Lightweight flow switching and running accompanying system and running accompanying method thereof

Abstract

The invention relates to a lightweight flow switching and running accompanying system and a running accompanying method thereof, and belongs to the field of financial services, the lightweight flow switching and running accompanying system comprises a service front-end module used for intercepting and distributing flow and dynamically adjusting flow distribution according to a system state; the service front running accompanying module is used for consuming the MQ message pushed by the service front, calling the new system module for running accompanying and comparing the difference between the new system module and the old system module; the old system module is used for undertaking the old accounting system function; the new system module is used as a target system for accompanying running and is used for receiving calling of the service front-end accompanying running module so as to trigger a new accounting system; and the difference recording and solving module is used for recording the accounting difference found in the accompanying running process, storing the difference information in a library, and providing difference analysis for marking comparison failure. The method has the advantages that service non-perception switching is achieved, seamless switching between a new core system and an old core system is achieved through the flow interception and distribution function of the service front-end system, and the service does not need to perceive the upgrading process.

Description

Technical Field

[0001] The invention relates to a lightweight traffic switching and accompanying system and an accompanying method thereof. The lightweight traffic switching and accompanying system is based on a service-front credit core system and belongs to the field of financial services. Background Art

[0002] In the traditional practice of upgrading the credit core system, a direct switching method is usually adopted, which involves completely stopping the existing business system in order to deploy and replace the new system. Although this upgrade method is direct, it has many problems and challenges, which are mainly manifested in the following aspects:

[0003] First, the suspension of business upgrades has brought great inconvenience to the daily operations of financial institutions. In the field of financial services, business continuity is crucial, and any form of service interruption may lead to customer dissatisfaction, business loss and even reputation risk. Direct switching to upgrade methods requires suspending all business operations during the upgrade, which not only affects the customer experience, but may also cause compliance issues because many financial transactions and operations have strict time requirements.

[0004] Secondly, the direct switch upgrade method has a high production risk. In the process of launching a new system, various unexpected situations may occur, such as technical failures, data migration errors, system compatibility issues, etc. These problems may cause the new system to fail to go online as planned, or serious problems may occur after going online, requiring an emergency rollback to the old system, which not only increases the complexity of operations, but may also cause business losses.

[0005] Furthermore, the direct switching method has a long process cycle. From the planned upgrade to the full launch of the new system, it is necessary to go through multiple stages such as demand analysis, system design, development, testing, and data migration. Delays may occur at each stage, resulting in a longer cycle for the entire upgrade project. The long upgrade cycle means that financial institutions need to maintain operations on the old system for a longer period of time, which limits business innovation and market response speed.

[0006] In addition, the direct switch upgrade method is also a challenge in terms of cost. In addition to the direct costs of the upgrade process, such as the investment of manpower, material resources and financial resources, there are also indirect costs, such as loss of revenue due to business interruption, customer loss and potential legal risks. These costs add up to make the overall cost of the upgrade project high.

[0007] In summary, there are many problems with the traditional way of upgrading the core credit system, which restricts the business development and technological progress of financial institutions. Therefore, the market urgently needs a new upgrade method that can achieve a smooth transition between the old and new systems without affecting business continuity, reduce upgrade risks, shorten upgrade cycles, and reduce upgrade costs. This new method will help financial institutions better adapt to market changes and improve their competitiveness. Summary of the invention

[0008] In order to overcome the defects of the prior art, the present invention provides a lightweight flow switching and accompanying running system and accompanying running method thereof. The technical solution of the present invention is:

[0009] A lightweight flow switching and accompanying running system and accompanying running method, comprising:

[0010] The service front-end module is used for traffic interception and distribution, and dynamically adjusts traffic distribution according to the system status;

[0011] The service front-end running module is used to consume the MQ messages pushed by the service front-end, call the new system module to run with it, and compare the differences between the new system module and the old system module;

[0012] The old system module is used to take over the functions of the old accounting system;

[0013] The new system module is used as the target system for running with the service provider, and receives the call of the service front-end running module to trigger the new accounting system.

[0014] The difference recording and resolution module is used to record the accounting differences found during the running process, store the difference information, and provide difference analysis when the marking comparison fails.

