpretable mathematical modeling methods such as partial differential equations and variational methods, not only can we get the optimization results, but also can we understand the internal optimization mechanism, with strong interpretability.

[0028] In summary, the scheme makes full use of the communication characteristics of micro-service nodes in industrial production environments, through real-time monitoring, dynamic analysis and coordinated optimization, effectively improving the communication performance and resource utilization efficiency of the entire production system, providing technical support for the development of intelligent factories. It solves the technical problem of insufficient optimization of communication paths between micro-services in the prior art. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 is a flowchart of the specific embodiments of the present application;

[0030] Figure 2 is a time series variation graph of the basic monitoring data in Example 2, including two subgraphs: the upper subgraph is a communication delay fluctuation graph over time, and the lower subgraph is a bandwidth B variation graph over time;

[0031] Figure 3 is a result graph of HHT decomposition of communication delay data in Example 2, including four subgraphs from top to bottom: high-frequency IMF component graph, medium-frequency IMF component graph, low-frequency IMF component graph and overall trend item graph;

[0032] Figure 4 is a node similarity heat map in Example 2;

[0033] Figure 5 is an optimized micro-service communication topology graph in Example 2. DETAILED DESCRIPTION

[0034] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0035] The first aspect of the present application provides an industrial production micro-service intercommunication path optimization method, comprising the following steps:

[0036] S10, collecting basic communication data between micro-service nodes in real time to generate a micro-service node basic data set, the micro-service node basic data set including micro-service node communication delay data, micro-service node bandwidth data, micro-service node processing capability data, micro-service node load rate data, micro-service node data transmission volume data, micro-service node historical failure rate data, micro-service node concurrent processing capability data, micro-service node network congestion degree data, micro-service node hop count weight data, micro-service node geographic distance data, micro-service node communication protocol compatibility data, micro-service node service quality level data, micro-service node fault tolerance capability data, micro-service node load balancing factor data, micro-service node resource utilization rate data, micro-service node energy consumption coefficient data, and micro-service node maintenance cost data;

[0037] S20, performing stable fluctuation decomposition calculation on the micro-service node communication delay data to obtain micro-service node communication delay stable components and micro-service node communication delay fluctuation components;

[0038] S30, performing stable fluctuation decomposition calculation on the micro-service node bandwidth data to obtain micro-service node bandwidth stable components and micro-service node bandwidth fluctuation components;

[0039] S40, establishing a micro-service node similarity calculation function and a micro-service node similarity threshold value, and calculating micro-service node similarity values between each two micro-service nodes based on the micro-service node basic data set;

[0040] S50, constructing a micro-service node communication path optimization double-layer model, the micro-service node communication path optimization double-layer model including a micro-service node communication path outer-layer optimization model and a micro-service node communication path inner-layer optimization model, the micro-service node communication path outer-layer optimization model being used for global communication path optimization, and the micro-service node communication path inner-layer optimization model being used for local communication path optimization;

[0041] S60, establishing a main communication path optimization equation set of the micro-service node communication path outer-layer optimization model, the main communication path optimization equation set including a micro-service node type judgment equation, a micro-service node path distance equation, a micro-service node efficiency constraint equation, and a micro-service node total cost equation;

[0042] S70, establish a local communication path optimization equation set of the micro-service node communication path inner layer optimization model, the local communication path optimization equation set includes a micro-service node power balance equation, a micro-service node bandwidth balance equation, and a micro-service node load balancing equation, and set an upper limit of the micro-service node iteration optimization times;

[0043] S80, perform iteration optimization of the micro-service node communication path outer layer optimization model to obtain a micro-service node global communication path initial scheme, perform iteration optimization of the micro-service node communication path inner layer optimization model by taking the micro-service node global communication path initial scheme as input, generate a micro-service node local communication path optimization scheme, and when the micro-service node load distribution value of a certain micro-service node is better than an existing micro-service node load distribution value, synchronize the micro-service node local communication path optimization scheme to other micro-service nodes with a micro-service node similarity value greater than a micro-service node similarity threshold value.

[0044] S90, repeat step S80 until the upper limit of the micro-service node iteration optimization times is reached to obtain a final micro-service node communication topology graph, and the final micro-service node communication topology graph contains output results of the main communication path optimization equation set and the local communication path optimization equation set.

[0045] The main communication path optimization equation set includes a micro-service node type determination equation, a micro-service node path distance equation, a micro-service node efficiency constraint equation, and a micro-service node total cost equation.

[0046] The micro-service node type determination equation is used to determine whether the micro-service node belongs to a first-level transmission micro-service node or a second-level transmission micro-service node, the input includes micro-service node load rate data, micro-service node processing capacity data, micro-service node historical failure rate data, micro-service node concurrent processing capacity data, and micro-service node network congestion data, and the output is a micro-service node type identification value.

[0047] The micro-service node path distance equation is used to calculate the shortest distance of the communication path between the micro-service nodes, the input includes micro-service node communication delay stable component data, micro-service node bandwidth stable component data, micro-service node hop count weight data, micro-service node geographic distance data, and micro-service node communication protocol compatibility data, and the output is a micro-service node shortest communication path value.

[0048] The microservice node efficiency constraint equation is used to define the efficiency threshold of the communication path between the microservice nodes, and the input includes the microservice node communication delay fluctuation component, the microservice node bandwidth fluctuation component, the microservice node quality of service level data, the microservice node fault tolerance capability data, and the microservice node load balancing factor data, and the output is a microservice node communication efficiency constraint value;

[0049] The microservice node total cost equation is used to calculate the overall resource consumption of the communication path of the microservice node, and the input includes the microservice node type identification value, the microservice node shortest communication path value, the microservice node communication efficiency constraint value, the microservice node resource utilization data, the microservice node energy consumption coefficient data, and the microservice node maintenance cost data, and the output is a microservice node communication path total cost value;

[0050] The local communication path optimization equation set includes a microservice node power balance equation, a microservice node bandwidth balance equation, and a microservice node load balancing equation.

[0051] The microservice node power balance equation is used to balance the calculation processing power distribution of the microservice node, and the input includes the microservice node processing capability data, the microservice node concurrent processing capability data, the microservice node load rate data, the microservice node resource utilization data, and the microservice node energy consumption coefficient data, and the output is a microservice node power distribution matrix.

[0052] The microservice node bandwidth balance equation is used to balance the bandwidth resource distribution of the microservice node, and the input includes the microservice node bandwidth stable component, the microservice node bandwidth fluctuation component, the microservice node data transmission amount data, the microservice node network congestion degree data, and the microservice node communication protocol compatibility data, and the output is a microservice node bandwidth allocation vector.

