Request shunting system based on configuration algorithm

Abstract

The invention relates to a request shunting system based on a configuration algorithm, and relates to the field of computer networks and software engineering. The system comprises a data acquisition module which is used for collecting server state information and network request characteristics; the configuration algorithm module is used for adjusting a load distribution strategy of the server; the strategy configuration module is used for customizing and testing a shunting strategy by a background administrator; the shunting decision module is used for performing shunting decision on the output result of the configuration algorithm; the execution module is used for executing the shunting decision and guiding a request for executing the shunting decision to the server or the processing logic; and the monitoring and self-adaption module is used for monitoring the shunting effect and adjusting the shunting rule. According to the invention, the method achieves the dynamic adjustment of the request distribution strategy, guarantees that the server is still effectively processed when a fault or overload occurs, achieves the differentiated processing mode of the request, simplifies the system configuration, and improves the operation and maintenance efficiency.

Description

Technical Field

[0001] The present invention relates to the fields of computer network and software engineering, and in particular to a request diversion system based on a configuration algorithm. Background Art

[0002] The rapid development of Internet services has led to a surge in user requests, which has put higher demands on the resource allocation and processing capabilities of the server. Request diversion systems usually use specific algorithms to optimize resource allocation and request routing. For example, in mobile edge networks, the efficiency of AI-generated content services can be significantly improved by optimizing resource allocation and request routing technologies. In addition, some systems, such as the Mafengwo ABTest multi-layer diversion system, use elements such as device ID, URI, and allocation strategy, and use algorithms such as MurmurHash to achieve stable diversion guarantees. However, the above algorithms all use a static approach to load balancing, which lacks flexibility and adaptability in the face of changing network conditions and request patterns. Summary of the invention

[0003] The purpose of the present invention is to solve the problems existing in the prior art and to provide a request diversion system based on a configuration algorithm, which realizes dynamic adjustment of the request allocation strategy, ensures that the server is still effectively processed when a failure or overload occurs, realizes differentiated processing of requests, simplifies system configuration, and improves operation and maintenance efficiency.

[0004] The present invention provides a request diversion system based on a configuration algorithm, comprising:

[0005] Data collection module: used to collect server status information and network request characteristics;

[0006] Configuration algorithm module: used to adjust the server load distribution strategy;

[0007] Strategy configuration module: used by backend administrators to customize and test diversion strategies;

[0008] Diversion decision module: used to make diversion decisions based on the output results of the configuration algorithm;

[0009] Execution module: used to execute the diversion decision and direct the execution diversion decision request to the server or processing logic;

[0010] Monitoring and adaptive module: used to monitor the diversion effect and adjust the diversion rules.

[0011] The data collection module includes a data collection unit for collecting, storing and monitoring server load, network status and request characteristics and a request characteristic extraction unit for extracting request type, size and frequency.

[0012] The data acquisition unit includes a data acquisition server, a storage server and a monitoring server, which are connected in sequence. The request feature extraction unit includes a feature extractor, which is connected to the acquisition server and the storage server respectively.

[0013] The data collection implementation steps include:

[0014] Step 1: Data collection: Start the data collection server to collect server load, response time, and real-time traffic data;

[0015] Step 2: Data storage: The collected data is stored in the storage server. At the same time, the collected data enters the feature extractor, and the data enters the storage server after extraction;

[0016] Step 3: Data monitoring: The monitoring server monitors the collected and extracted data in real time and feeds the monitoring data back to the data collection server.

[0017] The configuration algorithm module includes a parameter monitoring unit for collecting and real-time monitoring of parameters, a diversion screening and optimization unit for dynamically screening and optimizing the collected data, a diversion performance simulation and evaluation unit for simulating diversion performance and evaluating diversion effects, a configuration adjustment unit for dynamically adjusting diversion, and an adaptive unit for adjusting the diversion algorithm based on system response and user feedback to realize an adaptive mechanism to continuously optimize the algorithm.

[0018] The steps of diversion screening include:

[0019] Step 1: Initialize the parameter weight list: set the initial weight of each parameter, and adjust the weight according to business needs;

[0020] Step 2: Parameter matching analysis: Match the parameters in the transaction request with each rule and create a matching score for each parameter;

[0021] Step 3: Rule popularity calculation: Calculate the usage frequency of each rule based on historical data and adjust the rule score accordingly;

[0022] Step 4: Execution time evaluation: used to calculate and compare the average time required to execute each rule in history;

[0023] Step 5: Calculate the total score;

[0024] Step 6: Rule sorting and selection: used to sort all rules according to the calculated total score and select the rule with the highest score as the optimal rule;

[0025] Step 7: Dynamic update: used to regularly analyze the system processing effect and dynamically adjust the parameter weights.