[0015] The service front-end module further includes:

[0016] Status monitoring unit, used to monitor the operating status of the system in real time, including but not limited to system performance indicators, response time and error rate;

[0017] The decision engine unit dynamically adjusts the traffic distribution strategy based on the data provided by the status monitoring unit and the preset strategy to optimize system performance and user experience;

[0018] Configuration management unit, used to manage and update the configuration information of traffic distribution, ensuring that the service front-end module can perform traffic interception and distribution according to the latest configuration;

[0019] The specific allocation steps of the service front-end module are as follows:

[0020] 1.1 Traffic interception: As the first contact point of the system, the service front-end module intercepts the traffic entering the system to ensure that all requests are first processed by the service front-end module;

[0021] 1.2 Traffic distribution: Based on the current status and configuration of the system, the service front-end module distributes the intercepted traffic to different processing paths, including the old core system and the new core system;

[0022] 1.3 Dynamically adjust traffic distribution: The service front-end module dynamically adjusts the traffic distribution ratio between the old core system and the new core system according to the real-time operating status of the system and preset rules to achieve smooth transition and load balancing;

[0023] The decision engine in the service front-end module uses the following algorithm to dynamically adjust the traffic distribution: new =T old +α×(P new -P old ), where Tnew represents the traffic allocation ratio of the new core system, Told represents the traffic allocation ratio of the old core system, Pnew and Pold represent the performance indicators of the new and old core systems respectively, and α is the adjustment coefficient, which is used to dynamically adjust the traffic allocation according to the performance difference;

[0024] The adjustment factor α is calculated according to the following formula: Where W represents the weight vector, X represents the state feature vector, and b represents the bias term. This formula is a logistic regression model, which is used to dynamically calculate the adjustment coefficient according to the system state characteristics.

[0025] The service pre-running module further includes:

[0026] A message parsing unit, used to parse the MQ message pushed by the service front-end module and extract the request message and the response message;

[0027] A calling unit, used for calling the new system module and the old system module according to the parsed request message;

[0028] A comparison unit is used to compare the response messages of the new system module and the old system module, and record and analyze the differences;

[0029] The accompanying steps of the service front-end accompanying module are as follows:

[0030] 2.1 Message consumption: The service front-end running module receives the MQ message pushed by the service front-end module; 2.2 Calling the new system module: The service front-end running module parses the MQ message and calls the new system module for processing;

[0031] 2.3 Comparison of differences: The service pre-running module compares the differences in processing results between the new system module and the old system module;

[0032] The comparison unit uses the following algorithm to compare differences:

[0033]

[0034] Where D represents the difference metric, Rnew(i) and Rold(i) represent the response values ​​of the new and old system modules on the i-th field, respectively, and n is the total number of fields in the response message;

[0035] The service pre-running module uses the following formula to calculate the difference weight at the field level: Wherein, W(i) represents the difference weight of the i-th field, Wbase represents the basic weight vector, F(i) represents the feature vector of the i-th field, and b represents the bias term. The formula is a logistic regression model, which is used to dynamically calculate the difference weight according to the field characteristics; the comparison unit adjusts the calculation method of the difference metric D according to the difference weight W(i): Where D′ represents the weighted difference measure.

[0036] The old system module further includes:

[0037] Request processing unit, used to receive requests and process them according to preset business rules; Data accounting unit: used to perform data accounting, including but not limited to calculating, verifying and recording business data;

[0038] Response generation unit: used to generate a response message according to the processing result and return it to the service front-end module;

[0039] The data calculation unit uses the following algorithm to perform data calculation:

[0040] V=f(D;P), where V represents the calculation result, D represents the input data, P represents the calculation parameter, and f represents the calculation function, which is defined according to the business rules of the old accounting system; the calculation parameter P is dynamically adjusted according to the following formula:

[0041] P new =P old +β×(GP old ), where Pnew represents the new accounting parameter, Pold represents the old accounting parameter, G represents the target parameter value, and β is the adjustment coefficient, which is used to dynamically adjust the accounting parameter according to business needs and market changes.