[0053] The microservice node load balancing equation is used to balance the task load distribution of the microservice node, and the input includes the microservice node power distribution matrix, the microservice node bandwidth allocation vector, the microservice node fault tolerance capability data, the microservice node quality of service level data, and the microservice node load balancing factor data, and the output is a microservice node load distribution value.

[0054] The specific implementation of step S10 is: real-time collection of micro-service node basic communication data using a distributed sensor network, including communication delay data collection at a frequency of 100 times per second, bandwidth data collection at a frequency of 50 times per second, processing capability data collection at a frequency of 20 times per second, load rate data collection at a frequency of 10 times per second, data transmission volume collection at a frequency of 5 times per second, historical failure rate data collection at a frequency of 1 time per minute, concurrent processing capability data collection at a frequency of 2 times per second, network congestion degree data collection at a frequency of 1 time per second, hop weight data collection at a frequency of 1 time per hour, geographic distance data collection at a frequency of 1 time per day, communication protocol compatibility data collection at a frequency of 1 time per hour, service quality level data collection at a frequency of 1 time per hour, fault tolerance capability data collection at a frequency of 1 time per hour, load balancing factor data collection at a frequency of 1 time per minute, resource utilization rate data collection at a frequency of 1 time per second, energy consumption coefficient data collection at a frequency of 1 time per hour, and maintenance cost data collection at a frequency of 1 time per day. The collected data is stored in a distributed database, a time series database is used to store and manage the data, and a data preprocessing technique is used to filter abnormal data during data collection. The filtering rules include data range checking, data type checking, and data format checking, to ensure the accuracy and reliability of the collected data.

[0055] The specific implementation of step S20 is: stable fluctuation decomposition calculation of the collected micro-service node communication delay data. First, the Hilbert-Huang transform algorithm is used to decompose the communication delay time series, and the original communication delay time series is decomposed into the superposition form of stable components and fluctuation components. The stable component represents the long-term trend of the communication delay, and the fluctuation component represents the short-term fluctuation of the communication delay. In the decomposition process, the empirical mode decomposition technique is used to decompose the time series. Each intrinsic mode function component obtained by decomposition represents different scale fluctuation characteristics. By analyzing each intrinsic mode function component, the stable component and the fluctuation component of the communication delay can be obtained. The threshold of the stable component is set to 80% of the average communication delay, and the threshold of the fluctuation component is set to 20% of the average communication delay.

[0056] The specific implementation of step S30 is: stable fluctuation decomposition calculation of the collected micro-service node bandwidth data. Similarly, the Hilbert-Huang transform algorithm is used to decompose the original bandwidth time series into the form of stable components and fluctuation components. The stable component represents the long-term trend of the bandwidth, and the fluctuation component represents the short-term fluctuation of the bandwidth. In the decomposition process, the empirical mode decomposition technique is used to decompose the time series. Each intrinsic mode function component obtained by decomposition represents different scale fluctuation characteristics. By analyzing each intrinsic mode function component, the stable component and the fluctuation component of the bandwidth can be obtained. The threshold of the stable component is set to 85% of the average bandwidth, and the threshold of the fluctuation component is set to 15% of the average bandwidth.

[0057] The specific implementation of step S40 is to establish a micro-service node similarity calculation function, adopt a kernel function-based similarity calculation method, map the feature vectors of each micro-service node to a high-dimensional feature space for similarity calculation, the feature vectors include communication delay, bandwidth, processing capacity, and load rate, etc. parameters, adopt a Gaussian kernel function as the kernel function, the bandwidth parameter of the kernel function is determined by the cross-validation method, the time-varying weight function is considered in the similarity calculation process, the parameters of the weight function are determined by the least squares method, and the similarity threshold is set to 0.8. When the similarity of two micro-service nodes is greater than the threshold, they are considered to be similar nodes.

[0058] The specific implementation of step S50 is to construct a double-layer model for micro-service node communication path optimization. The outer optimization model adopts a hierarchical optimization strategy to decompose the global communication path optimization problem into multiple sub-problems, each sub-problem corresponds to an optimization objective, including node type determination, path distance calculation, efficiency constraint, and total cost calculation. The inner optimization model adopts a local optimization strategy to decompose the local communication path optimization problem into three sub-problems: power balance, bandwidth balance, and load balance. The solution of the double-layer optimization model adopts an iterative optimization method. The solution of the outer optimization model adopts a genetic algorithm, and the solution of the inner optimization model adopts a particle swarm optimization algorithm.

[0059] The specific implementation of step S60 is to establish a main communication path optimization equation set in the outer optimization model, including a micro-service node type determination equation, a micro-service node path distance equation, a micro-service node efficiency constraint equation, and a micro-service node total cost equation. The micro-service node type determination equation adopts a deep neural network model, the input features include load rate, processing capacity, historical failure rate, concurrent processing capacity, and network congestion degree, and the output is a node type identifier. The weight matrix of the neural network is obtained by training through the back propagation algorithm. The micro-service node path distance equation adopts a shortest path algorithm, the input includes communication delay stable component, bandwidth stable component, hop weight, geographic distance, and communication protocol compatibility, and the output is the shortest communication path value between nodes. The micro-service node efficiency constraint equation adopts a nonlinear constraint optimization method, the input includes communication delay fluctuation component, bandwidth fluctuation component, service quality level, fault tolerance capability, and load balancing factor, and the output is a communication efficiency constraint value. The micro-service node total cost equation adopts a multi-objective optimization method, the input includes node type, shortest path distance, efficiency constraint value, resource utilization rate, energy consumption coefficient, and maintenance cost, and the output is a communication path total cost value.

[0060] The specific implementation of step S70 is to establish a local communication path optimization equation set in the inner layer optimization model, including a microservice node power balance equation, a microservice node bandwidth balance equation, and a microservice node load balancing equation, wherein the microservice node power balance equation adopts a matrix differential equation solving method, the inputs include processing capacity, concurrent processing capacity, load rate, resource utilization rate, and energy consumption coefficient, and the output is a power distribution matrix, the microservice node bandwidth balance equation adopts a vector differential equation solving method, the inputs include bandwidth stable component, bandwidth fluctuation component, data transmission volume, network congestion degree, and communication protocol compatibility, and the output is a bandwidth distribution vector, the microservice node load balancing equation adopts a nonlinear integral differential equation solving method, the inputs include the power distribution matrix, the bandwidth distribution vector, fault tolerance capability, service quality level, and load balancing factor, and the output is a load distribution value, and the upper limit of the iteration optimization times is set to 1000 times.