[0026] The diversion screening adopts a dynamic rule matching score algorithm, and the expression of the dynamic rule matching score algorithm is:

[0027] S j =∑ n i＝1 (M i ×W i )+H j -T j

[0028] Where: S j is the total score, S j The rule with the highest value is taken as the optimal rule;

[0029] M i is the matching score of the i-th parameter;

[0030] W i is the weight score of the i-th parameter;

[0031] H j is the heat score of the jth rule;

[0032] T j is the average time required to execute the jth rule.

[0033] The diversion performance simulation adopts a simulation algorithm, and the expression of the simulation algorithm is:

[0034] System throughput formula: T = ∑ n i＝1 (1 / T i ) / n

[0035] Where T is the overall throughput of the system, n is the number of rules, and Ti is the processing time of the i-th diversion rule;

[0036] Hit rate formula: H = number of successful matches / total number of requests × 100%

[0037] Where H is the hit rate of the rule;

[0038] Response time simulation formula: R = ∑ m j＝1 p j × j

[0039] Where R is the expected average response time, m is the number of diversion points, and p j is the request ratio of the jth diversion point,

[0040] r j is the average processing time of the jth diversion point.

[0041] Resource utilization evaluation formula: U = used resources / total resources

[0042] Among them, U is the resource utilization rate, which is used to measure the efficiency of resource allocation;

[0043] Load balancing formula: A = ∑ o k＝1 min(L k ) / O

[0044] Among them, A is the load balancing degree, o is the number of split sets, and L k is the load value in the kth flow distribution set;

[0045] Comprehensive evaluation effect formula: E = αT + βH + γR + δU + ∈A

[0046] Among them, E is the comprehensive evaluation result;

[0047] α is the system throughput weight coefficient,;

[0048] β is the hit rate weight coefficient;

[0049] γ is the response time weight coefficient;

[0050] δ is the resource utilization weight coefficient;

[0051] ∈ is the load balancing weight coefficient; α, β, γ, δ and ∈ are all used to adjust the relative importance of each indicator in the total score.

[0052] The diversion performance simulation step comprises:

[0053] Step 1: Design of simulation scenarios: Design transaction scenarios and corresponding parameters based on historical data and business development forecasts;

[0054] Step 2: Parameter input: input configuration parameters, weights and total resources;

[0055] Step 3: Rule matching and execution: During the simulation process, the rules are matched and executed;

[0056] Step 4: Calculate performance indicators: Calculate system throughput, hit rate, response time, resource utilization and load balance according to the simulation algorithm;

[0057] Step 5: Result analysis and optimization: Analyze simulation results to identify performance bottlenecks and improvement points;

[0058] Step 6: Repeat steps 2 to 6 until the optimal diversion rule configuration is found.

[0059] The diversion decision module includes a data parsing unit for extracting and parsing request data, a temporary queue management unit for temporarily storing predetermined priorities, a decision unit for determining and routing priorities, an execution diversion unit for request scheduling, and a monitoring feedback unit for monitoring and feeding back execution and response.

[0060] Compared with the prior art, the present invention has the following beneficial effects:

[0061] 1. Improve resource utilization: Allocate requests based on actual needs to avoid waste of resources.

[0062] 2. Enhance the system's fault tolerance: Reduce the load pressure on a single server through effective request diversion.

[0063] 3. Improve user experience: Give priority to high-priority requests and improve the processing efficiency of key businesses.