[0042] The new system modules include

[0043] A call receiving unit, used for receiving a call request of a service front-end accompanying module;

[0044] The accounting logic unit is used to execute new accounting logic, including but not limited to calculating, verifying and recording business data;

[0045] A result generating unit is used to generate a response message according to the processing result of the accounting logic unit and return it to the service front-end accompanying module;

[0046] The accounting logic unit uses the following algorithm to calculate business data:

[0047] V new =g(D;Q), where Vnew represents the accounting result of the new accounting system, D represents the input data, Q represents the new accounting parameter, and g represents the new accounting function, which is defined according to the business rules of the new accounting system; the new accounting parameter Q is dynamically adjusted according to the following formula: Q new =Q old +γ×(HQ old ), Qnew represents the new accounting parameter, Qold represents the old accounting parameter, H represents the target parameter value, and γ is the adjustment coefficient, which is used to dynamically adjust the accounting parameters according to business needs and market changes.

[0048] The difference recording and solving module further comprises:

[0049] Detection unit, used to detect the difference in results between the new and old accounting systems;

[0050] A recording unit, for storing the detected difference records in a database;

[0051] Analytical unit, used to analyze the causes of differences and provide solutions;

[0052] The detection unit uses the following algorithm to perform difference detection:

[0053] D i =|R new (i)-R old (i)|; wherein Di represents the difference of the i-th calculation result, Rnew(i) and Rold(i) represent the results of the new and old calculation systems at the i-th calculation point respectively; the detection unit also includes a difference threshold judgment, which is judged by the following formula:

[0054] Among them, Ti represents the mark of whether the i-th calculation point exceeds the difference threshold θ;

[0055] The recording unit stores the difference information in the following structured format:

[0056] Record = {D i , T i , R new (i), R old(i), Timestamp}; where Record represents a difference record, including the difference value Di, the threshold judgment result Ti, the new calculation result Rnew(i), the old calculation result Rold(i), and the timestamp Timestamp;

[0057] The analysis unit uses the following algorithm to perform difference analysis:

[0058] C = A (D, H), where C represents the cause of the difference, D represents the difference data set, H represents the historical difference case library, and A represents the analysis function, which performs pattern matching and cause analysis based on historical cases and current difference data; the analysis function A uses a machine learning method, specifically: in, represents the predicted difference cause, and P(C=c|D,H;θ) represents the posterior probability of the difference cause c given the difference data set D, the historical difference case library H and the model parameter θ.

[0059] A running-accompanying method based on the lightweight flow switching and running-accompanying system comprises the following steps:

[0060] Traffic interception: As the first contact point of the system, the service front-end module intercepts the traffic entering the system to ensure that all requests are first processed by the service front-end module;

[0061] Traffic distribution: Based on the current status and configuration of the system, the service front-end module distributes the intercepted traffic to different processing paths, including old system modules and new system modules;

[0062] Dynamically adjust traffic distribution: The service front-end module dynamically adjusts the traffic distribution ratio between the old system modules and the new system modules according to the real-time operating status of the system and preset rules to achieve smooth transition and load balancing.

[0063] The advantages of the present invention are:

[0064] Realize business-unaware switching: Through the traffic interception and distribution function of the service front-end system, seamless switching between the old and new core systems is achieved, and the business does not need to be aware of the upgrade process.

[0065] Reduce upgrade risks: The accompanying mechanism can be used to discover and resolve differences between the old and new accounting systems in real time, ensure the accuracy and stability of the new core system, and reduce risks during the upgrade process.

[0066] Shorten the upgrade cycle: reduce upgrade delays caused by business suspension, rollback, etc., and improve upgrade efficiency.

[0067] Improve system availability: Timely detect and handle potential risks through monitoring and alarm mechanisms to ensure high system availability.

[0068] In summary, the lightweight traffic switching and accompanying system and its accompanying method, through their innovative design and implementation, provide financial institutions with an efficient, reliable and cost-effective system upgrade solution, which helps financial institutions remain competitive in a rapidly changing market environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is the main structure frame diagram of the present invention. DETAILED DESCRIPTION

[0070] The present invention will be further described below in conjunction with specific embodiments, and the advantages and features of the present invention will become clearer as the description proceeds. However, these embodiments are exemplary only and do not constitute any limitation to the scope of the present invention. It should be understood by those skilled in the art that the details and forms of the technical solution of the present invention may be modified or replaced without departing from the spirit and scope of the present invention, but these modifications and replacements all fall within the scope of protection of the present invention.