[0061] The specific implementation of step S80 is to perform iteration optimization of the outer layer optimization model, first, a genetic algorithm is used to solve the outer layer optimization model, the population size is set to 100, the crossover probability is set to 0.8, the mutation probability is set to 0.1, the iteration times are set to 500 times, and an initial scheme of the global communication path of the microservice node is obtained, then the initial scheme is taken as input to perform iteration optimization of the inner layer optimization model, a particle swarm optimization algorithm is used to solve the inner layer optimization model, the particle swarm size is set to 50, the inertia weight is set to 0.7, the learning factor is set to 2.0, the iteration times are set to 200 times, a local communication path optimization scheme is generated, and when the load distribution value of a certain node is better than the existing load distribution value, the local optimization scheme is synchronized to other nodes with a similarity greater than a threshold.

[0062] The specific implementation of step S90 is to repeatedly perform step S80 until the upper limit of the microservice node iteration optimization times is reached, and finally an optimal microservice node communication topology graph is obtained, which contains the output results of the outer layer optimization model and the inner layer optimization model, the generation of the communication topology graph adopts a graph layout algorithm, the connection relationship between nodes is determined according to the communication path optimization result, the position of the node is determined according to the geographical distance and the communication delay, the size of the node is determined according to the processing capacity and the load rate, the color of the node is determined according to the service quality level, the thickness of the connection line is determined according to the bandwidth distribution, and the color of the connection line is determined according to the communication efficiency.

[0063] The parameter acquisition and equation design principle of the whole optimization method are described as follows. The system parameters can be obtained by least square fitting of historical data, the weight matrix can be obtained by deep learning method training, the error term range can be determined by Monte Carlo simulation, the kernel function parameters can be determined by cross validation, the control parameters can be obtained by optimal control theory calculation, the equation design adopts partial differential equation to describe the system dynamic characteristics, uses integral term to describe the influence of historical information, uses nonlinear term to express the complex interaction relationship, introduces Laplace operator to process spatial correlation, and combines variational method to solve the optimization problem.

[0064] The equations or calculation processes involved in the present application are described in detail below.

[0065] 1. Stable fluctuation decomposition calculation equation:

[0066] The communication delay data decomposition adopts Hilbert-Huang transform (HHT), which is specifically represented as follows:

[0067] ;

[0068] ;

[0069] ;

[0070] In the formula, is a communication delay time series; is a stable component; is an i-th intrinsic mode function (IMF) component; is an instantaneous amplitude; is an instantaneous frequency; is an external force function; is a decomposition error term.

[0071] The bandwidth data decomposition adopts the same method:

[0072] ;

[0073] In the formula, is a bandwidth time series; is a stable component; is a bandwidth IMF component; is a decomposition error term.

[0074] 2. Microservice node similarity calculation function:

[0075] ;

[0076] ;

[0077] ;

[0078] where, is the eigenvector of node i; is the time-varying weight function; is the adaptive bandwidth function; is the tuning parameter; is the variance operator; is the similarity error term.

[0079] 3. Microservice node type determination equation:

[0080] ;

[0081] ;

[0082] where, is the node type identification function; is the weight matrix; is the input eigenvector; is the Laplacian operator; is the type determination error term.

[0083] 4. Microservice node path distance equation:

[0084] ;

[0085] ;

[0086] where, is the path curve; is the metric tensor; is the optimization domain; is the boundary; is the path distance error term.

[0087] 5. Microservice node efficiency constraint equation:

[0088] ;

[0089] ;

[0090] ;

[0091] where, is the efficiency field function; is the flow field velocity; is the diffusion coefficient; is the source term function; is the viscosity coefficient; is the efficiency constraint error term.

[0092] 6. Microservice node total cost equation:

[0093] ;

[0094] ;

[0095] ;

[0096] where, is the control vector; is the Lagrangian operator; is the regularization parameter; is the Frobenius norm; is the total cost error term.

[0097] 7. Microservice node power balancing equation:

[0098] ;

[0099] where, is the power allocation matrix; is the system matrix; is the input matrix; is the control matrix; is the constraint matrix; is the power error matrix.

[0100] 8. Microservice node bandwidth balancing equation:

[0101] ;

[0102] where, is the bandwidth vector field; is the potential function; is the dynamic viscosity coefficient; is the external force term; is the bandwidth error vector.

[0103] 9. Microservice node load balancing equation:

[0104] ;

[0105] where, is the load distribution function; is the conduction coefficient; is the memory kernel function; is the nonlinear coupling function; is the load balancing error term.

[0106] Parameter acquisition method: system parameters The historical data is fitted by least squares method; the weight matrix is obtained by deep learning method; the error term range is determined by Monte Carlo simulation; the kernel function parameters are determined by cross-validation; the control parameters are obtained by optimal control theory.

[0107] The derivation process and parameter meanings of each equation are explained in detail below:

[0108] 1. Derivation of the calculation equation of stable fluctuation decomposition:

[0109] First, consider the empirical mode decomposition (EMD) of the time series signal, which is obtained by the following steps:

[0110] Step 1: Identify all local extreme points of the time series;

[0111] Step 2: Obtain the upper envelope and the lower envelope by cubic spline interpolation;

[0112] Step 3: Calculate the mean envelope: ;

[0113] Step 4: Extract the detail component: .

[0114] Then introduce the Hilbert transform to obtain the analytical signal:

[0115] ;

[0116] where is the Hilbert transform operator.

[0117] Finally, the expression of the IMF component is obtained:

[0118] ;

[0119] where and are obtained by the modulus and phase derivative of the analytical signal, respectively.

[0120] 2. Derivation of the similarity calculation function of microservice nodes:

[0121] Based on the kernel method theory, first construct the Gaussian kernel function:

[0122] ;

[0123] Consider the time-varying characteristics, introduce the time weight function:

[0124] ;

[0125] where is the time decay coefficient.

[0126] Combine gradient information to enhance local features:

[0127] ;

[0128] Finally, the complete similarity calculation function is obtained:

[0129] .

[0130] 3. Derivation of microservice node type determination equation:

[0131] Based on the theory of deep learning, first construct the basic activation function:

[0132] ;

[0133] Introduce time dynamics:

[0134] ;

[0135] Consider the spatial diffusion effect:

[0136] ;

[0137] Combine to get the final equation:

[0138] .