[0064] 4. Simplify system configuration and maintenance: Configuration algorithms simplify system configuration, reduce human errors and improve operation and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a system schematic diagram of Embodiment 1 of the present invention;

[0066] Figure 2 This is a data collection flow chart of Example 1 of the present invention;

[0067] Figure 3 This is a flowchart of the configuration algorithm module of Example 1 of the present invention;

[0068] Figure 4 This is a policy configuration flow chart of Example 1 of the present invention;

[0069] Figure 5 This is a flow chart of the diversion decision-making in Embodiment 1 of the present invention;

[0070] Figure 6 This is a flowchart of the execution module of Example 1 of the present invention;

[0071] Figure 7 This is a workflow diagram of the monitoring and self-adaptation module of Example 1 of the present invention. DETAILED DESCRIPTION

[0072] The present invention will be further described below in conjunction with embodiments:

[0073] Example 1

[0074] like Figure 1-Figure 7 As shown, the request diversion system based on the configuration algorithm described in the present invention includes

[0075] Data collection module: used to collect server status information and network request characteristics; the data collection module includes a data collection unit for collecting, storing and monitoring server load, network status and request characteristics, and a request feature extraction unit for extracting request type, size and frequency. The server load specifically includes CPU usage, memory usage and I / O operation frequency, and the network status specifically includes bandwidth usage, delay time and packet loss rate. The modular design of the data collection unit and the request feature extraction unit allows the system to be easily expanded as needed to adapt to different scales and needs.

[0076] The data collection unit includes a data collection server, a storage server and a monitoring server. The collection server, the storage server and the monitoring server are connected in sequence. The request feature extraction unit includes a feature extractor, which is connected to the collection server and the storage server respectively. The data collection server collects key indicators such as server load, response time, and real-time traffic in real time, ensuring a comprehensive understanding of the server status and network request characteristics. The collected data is directly stored in the storage server, ensuring the security and accessibility of the data, and facilitating historical data analysis. The monitoring server monitors the collected data in real time, can promptly discover and respond to potential problems, and improves the stability and reliability of the system. The request feature extraction unit extracts key features such as request type, size, and frequency through the feature extractor, providing accurate data support for subsequent data analysis and diversion decisions.

[0077] The steps to implement data collection include:

[0078] Step 1: Data collection: Start the data collection server to collect server load, response time, and real-time traffic data;

[0079] Step 2: Data storage: The collected data is stored in the storage server. At the same time, the collected data enters the feature extractor, and the data enters the storage server after extraction; the automated data collection and storage process reduces manual intervention, reduces operating costs, and improves the accuracy of data processing.

[0080] Step 3: Data monitoring: The monitoring server monitors the collected and extracted data in real time and feeds the monitoring data back to the data collection server.

[0081] Configuration algorithm module: used to adjust the load distribution strategy of the server; the configuration algorithm module includes a parameter monitoring unit for collecting and real-time monitoring of parameters, a diversion screening and optimization unit for dynamically screening and optimizing the collected data, a diversion performance simulation and evaluation unit for simulating diversion performance and evaluating diversion effects, and a configuration adjustment unit for dynamically adjusting diversion, and an adaptive unit for adjusting the diversion algorithm based on system response and user feedback to realize the adaptive mechanism and continuous optimization algorithm. The workflow of the configuration algorithm module is to pass the data to the parameter monitoring unit through the input module, and then perform rule screening and optimization in the diversion screening and optimization unit, perform performance simulation and evaluation on the optimized data, and adjust the configuration algorithm according to the evaluation results. The adjusted results are continuously optimized by the adaptive unit and then enter the diversion decision module through the output module.

[0082] The steps of triage screening include:

[0083] Step 1: Initialize the parameter weight list: set the initial weight of each parameter, and adjust the weight according to business needs;

[0084] Step 2: Parameter matching analysis: Match the parameters in the transaction request with each rule and create a matching score for each parameter;

[0085] Step 3: Rule heat calculation: Calculate the usage frequency of each rule based on historical data and adjust the rule score accordingly; Step 4: Execution time evaluation: Used to calculate and compare the average time required to execute each rule in history;

[0086] Step 5: Calculate the total score;

[0087] Step 6: Rule sorting and selection: used to sort all rules according to the calculated total score and select the rule with the highest score as the optimal rule;

[0088] Step 7: Dynamic update: used to regularly analyze the system processing effect and dynamically adjust the parameter weights.

[0089] The diversion screening adopts the dynamic rule matching score algorithm, and the expression of the dynamic rule matching score algorithm is:

[0090] S j =∑ n i＝1 (M i ×W i )+H j -T j

[0091] Where: S j is the total score, S j The rule with the highest value is taken as the optimal rule;

[0092] M i is the matching score of the i-th parameter;

[0093] W i is the weight score of the i-th parameter;

[0094] H j is the heat score of the jth rule;

[0095] T j is the average time required to execute the jth rule.