[0071] See also Figure 1 The present invention relates to a lightweight flow switching and accompanying running system, comprising:

[0072] Service front-end module 1, used for traffic interception and distribution, dynamically adjusting traffic distribution according to system status;

[0073] Service pre-running module 2 is used to consume the MQ message pushed by the service pre-running, call the new system module to run with it, and compare the differences between the new system module and the old system module;

[0074] Old system module 3, used to take over the functions of the old accounting system;

[0075] New system module 4, used as the target system for accompanying running, receiving the call of the service front-end accompanying running module to trigger the new accounting system;

[0076] The difference recording and resolution module 5 is used to record the accounting differences found during the accompanying process, store the difference information in the database, and provide difference analysis when the marking comparison fails.

[0077] Based on the settings of the above modules, the following are achieved:

[0078] Seamless service transition: Through the traffic interception and distribution function of the service front-end module 1, the system can achieve a smooth transition between the old and new systems without interrupting the service, ensuring business continuity.

[0079] Risk control: During the parallel operation of the old and new systems, the service pre-running module 2 can monitor and compare the differences between the two systems in real time, discover problems in time, and thus reduce the risk of system upgrades.

[0080] Performance optimization: Real-time monitoring of system status and dynamic traffic allocation adjustment enable the system to optimize resource allocation according to actual operating conditions and improve the overall performance of the system.

[0081] Flexibility and adaptability: As the target system for accompanying the running, the new system module 4 can dynamically adjust the accounting parameters according to the feedback from the service pre-accompanying module 2, making the system more flexible and adaptable to business changes.

[0082] Data consistency assurance: The difference recording and resolution module 5 can record and analyze the differences in the accounting process, ensure the consistency and accuracy of the data, and reduce data errors caused by system upgrades.

[0083] Fast fault recovery: When a problem occurs in the new system module 4, the old system module 3 can quickly take over the business to ensure the stability and reliability of the system.

[0084] Intelligent decision-making: Using advanced algorithms and models, the system can intelligently make decisions based on historical data and current status, improving the accuracy and efficiency of decision-making.

[0085] Automation and efficiency improvement: Automated traffic management and discrepancy detection reduce manual intervention and improve operational efficiency and processing speed.

[0086] The service front-end module 1 further comprises:

[0087] Status monitoring unit, used to monitor the operating status of the system in real time, including but not limited to system performance indicators, response time and error rate;

[0088] The decision engine unit dynamically adjusts the traffic distribution strategy based on the data provided by the status monitoring unit and the preset strategy to optimize system performance and user experience;

[0089] Configuration management unit, used to manage and update the configuration information of traffic distribution, ensuring that the service front-end module can perform traffic interception and distribution according to the latest configuration;

[0090] The specific allocation steps of the service front-end module 1 are as follows:

[0091] 1.1 Traffic interception: As the first contact point of the system, the service front-end module intercepts the traffic entering the system to ensure that all requests are first processed by the service front-end module;

[0092] 1.2 Traffic distribution: According to the current status and configuration of the system, the service front-end module distributes the intercepted traffic to different processing paths, including the old core system and the new core system; 1.3 Dynamically adjust traffic distribution: The service front-end module dynamically adjusts the traffic distribution ratio between the old core system and the new core system according to the real-time operating status of the system and the preset rules to achieve smooth transition and load balancing;

[0093] The decision engine in the service front-end module uses the following algorithm to dynamically adjust the traffic distribution: new =T old +α×(P new -P old ), where Tnew represents the traffic allocation ratio of the new core system, Told represents the traffic allocation ratio of the old core system, Pnew and Pold represent the performance indicators of the new and old core systems respectively, and α is the adjustment coefficient, which is used to dynamically adjust the traffic allocation according to the performance difference;

[0094] The adjustment factor α is calculated according to the following formula: Where W represents the weight vector, X represents the state feature vector, and b represents the bias term. This formula is a logistic regression model, which is used to dynamically calculate the adjustment coefficient according to the system state characteristics.

[0095] The design and implementation of the service front-end module 1 has the following advantages:

[0096] Real-time performance monitoring: The condition monitoring unit can monitor the key performance indicators of the system in real time, ensure the transparency and immediacy of the system operation status, and provide accurate data support for decision-making.

[0097] Dynamic traffic management: The decision engine unit dynamically adjusts traffic distribution based on real-time data and preset strategies, improving the system's responsiveness and flexibility and optimizing resource allocation based on actual load conditions.

[0098] Optimize user experience: By dynamically adjusting traffic distribution, the system can balance load, reduce response time, and increase throughput, thereby improving user experience.