[0139] 4. Derivation of microservice node path distance equation:

[0140] Based on the shortest path theory, first define the metric tensor:

[0141] ;

[0142] In the formula, is each influencing factor.

[0143] Introduce the variational problem:

[0144] ;

[0145] Consider the boundary conditions:

[0146] ;

[0147] Finally, the complete equation is obtained:

[0148] .

[0149] 5. Derivation of microservice node efficiency constraint equation:

[0150] Based on the fluid dynamics equation, first consider the convection term:

[0151] ;

[0152] Introducing diffusion term:

[0153] ;

[0154] Constructing source term function:

[0155] ;

[0156] Combining to get final equation:

[0157] .

[0158] 6. Derivation of total cost equation of microservice node:

[0159] Based on optimal control theory, first construct Lagrange quantity:

[0160] ;

[0161] Consider the space-time evolution:

[0162] ;

[0163] Introducing integral constraints:

[0164] ;

[0165] Get the final equation:

[0166] .

[0167] 7. Derivation of power balance equation of microservice node:

[0168] Based on Lyapunov stability theory, construct state space equation:

[0169] ;

[0170] Introducing quadratic Lyapunov function:

[0171] ;

[0172] Consider the algebraic Lyapunov equation:

[0173] ;

[0174] Finally get the power balance equation:

[0175] .

[0176] 8. Derivation of microservice node bandwidth balancing equation:

[0177] Based on Navier-Stokes equation, first consider the inertia term:

[0178] ;

[0179] Introduce the viscosity term:

[0180] ;

[0181] Consider the pressure gradient:

[0182] ;

[0183] Add external force and error term to get the final equation:

[0184] .

[0185] 9. Derivation of microservice node load balancing equation:

[0186] Based on heat conduction equation, first consider the diffusion term:

[0187] ;

[0188] Introduce memory effect:

[0189] ;

[0190] Construct nonlinear coupling function:

[0191] ;

[0192] Get the final equation:

[0193] .

[0194] Optimization effect of each equation: stable fluctuation decomposition can effectively separate the stable trend and fluctuation component of the signal; similarity calculation considers time evolution and spatial correlation; type determination has good classification performance; path distance optimization ensures global optimality; efficiency constraint realizes multi-objective balance; total cost calculation considers the overall performance of the system; power balance ensures system stability; bandwidth allocation realizes dynamic adjustment of resources; load balancing considers historical influence and spatial distribution.

[0195] The second aspect of the application provides a computer readable storage medium, the computer readable storage medium stores program instructions, the program instructions are used to execute the above-mentioned industrial production microservice communication path optimization method when running in the computer.

[0196] The third aspect of the present application provides an industrial production microservice intercommunication path optimization method, comprising the computer readable storage medium described above.

[0197] The existing industrial production microservice intercommunication path determination method mainly adopts a single-layer optimization model, usually based on Dijkstra, Floyd and other shortest path algorithms, and combines load balancing strategies to select a communication path. In actual application, the prior art first establishes a connection relationship graph between microservice nodes, takes communication delay, bandwidth and other parameters as the weight of the edge, and then uses the shortest path algorithm to calculate the optimal communication path between nodes. This method has many limitations: the optimization goal is too single, only focuses on minimizing communication delay or maximizing bandwidth utilization, etc., and it is difficult to realize the coordinated optimization of multiple objectives; the static path planning method cannot adapt to the dynamic changes of microservice node states and network environment in industrial production environment; there is no hierarchical optimization mechanism, all optimization objectives are considered at the same level, which leads to the optimization process being easily trapped in local optimum, and the calculation complexity is high; the use of historical data and real-time monitoring data is not sufficient, and the regularity information contained in the data cannot be effectively extracted and utilized; the optimization granularity is usually fixed, and there is no adaptive adjustment ability at different spatial scales.

[0198] The microservice node communication path optimization double-layer model proposed by the present application realizes the organic combination of global optimization and local optimization through the innovative hierarchical optimization architecture. The outer model is responsible for macro path planning to ensure the optimal overall performance of the system, while the inner model focuses on micro resource scheduling to ensure efficient operation of the local area. The two-layer model realizes collaborative optimization through information interaction, which significantly improves the overall optimization effect. At the same time, the present application introduces a stable fluctuation decomposition technology to decompose the communication delay and bandwidth data into stable components and fluctuation components, wherein the stable components are used for long-term planning, and the fluctuation components are used for short-term adjustment, which greatly improves the accuracy of decision-making. Through similarity calculation and threshold judgment mechanism, the effective transmission and reuse of optimization experience is realized, which greatly improves the overall efficiency of the system.

[0199] In terms of dynamic optimization capability, the model of the present application can dynamically adjust the optimization strategy according to real-time monitoring data, fully considers the system dynamic characteristics at multiple time scales, and can quickly respond to changes in network environment and node state. The main communication path optimization equation set considers multiple objectives such as node type, path distance, efficiency constraint and total cost, while the local communication path optimization equation set focuses on balancing multiple resource dimensions such as power, bandwidth and load. Through reasonable weight setting, the compromise optimization of multiple objectives is realized, so that the system can achieve a good balance in various performance indicators.

[0200] The application also has significant advantages in robustness and reliability. By introducing error terms and constraint conditions, the robustness of the model is enhanced; considering the node failure rate and fault tolerance, the system reliability is improved; the optimization result has good stability to parameter disturbance. In terms of computational efficiency, the hierarchical optimization structure effectively reduces the complexity of the problem, reduces repeated calculation through the similarity transmission mechanism, and ensures the convergence of the algorithm by using the iterative optimization strategy. In addition, the model design of the application fully considers the actual needs of industrial production, the parameters can be flexibly adjusted according to the specific application scene, the framework has good scalability, and it is easy to integrate new optimization objectives and constraint conditions.

[0201] In summary, the application realizes significant improvement in key indicators such as communication delay, resource utilization, system response speed, load balancing degree and network congestion through the innovative design of a double-layer optimization model, and provides a new technical solution for microservice communication optimization in industrial production environment. This solution overcomes many limitations in the prior art and has strong optimization ability and adaptability in practical application, providing important technical support for the development of industrial internet.

[0202] The application scheme can solve the problems existing in the prior art, mainly due to the following key technical principles:

[0203] 1. Real-time monitoring and stable fluctuation decomposition. Real-time collection of communication data between microservice nodes, including delay, bandwidth, processing capacity, etc., to build a complete basic data set. Then use HHT algorithm to decompose the time series data into stable components and fluctuation components. This not only reflects the dynamic changes of node communication characteristics, but also provides more accurate input features for subsequent similarity calculation and optimization modeling.