[0096] The simulation algorithm is used for the diversion performance simulation. The expression of the simulation algorithm is:

[0097] System throughput formula: T = ∑ n i＝1 (1 / T i ) / n

[0098] Where T is the overall throughput of the system, n is the number of rules, and Ti is the processing time of the i-th diversion rule;

[0099] Hit rate formula: H = number of successful matches / total number of requests × 100%

[0100] Where H is the hit rate of the rule;

[0101] Response time simulation formula: R = ∑ m j＝1 p j × j

[0102] Where R is the expected average response time, m is the number of diversion points, and p j is the request ratio of the jth diversion point,

[0103] And r j is the average processing time of the jth diversion point.

[0104] Resource utilization evaluation formula: U = used resources / total resources

[0105] Among them, U is the resource utilization rate, which is used to measure the efficiency of resource allocation;

[0106] Load balancing formula: A = ∑ o k＝1 min(L k ) / O

[0107] Among them, A is the load balancing degree, o is the number of split sets, and L k is the load value in the kth flow distribution set;

[0108] Comprehensive evaluation effect formula: E = αT + βH + γR + δU + ∈A

[0109] Among them, E is the comprehensive evaluation result;

[0110] α is the system throughput weight coefficient,;

[0111] β is the hit rate weight coefficient;

[0112] γ is the response time weight coefficient;

[0113] δ is the resource utilization weight coefficient;

[0114] ∈ is the load balancing weight coefficient; α, β, γ, δ and ∈ are all used to adjust the relative importance of each indicator in the total score.

[0115] The steps of shunting performance simulation include:

[0116] Step 1: Design of simulation scenarios: Design transaction scenarios and corresponding parameters based on historical data and business development forecasts;

[0117] Step 2: Parameter input: Input configuration parameters, weights, and total resources; specifically, the specific parameters and weights to be used for each diversion rule in the configuration simulation, as well as the total resources that the system can use during the simulation.

[0118] Step 3: Rule matching and execution: During the simulation, the steps in the algorithm are used to match and execute rules for each transaction scenario;

[0119] Step 4: Calculate performance indicators: Calculate system throughput, hit rate, response time, resource utilization and load balance according to the simulation algorithm;

[0120] Step 5: Result analysis and optimization: Analyze simulation results, identify performance bottlenecks and improvement points, and optimize rule settings;

[0121] Step 6: Repeat steps 2 to 6, adjust the parameters according to the optimization results, and re-simulate until the optimal diversion rule configuration is found.

[0122] By running simulation algorithms, we evaluate the performance of different diversion rules in processing various transaction requests. Specifically, performance comparison: compare the comprehensive scores of different diversion rules in the simulation and select the rule configuration with the highest score. Feasibility analysis: analyze the feasibility and potential risks of the selected rule configuration in the real environment. Optimization adjustment: adjust the rule parameters according to the simulation results, optimize the diversion rules, and perform another simulation to verify the optimization effect.

[0123] Through traffic diversion performance simulation and evaluation, the performance of traffic diversion rules can be predicted, their effectiveness verified, and optimized before the actual situation occurs. This process ensures that when these rules are implemented, the system can operate in the best state, maintain system performance under peak load conditions, and provide better customer service.

[0124] Policy configuration module: used by backend administrators to customize and test diversion strategies; includes a policy configuration unit, specifically inputs or adjusts diversion strategies to the policy configuration unit, calculates and optimizes the strategies, determines whether it is a valid strategy evaluation, and if it is a valid strategy evaluation, applies the strategy to the diversion decision module, monitors the application strategy data, and feeds back the monitoring results to the policy configuration unit. If it is an invalid strategy evaluation, it needs to be adjusted and re-evaluated.

[0125] Diversion decision module: used to make diversion decisions on the output results of the configuration algorithm; the diversion decision module includes a data parsing unit for extracting and parsing request data, a temporary queue management unit for temporarily storing predetermined priorities, a decision unit for determining and routing priorities, an execution diversion unit for request scheduling, and a monitoring feedback unit for monitoring and feedback of execution and response. The specific process of diversion decision is as follows:

[0126] Parse request characteristics: Parse the received requests and extract the key data required for diversion decisions. Categorize based on request type, size, priority, etc.

[0127] Queue management: temporarily store requests in queues according to predetermined priorities, and implement dynamic request buffering and traffic shaping.