[0204] 2. Similarity calculation and local optimization propagation. A microservice node similarity calculation function is established, which considers the communication delay, bandwidth, processing capacity and load rate of the node, and quantitatively describes the correlation between nodes. When the local optimization result of a node is better than the existing scheme, it will be synchronized to other nodes with high similarity. This similarity-based optimization result propagation mechanism can promote the penetration of local optimization to global optimization and improve the overall optimization effect.

[0205] 3. Double-layer optimization model coordination mechanism. A double-layer model for optimizing the communication path of microservice nodes is constructed. The outer layer model is responsible for global optimization, including node type determination, shortest path distance calculation, communication efficiency constraints, and overall resource consumption. The inner layer model is responsible for local optimization, including power balance, bandwidth balance, and load balancing. This hierarchical optimization framework can coordinate the communication path of microservice nodes at both global and local levels, ensuring optimal overall communication performance and resource utilization efficiency.

[0206] 4. Explainable mathematical modeling method. Mathematical modeling methods such as partial differential equations and variational methods are used to describe system dynamic characteristics, historical information influence, and complex interaction relationships. This modeling method not only solves the optimization result, but also understands the internal optimization mechanism, thereby improving the explainability of the scheme. Compared with black-box machine learning models, this explainable mathematical modeling method is more conducive to analysis and adjustment in engineering applications.

[0207] To better understand and implement the present application, a specific embodiment 1 of the present application is provided below, and the specific implementation of each step in embodiment 1 is described in detail as follows: the specific implementation of step S10 is as follows: first, real-time collection of basic communication data between microservice nodes is required. This includes microservice node communication delay data , microservice node bandwidth data , microservice node processing capability data , microservice node load rate data , microservice node data transmission volume data , microservice node historical failure rate data , microservice node concurrent processing capability data , microservice node network congestion data , microservice node hop weight data , microservice node geographic distance data , microservice node communication protocol compatibility data , microservice node service quality level data , microservice node fault tolerance capability data , microservice node load balancing factor data , microservice node resource utilization rate data , microservice node energy consumption coefficient data , and microservice node maintenance cost data The communication data can be collected in real time by means of sensor technology, system monitoring, log analysis, etc. Through the collection and analysis of the basic data, a basic data set of the microservice nodes can be established. The data set contains key characteristic parameters of the communication between the microservice nodes, and provides basic data support for subsequent stable fluctuation decomposition, similarity calculation and optimization modeling.

[0208] The specific implementation of step S20 is to perform stable fluctuation decomposition calculation on the collected microservice node communication delay data . The Hilbert-Huang Transform (HHT) algorithm can be used to decompose the communication delay time series. The HHT algorithm can decompose the original communication delay time series into a superimposed form of stable components and fluctuation components , that is:

[0209] ;

[0210] wherein, , is the instantaneous amplitude, is the instantaneous frequency, is the external force function, is the decomposition error term. In this way, the stable component and the fluctuation component of the communication delay can be obtained.

[0211] The specific implementation of step S30 is to also perform stable fluctuation decomposition calculation on the collected microservice node bandwidth data . Similarly, the HHT algorithm is used to decompose the original bandwidth time series into a form of stable components and fluctuation components :

[0212] ;

[0213] wherein, represents the IMF component of the bandwidth, is the decomposition error term. In this way, the stable component and the fluctuation component of the bandwidth are obtained.

[0214] Through the above two steps, the communication delay and bandwidth data of the microservice nodes are decomposed into stable components and fluctuation components of these key characteristic parameters. This lays a foundation for subsequent similarity calculation and optimization modeling.

[0215] The specific implementation of step S40 is: first, a microservice node similarity calculation function needs to be established and a similarity threshold . The microservice node similarity calculation function can take the following form:

[0216] ;

[0217] wherein, is the feature vector of node at time , containing parameters such as communication delay, bandwidth, processing capacity, and load rate; is a time-varying weight function; is an adaptive bandwidth function, determined by and two adjustment parameters; is a weight coefficient; is a similarity calculation error term.

[0218] Then, a similarity threshold can be set, and when the similarity of two microservice nodes is greater than the threshold, they are considered to be similar nodes. This threshold can be adjusted according to the actual application scenario, and is usually taken to be or so.

[0219] By establishing a similarity calculation function and setting a similarity threshold, the similarity between microservice nodes can be quantitatively described, providing a basis for subsequent optimization decisions.

[0220] The specific implementation of step S50 is: a double-layer microservice node communication path optimization model is constructed. The outer optimization model is responsible for global communication path optimization, and the inner optimization model is responsible for local communication path optimization.

[0221] The outer optimization model includes the following equations:

[0222] 1) Microservice node type determination equation:

[0223] ;

[0224] wherein, is a node type identification function, is an input feature vector, and are weight matrices, is a Laplace operator, is a type determination error term. Through this equation, it can be determined whether the node is a first-level transmission node or a second-level transmission node.

[0225] 2) Microservice node path distance equation:

[0226] ;

[0227] ;

[0228] where, is the path curve, is the optimization domain, is the boundary, is the regularization parameter, is the path distance error term. Through this equation, the shortest communication path distance between nodes can be calculated.

[0229] 3) Microservice node efficiency constraint equation:

[0230] ;

[0231] ;

[0232] ;

[0233] where, is the efficiency field function, is the flow field velocity, is the diffusion coefficient, is the source term function, is the viscosity coefficient, is the efficiency constraint error term. Through this equation, the efficiency of the communication path can be constrained.

[0234] 4) Microservice node total cost equation:

[0235] ;

[0236] ;

[0237] ;

[0238] where, is the control vector, is the Lagrange operator, is the regularization parameter, is the Frobenius norm, is the total cost error term. Through this equation, the overall resource consumption of the communication path can be calculated.

[0239] The inner optimization model includes the following several equations:

[0240] 1) Microservice node power balance equation:

[0241] ;

[0242] wherein, is a power allocation matrix, is a system matrix, is an input matrix, is a control matrix, is a constraint matrix, is a power error matrix. Through this equation, the calculation processing power allocation of the nodes can be balanced.

[0243] 2) Microservice node bandwidth balancing equation:

[0244] ;

[0245] wherein, is a bandwidth vector field, is a potential function, is a dynamic viscosity coefficient, is an external force term, is a bandwidth error vector. Through this equation, the bandwidth resource allocation of the nodes can be balanced.

[0246] 3) Microservice node load balancing equation:

[0247] ;

[0248] wherein, is a load allocation function, is a conduction coefficient, is a memory kernel function, is a nonlinear coupling function, is a load balancing error term. Through this equation, the task load allocation of the nodes can be balanced.

[0249] By constructing such a double-layer optimization model, the communication path of the microservice nodes can be coordinated and optimized at both global and local levels, considering both the overall communication efficiency and the resource balance between nodes. This method can improve the communication performance and resource utilization efficiency of the industrial production system.

[0250] The specific implementation of step S60 is: in the outer-layer optimization model, first establish a microservice node type judgment equation. This equation can take the following form:

[0251] ;

[0252] wherein, is a node type identification function, is an input feature vector, and are weight matrices, is the Laplace operator, is the type judgment error term. By training the deep learning model, the value of and can be determined. In this way, it can be determined whether the node is a first-level transmission node or a second-level transmission node.

[0253] Secondly, the micro-service node path distance equation is established. This equation can take the following form:

[0254] ;

[0255] ;

[0256] wherein, is the path curve, is the optimization domain, is the boundary, is the regularization parameter, is the path distance error term. By solving this optimization problem by the variational method, the shortest communication path distance between nodes can be obtained.

[0257] Thirdly, the micro-service node efficiency constraint equation is established. This equation can take the following form:

[0258] ;

[0259] ;

[0260] ;

[0261] wherein, is the efficiency field function, is the flow field velocity, is the diffusion coefficient, is the source term function, is the viscosity coefficient, is the efficiency constraint error term. By solving this partial differential equation system, the efficiency constraint value of the communication path can be obtained.

[0262] Finally, the micro-service node total cost equation is established. This equation can take the following form:

[0263] ;

[0264] ;

[0265] ;

[0266] wherein, is the control vector, is the Lagrangian operator, is the regularization parameter, is the Frobenius norm, is the total cost error term. By solving this partial differential equation, the overall resource consumption of the communication path can be obtained .

[0267] The outer optimization model composed of the above four equations can obtain the optimal communication path in the global range. These equations involve key factors such as node type determination, path distance calculation, communication efficiency constraint and overall resource consumption, which can provide a basis for subsequent local optimization.

[0268] The specific implementation of step S70 is: in the inner optimization model, first, the micro-service node power balance equation is established. This equation can take the following form:

[0269] ;

[0270] wherein, is the power allocation matrix, is the system matrix, is the input matrix, is the control matrix, is the constraint matrix, is the power error matrix. By solving this matrix differential equation, the power allocation scheme of the node can be obtained .

[0271] Secondly, the micro-service node bandwidth balance equation is established. This equation can take the following form:

[0272] ;

[0273] wherein, is the bandwidth vector field, is the potential function, is the dynamic viscosity coefficient, is the external force term, is the bandwidth error vector. By solving this vector differential equation, the bandwidth allocation scheme of the node can be obtained .

[0274] Thirdly, the micro-service node load balancing equation is established. This equation can take the following form:

[0275] ;

[0276] wherein, is the load allocation function, is the conduction coefficient, is the memory kernel function, is a nonlinear coupling function, is a load balancing error term. By solving this nonlinear integral differential equation, the load distribution scheme of the nodes can be obtained .

[0277] The above three equations constitute the inner optimization model. An upper limit of the number of iterations of the micro-service node optimization can also be set , and the inner optimization is terminated when the upper limit is reached.

[0278] Through the inner optimization model, the power, bandwidth and load of the nodes can be balanced in a local range, ensuring the full use of resources. These local optimization results will provide important input for subsequent global optimization.

[0279] The specific implementation of step S80 is: first, the iterative optimization of the outer optimization model is performed to obtain an initial scheme of the global communication path of the micro-service node. Then, the initial scheme is taken as input, and the iterative optimization of the inner optimization model is performed to generate a local communication path optimization scheme. When the load distribution value of a certain node is better than the existing load distribution value, the local optimization scheme is synchronized to other nodes with a similarity greater than a similarity threshold.

[0280] The purpose of this is to enable the local optimization results to be shared and propagated among similar nodes, further optimizing the global communication effect. This process is repeated until the upper limit of the number of iterations of the micro-service node optimization is reached .

[0281] The specific implementation of step S90 is: after experiencing multiple rounds of iterative optimization of the outer and inner layers, the optimal micro-service node communication topology graph is finally obtained. This topology graph contains the output results of the outer and inner optimization models, including: node type determination results ; the shortest communication path between nodes ; the communication path efficiency constraint value ; the overall resource consumption of the communication path ; the node power allocation matrix ; the node bandwidth allocation vector ; and the node load distribution value .

[0282] These output results reflect the optimization of the entire system at the global and local levels, and can provide a decision basis for the communication management of industrial production systems.

[0283] The parameter acquisition and equation design principles of the entire optimization method are as follows:

[0284] 1) System parameters​ 、 、 、 , etc. can be obtained by fitting historical data with the least square method;

[0285] 2) Weight matrix 、 , etc. can be obtained by training with deep learning methods;

[0286] 3) Error term range 、 、 、 、 , etc. can be determined by Monte Carlo simulation;

[0287] 4) Kernel function parameters can be determined by cross-validation;

[0288] 5) Control parameters can be obtained by optimal control theory.

[0289] The following provides an embodiment 2 of a specific application scenario of the present application: A certain intelligent pharmaceutical is building a production management system based on micro-service architecture. The system divides the entire pharmaceutical production process into multiple independent micro-service nodes, including raw material procurement management, formula preparation control, finished product packaging monitoring, etc. These micro-service nodes dynamically communicate and cooperate through high-speed networks to realize the automation and optimization of the production process.

[0290] In order to improve the communication performance and resource utilization efficiency of the micro-service system, it is decided to use the industrial production micro-service intercommunication path optimization method proposed by the present application. The following is the specific implementation of this method in the production system.

[0291] First, a real-time monitoring system is deployed to collect basic communication data between micro-service nodes. According to the requirements of the present application, the monitoring content includes:

[0292] 1. Communication delay data : By deploying delay detectors on the links between nodes, real-time communication delay time series between micro-service nodes are collected. Through statistical analysis, the communication delay data of the pharmaceutical system fluctuates within .

[0293] 2. Bandwidth data : Bandwidth monitoring probes are deployed on the backbone network devices to collect real-time available bandwidth data between micro-service nodes. The statistical results show that the bandwidth between nodes changes within .