[0128] Application configuration algorithm rules: Make decisions based on the diversion rules and configuration parameters provided by the configuration algorithm module. Prioritize and route requests in all queues according to the diversion rules.

[0129] Execute task dispatch: distribute requests to servers or processing units according to the diversion strategy. Realize real-time request scheduling.

[0130] Monitor the diversion effect: Monitor the execution and response of each request. Provide execution feedback to the configuration algorithm module for further adjustment.

[0131] Exception handling mechanism: Detects abnormal situations during request execution and implements failover or rerouting strategies.

[0132] Execution module: used to execute the diversion decision and direct the execution diversion decision request to the server or processing logic; the execution module workflow is as follows Figure 6 As shown,

[0133] Diversion decision engine output: The system first receives a request and analyzes the request through the diversion decision engine.

[0134] Request routing: Based on the analysis results of the diversion decision engine, the system decides which server the request should be routed to.

[0135] Server Storage Queue: The selected server places the request in its storage queue.

[0136] Allocation successful: If the request is successfully allocated to the server, the system will record monitoring data for subsequent analysis and optimization.

[0137] Task execution: The server takes the request from the queue and performs task processing.

[0138] Reassignment: If during task execution it is found that a request cannot be completed on the current server, the system will reallocate the request to another server.

[0139] Successful execution feedback: Once the request is successfully processed, the system will feed back the execution result to the diversion decision engine.

[0140] Return user response: Finally, the processing result is returned to the user, completing the entire request processing flow.

[0141] The entire process is a dynamic adjustment process. The system will continuously adjust and optimize the request processing path based on real-time data and preset rules to ensure efficient and stable performance.

[0142] Monitoring and Adaptive Module: used to monitor the diversion effect and adjust the diversion rules. The workflow of the Monitoring and Adaptive Module is as follows: Figure 7 As shown,

[0143] System status monitoring: The system continuously monitors its operating status and collects health status and performance data.

[0144] Collect abnormal signals: During the monitoring process, the system detects and collects any abnormal signals that may indicate system performance degradation or potential failure.

[0145] Exception handling: Once an exception is detected, the system initiates the exception handling procedure to resolve or mitigate the problem.

[0146] Adaptive Adjustment: The system uses adaptive adjustment to make dynamic adjustments based on monitoring data and the results of exception handling.

[0147] Feedback adjustment strategies to the configuration algorithm module: The adjustment strategies are fed back to the configuration algorithm module so that the system can adjust itself according to these strategies.

[0148] Strategy execution monitoring: The system continues to monitor the effectiveness of strategy execution to ensure that adjustment measures are effective.

[0149] Monitoring result data: All monitoring result data will be recorded for subsequent analysis and optimization.

[0150] The entire process is a cyclic monitoring and adjustment process. The system maintains its optimal performance and stability by continuously collecting data, handling exceptions, adjusting strategies and monitoring results.

[0151] The description of the direction and relative position relationship of the structure in the present invention, such as the description of front, back, left, right, up, and down, does not constitute a limitation of the present invention, but is only for the convenience of description.

Claims

1. A request diversion system based on a configuration algorithm, characterized in that: include Data collection module: used to collect server status information and network request characteristics; Configuration algorithm module: used to adjust the server load distribution strategy; Strategy configuration module: used by backend administrators to customize and test diversion strategies; Diversion decision module: used to make diversion decisions based on the output results of the configuration algorithm; Execution module: used to execute the diversion decision and direct the execution diversion decision request to the server or processing logic; Monitoring and adaptive module: used to monitor the diversion effect and adjust the diversion rules.

2. The request diversion system based on configuration algorithm according to claim 1, characterized in that: The data collection module includes a data collection unit for collecting, storing and monitoring server load, network status and request characteristics, and a request feature extraction unit for extracting request type, size and frequency.

3. The request diversion system based on configuration algorithm according to claim 2, characterized in that: The data collection unit includes a data collection server, a storage server and a monitoring server, which are connected in sequence. The request feature extraction unit includes a feature extractor, which is connected to the collection server and the storage server respectively.

4. The request diversion system based on configuration algorithm according to claim 3, characterized in that: The steps to implement data collection include: Step 1: Data collection: Start the data collection server to collect server load, response time, and real-time traffic data; Step 2: Data storage: The collected data is stored in the storage server. At the same time, the collected data enters the feature extractor, and the data enters the storage server after extraction; Step 3: Data monitoring: The monitoring server monitors the collected and extracted data in real time and feeds the monitoring data back to the data collection server.