[0294] 3. Processing capacity data The CPU utilization, memory occupation and other indicators of each micro-service node are collected by the performance monitoring module of the node itself and converted into the processing capacity parameters of the node. In the pharmaceutical system, the processing capacity of the node is between .

[0295] 4. Load rate data The CPU load, memory occupation and other indicators of each micro-service node are also collected in real time by the monitoring module of the node itself, and the actual load rate of the node is calculated. In the system, the node load rate varies between .

[0296] 5. Other parameters: In addition, the data transmission volume, historical failure rate, concurrent processing capacity, network congestion, hop weight, geographic distance, communication protocol compatibility, service quality level, fault tolerance capability, load balancing factor, resource utilization rate, energy consumption coefficient and maintenance cost of the micro-service node are also collected.

[0297] Through the above means, a complete set of micro-service node basic data is established, laying a foundation for subsequent stable fluctuation decomposition, similarity calculation and optimization modeling.

[0298] Figure 2 The time series changes of the basic monitoring data are shown, including two subgraphs: the upper graph shows the fluctuation of communication delay D(t) over time, and the delay range is between 5-50 ms; the lower graph shows the change of bandwidth B(t) over time, and the bandwidth range is between 100-500 Mbps. The time series data of the two key indicators intuitively shows the dynamic characteristics of the system communication performance.

[0299] Secondly, the Hilbert-Huang Transform (HHT) algorithm is used to decompose the collected communication delay data and bandwidth data .

[0300] For the decomposition of the communication delay data , the specific process is as follows:

[0301] ;

[0302] ;

[0303] ;

[0304] wherein is the stable component of the communication delay, is the th intrinsic mode function (IMF) component, is the instantaneous amplitude of the IMF component, instantaneous frequency of the IMF component, external force function, decomposition error term. After HHT decomposition, the stable component and the fluctuation component of the communication delay are obtained.

[0305] Similarly, the decomposition of bandwidth data also adopts a similar HHT method, obtaining the stable component and the fluctuation component of the bandwidth.

[0306] Figure 3 The results of HHT decomposition of communication delay data are shown, including three IMF components and one trend term. From top to bottom are: high-frequency IMF component, medium-frequency IMF component, low-frequency IMF component, and overall trend term. This decomposition helps to understand the multi-scale dynamic characteristics of communication delay.

[0307] Through the above stable fluctuation decomposition process, the dynamic change law of the communication characteristics of microservice nodes is mastered, laying a foundation for subsequent similarity calculation and optimization modeling.

[0308] Next, the microservice node similarity calculation function is established, which is specifically as follows:

[0309] ;

[0310] ;

[0311] ;

[0312] where, is the feature vector of node at time , including communication delay, bandwidth, processing capacity, and load rate parameters; is a time-varying weight function; is an adaptive bandwidth function, determined by and two adjustment parameters; is a weight coefficient; is the error term of similarity calculation.

[0313] According to the actual application requirements, the similarity threshold is set. When the similarity of two microservice nodes is greater than the threshold, they are considered to be similar nodes.

[0314] Figure 4This is a 10×10 node similarity heatmap, showing the similarity metrics between different nodes. The darker the color, the higher the similarity. The self-similarity on the diagonal is 1, and it can be seen that the similarity between adjacent nodes is generally high.

[0315] With the basic dataset and similarity calculation function for microservice nodes, we then constructed a two-layer model for optimizing communication paths between microservice nodes.

[0316] The outer optimization model includes the following equations:

[0317] 1. Microservice node type determination equation:

[0318] ;

[0319] ;

[0320] This equation can determine the nodes. Is it a primary or secondary transmission node? Input includes load rate. Processing capacity Historical failure rate Concurrency processing capability and network congestion Parameters, the output is a node type identifier. The weight matrix was determined using deep learning methods. and The value of .

[0321] 2. Microservice node path distance equation:

[0322] ;

[0323] ;

[0324] This equation can calculate the nodes. and nodes Shortest communication path distance between The input includes a stable component of communication delay. Bandwidth stable components hop count weight Geographical distance and communication protocol compatibility Parameters such as [parameters are specified]. The optimization problem was solved using the variational method, yielding the shortest path distances between each node.

[0325] 3. Efficiency constraint equation for microservice nodes:

[0326] ;

[0327] ;

[0328] ;

[0329] This equation can constrain the efficiency of the communication path, and the input includes the communication delay fluctuation component, bandwidth fluctuation component, quality of service level , fault tolerance capability , and load balancing factor and other parameters. By solving the partial differential equation, the efficiency constraint value of the communication path between nodes .

[0330] 4. Microservice node total cost equation:

[0331] ;

[0332] ;

[0333] ;

[0334] This equation can calculate the overall resource consumption of the communication path , and the input includes node type , shortest path distance , efficiency constraint value , resource utilization , energy consumption coefficient and maintenance cost and other parameters. By solving the partial differential equation, the overall resource consumption of the communication path is obtained.

[0335] The inner optimization model includes the following equations:

[0336] 1. Microservice node power balance equation:

[0337] ; , , , , ;

[0338] This equation can balance the computing processing power distribution of the node , and the input includes processing capability , concurrent processing capability , load rate , resource utilization and energy consumption coefficient and other parameters. By solving the matrix differential equation, the power distribution scheme of each node is obtained.

[0339] 2. Microservice node bandwidth balancing equation:

[0340] ;

[0341] , ;

[0342] This equation can balance the bandwidth resource allocation of nodes , and the input includes parameters such as bandwidth stable component , bandwidth fluctuation component , data transmission volume , network congestion degree , and communication protocol compatibility . By solving this vector differential equation, the bandwidth allocation scheme of each node is obtained.

[0343] 3. Microservice node load balancing equation:

[0344] ;

[0345] This equation can balance the task load allocation of nodes , and the input includes parameters such as power allocation , bandwidth allocation , fault tolerance capability , quality of service level , and load balancing factor . By solving this nonlinear integral differential equation, the load allocation scheme of each node is obtained.

[0346] With the outer and inner optimization models, multiple rounds of iterative optimization are performed.

[0347] First, the iteration of the outer optimization model is performed to obtain the initial scheme of the global communication path of the microservice node. For example, according to the type determination equation, 20% of the nodes in the pharmaceutical system are identified as first-level transmission nodes, and 80% of the nodes are identified as second-level transmission nodes. According to the path distance equation, the shortest communication path distance between two first-level transmission nodes is , and the shortest path distance from a first-level node to a second-level node is . According to the efficiency constraint equation, the efficiency constraint value of the communication path between nodes is between 0.8 and 0.9. According to the total cost equation, the total resource consumption of the entire communication path is about 1000 units.

[0348] Then, the initial scheme of the global communication path is taken as input, and the iteration of the inner optimization model is performed. For example, according to the power balancing equation, the power allocation of a node is ; according to the bandwidth balancing equation, the bandwidth allocation of a node is ;according to the load balancing equation, the load distribution of a certain node .

[0349] When the load distribution value of a certain node is better than the existing load distribution value, the local optimization scheme is synchronized to other nodes with a similarity greater than 0.8.

[0350] The above process is repeated until the set upper limit of the number of iterations of optimization times. Finally, the optimized micro-service node communication topology graph is obtained, including all output results of the outer and inner models.

[0351] For example, after optimization, the first-level transmission nodes in the pharmaceutical system account for 30%, and the second-level transmission nodes account for 70%. The shortest communication path distance between two first-level transmission nodes is shortened to , and the shortest path distance from a first-level node to a second-level node is reduced to . The efficiency constraint value of the communication path between nodes is increased to between 0.9 and 0.95. The overall resource consumption of the entire communication path is reduced to about 800 units. The power allocation of a certain node is adjusted to , the bandwidth allocation is adjusted to , and the load distribution is reduced to .

[0352] Through the above optimization process, the micro-service system realizes significant improvement in communication performance and substantial improvement in resource utilization efficiency. Compared with the previous scheme, the end-to-end communication latency of the system is reduced by 30%, the system throughput is increased by 40%, and the node resource utilization is also increased by 25%. This creates a good network foundation condition for the intelligentization and automation of pharmaceutical production.

[0353] Figure 5 The optimized micro-service communication topology graph is shown, where green nodes represent first-level transmission nodes (accounting for 30%), and yellow nodes represent second-level transmission nodes (accounting for 70%). The connection between nodes represents the communication path, and the thickness of the line represents the bandwidth size.

[0354] It should be noted that the variables involved in the present application are explained in detail as shown in Table 1.

[0355] Table 1: Variable Explanation Table

[0356]

[0357] The above merely illustrates the specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. An industrial production inter-microservice communication path optimization method, characterized in that, The method comprises the following steps: Real-time collection of basic communication data between micro-service nodes to generate a micro-service node basic data set containing multi-dimensional indicators; Stable fluctuation decomposition calculation of micro-service node communication delay data and micro-service node bandwidth data in the basic data set to obtain corresponding stable components and fluctuation components; Establishment of a micro-service node similarity calculation function and a micro-service node similarity threshold, calculation of node similarity values based on the micro-service node basic data set, construction of a micro-service node communication path optimization double-layer model containing an outer optimization model and an inner optimization model, establishment of a main communication path optimization equation group of the micro-service node communication path outer optimization model and a local communication path optimization equation group of the inner optimization model, and execution of iterative optimization of the double-layer model and synchronization of optimization schemes when optimization conditions are met; Repetition of the following steps until the upper limit of the number of micro-service node iterative optimizations is reached to obtain a final micro-service node communication topology graph: execution of iterative optimization of the micro-service node communication path outer optimization model to obtain a micro-service node global communication path initial scheme, execution of iterative optimization of the micro-service node communication path inner optimization model with the micro-service node global communication path initial scheme as input to generate a micro-service node local communication path optimization scheme, and synchronization of the micro-service node local communication path optimization scheme to other micro-service nodes with micro-service node similarity values greater than the micro-service node similarity threshold when the micro-service node load distribution value of a certain micro-service node is better than the existing micro-service node load distribution value; The micro-service node load distribution value is the output of a micro-service node load balancing equation for balancing task load distribution of the micro-service nodes.

2. The method of claim 1, wherein, The micro-service node basic data set includes micro-service node communication delay data, micro-service node bandwidth data, micro-service node processing capability data, micro-service node load rate data, micro-service node data transmission volume data, micro-service node historical failure rate data, micro-service node concurrent processing capability data, micro-service node network congestion degree data, micro-service node hop weight data, micro-service node geographic distance data, micro-service node communication protocol compatibility data, micro-service node service quality level data, micro-service node fault tolerance capability data, micro-service node load balancing factor data, micro-service node resource utilization rate data, micro-service node energy consumption coefficient data, and micro-service node maintenance cost data.

3. The method of claim 1, wherein, Stable fluctuation decomposition calculation of the micro-service node communication delay data to obtain micro-service node communication delay stable components and micro-service node communication delay fluctuation components; Stable fluctuation decomposition calculation of the micro-service node bandwidth data to obtain micro-service node bandwidth stable components and micro-service node bandwidth fluctuation components.

4. The method of claim 1, wherein, The micro-service node communication path outer optimization model is used for global communication path optimization, and the micro-service node communication path inner optimization model is used for local communication path optimization.

5. The method of claim 1, wherein, The main communication path optimization equation set includes a microservice node type determination equation, a microservice node path distance equation, a microservice node efficiency constraint equation, and a microservice node total cost equation.

6. The industrial production inter-microservice communication path optimization method according to claim 5, characterized in that, The microservice node type determination equation is used to determine whether the microservice node belongs to a first-level transmission microservice node or a second-level transmission microservice node. The microservice node path distance equation is used to calculate the shortest distance of the communication path between the microservice nodes. The microservice node efficiency constraint equation is used to limit the efficiency threshold of the communication path between the microservice nodes. The microservice node total cost equation is used to calculate the overall resource consumption of the communication path of the microservice node.

7. The method of claim 1, wherein, The local communication path optimization equation set includes a microservice node power balance equation, a microservice node bandwidth balance equation, and a microservice node load balancing equation.

8. The industrial production inter-microservice communication path optimization method according to claim 7, characterized in that, The microservice node power balance equation is used to balance the computational processing power allocation of the microservice node; the microservice node bandwidth balance equation is used to balance the bandwidth resource allocation of the microservice node, and the input includes the microservice node bandwidth stable component, the microservice node bandwidth fluctuation component, the microservice node data transmission amount data, the microservice node network congestion degree data, and the microservice node communication protocol compatibility data. The microservice node load balancing equation is used to balance the task load allocation of the microservice node.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores program instructions, which when running in the computer, are used to execute the industrial production microservice intercommunication path optimization method of any one of claims 1-8.

10. An industrial production inter-microservice communication path optimization system, characterized in that, The computer readable storage medium of claim 9 is included.

Images