5. The request diversion system based on configuration algorithm according to any one of claims 1 to 4, characterized in that: The configuration algorithm module includes a parameter monitoring unit for collecting and real-time monitoring of parameters, a diversion screening and optimization unit for dynamically screening and optimizing the collected data, a diversion performance simulation and evaluation unit for simulating diversion performance and evaluating the diversion effect, a configuration adjustment unit for dynamically adjusting diversion, and an adaptive unit for adjusting the diversion algorithm based on system response and user feedback to realize an adaptive mechanism and continuously optimize the algorithm.

6. The request diversion system based on configuration algorithm according to claim 5, characterized in that: The steps of triage screening include: Step 1: Initialize the parameter weight list: set the initial weight of each parameter, and adjust the weight according to business needs; Step 2: Parameter matching analysis: Match the parameters in the transaction request with each rule and create a matching score for each parameter; Step 3: Rule popularity calculation: Calculate the usage frequency of each rule based on historical data and adjust the rule score accordingly; Step 4: Execution time evaluation: used to calculate and compare the average time required to execute each rule in history; Step 5: Calculate the total score; Step 6: Rule sorting and selection: used to sort all rules according to the calculated total score and select the rule with the highest score as the optimal rule; Step 7: Dynamic update: used to regularly analyze the system processing effect and dynamically adjust the parameter weights.

7. The request diversion system based on configuration algorithm according to claim 6, characterized in that: The diversion screening adopts the dynamic rule matching score algorithm, and the expression of the dynamic rule matching score algorithm is: S j =∑ n i＝1 (M i ×W i )+H j -T j Where: S j is the total score, S j The rule with the highest value is taken as the optimal rule; M i is the matching score of the i-th parameter; W i is the weight score of the i-th parameter; H j is the heat score of the jth rule; T j is the average time required to execute the jth rule.

8. The request diversion system based on configuration algorithm according to claim 5, characterized in that: The simulation algorithm is used for the diversion performance simulation. The expression of the simulation algorithm is: System throughput formula: T = ∑ n i＝1 (1 / T i ) / n Where T is the overall throughput of the system, n is the number of rules, and Ti is the processing time of the i-th diversion rule; Hit rate formula: H = number of successful matches / total number of requests × 100% Where H is the hit rate of the rule; Response time simulation formula: R = ∑ m j＝1 p j × j Where R is the expected average response time, m is the number of diversion points, and p j is the request ratio of the jth diversion point, r j is the average processing time of the jth diversion point. Resource utilization evaluation formula: U = used resources / total resources Among them, U is the resource utilization rate, which is used to measure the efficiency of resource allocation; Load balancing formula: A = ∑ o k＝1 min(L k ) / O Among them, A is the load balancing degree, o is the number of split sets, and L k is the load value in the kth flow distribution set; Comprehensive evaluation effect formula: E = αT + βH + γR + δU + ∈A Among them, E is the comprehensive evaluation result; α is the system throughput weight coefficient,; β is the hit rate weight coefficient; γ is the response time weight coefficient; δ is the resource utilization weight coefficient; ∈ is the load balancing weight coefficient; α, β, γ, δ and ∈ are all used to adjust the relative importance of each indicator in the total score.

9. The request diversion system based on configuration algorithm according to claim 8, characterized in that: The steps of shunting performance simulation include: Step 1: Design of simulation scenarios: Design transaction scenarios and corresponding parameters based on historical data and business development forecasts; Step 2: Parameter input: input configuration parameters, weights and total resources; Step 3: Rule matching and execution: During the simulation process, the rules are matched and executed; Step 4: Calculate performance indicators: Calculate system throughput, hit rate, response time, resource utilization and load balance according to the simulation algorithm; Step 5: Result analysis and optimization: Analyze simulation results to identify performance bottlenecks and improvement points; Step 6: Repeat steps 2 to 6 until the optimal diversion rule configuration is found.

10. The request diversion system based on configuration algorithm according to claim 9, characterized in that: The diversion decision module includes a data parsing unit for extracting and parsing request data, a temporary queue management unit for temporarily storing predetermined priorities, a decision unit for determining and routing priorities, an execution diversion unit for request scheduling, and a monitoring feedback unit for monitoring and providing feedback on execution and response.

Images