[0099] Improved system performance: Dynamically adjusting traffic distribution helps avoid system overload and ensures system stability and performance under high load conditions.

[0100] Flexible configuration management: The configuration management unit enables the system to quickly adapt to changes and adjust traffic distribution strategies by updating configuration information without system downtime or large-scale modifications.

[0101] Smooth transition: By dynamically adjusting the traffic distribution ratio, the system can achieve a smooth transition between the old and new systems, reducing business interruptions caused by system upgrades.

[0102] Load balancing: The service front-end module 1 can achieve load balancing according to the real-time operating status to improve the reliability and availability of the system.

[0103] The service pre-running module 2 further includes:

[0104] A message parsing unit, used to parse the MQ message pushed by the service front-end module and extract the request message and the response message;

[0105] A calling unit, used for calling the new system module and the old system module according to the parsed request message;

[0106] A comparison unit is used to compare the response messages of the new system module and the old system module, and record and analyze the differences;

[0107] The accompanying steps of the service front-end accompanying module are as follows:

[0108] 2.1 Message consumption: The service front-end running module receives the MQ message pushed by the service front-end module; 2.2 Calling the new system module: The service front-end running module parses the MQ message and calls the new system module for processing;

[0109] 2.3 Comparison of differences: The service pre-running module compares the differences in processing results between the new system module and the old system module;

[0110] The comparison unit uses the following algorithm to compare differences:

[0111]

[0112] Where D represents the difference metric, Rnew(i) and Rold(i) represent the response values ​​of the new and old system modules on the i-th field, respectively, and n is the total number of fields in the response message;

[0113] The service pre-running module uses the following formula to calculate the difference weight at the field level: Wherein, W(i) represents the difference weight of the i-th field, Wbase represents the basic weight vector, F(i) represents the feature vector of the i-th field, and b represents the bias term. The formula is a logistic regression model, which is used to dynamically calculate the difference weight according to the field characteristics; the comparison unit adjusts the calculation method of the difference metric D according to the difference weight W(i): Where D′ represents the weighted difference measure.

[0114] This solution has the following advantages through the design and implementation of the service front-end running module 2:

[0115] Efficient message processing: The message parsing unit can efficiently parse MQ messages, ensuring fast extraction of request and response messages, and improving the response speed of the system.

[0116] Flexible system calls: The calling unit flexibly calls old and new system modules according to the parsed request message, so that the system can process business in parallel, increasing the flexibility of the system.

[0117] Accurate difference comparison: The comparison unit compares the response messages of the new and old system modules through precise algorithms to ensure the accuracy of difference analysis.

[0118] Enhanced data consistency: By comparing the differences between the processing results of the new and old systems, the system can promptly detect and correct data inconsistencies, thereby enhancing data consistency.

[0119] Intelligent weight adjustment: The logistic regression model is used to calculate the difference weight at the field level, so that the system can intelligently adjust the calculation of the difference metric according to the importance of the field, thereby improving the accuracy of the difference analysis.

[0120] Optimized resource allocation: The weighted difference measurement enables the system to optimize resource allocation according to business needs and field characteristics, improving resource utilization efficiency.

[0121] Real-time business monitoring: The service pre-running module 2 can monitor the operating status of the new and old systems in real time, discover problems in time and make adjustments, thereby improving the stability of the system.

[0122] The old system module 3 further comprises:

[0123] Request processing unit, used to receive requests and process them according to preset business rules; Data accounting unit: used to perform data accounting, including but not limited to calculating, verifying and recording business data;

[0124] Response generation unit: used to generate a response message according to the processing result and return it to the service front-end module;

[0125] The data calculation unit uses the following algorithm to perform data calculation:

[0126] V=f(D;P), where V represents the calculation result, D represents the input data, P represents the calculation parameter, and f represents the calculation function, which is defined according to the business rules of the old accounting system; the calculation parameter P is dynamically adjusted according to the following formula:

[0127] P new =P old +β×(GP old ), where Pnew represents the new accounting parameter, Pold represents the old accounting parameter, G represents the target parameter value, and β is the adjustment coefficient, which is used to dynamically adjust the accounting parameter according to business needs and market changes.

[0128] This solution is designed and implemented through the old system module 3, and has the following advantages:

[0129] Business rule compliance: The request processing unit ensures that all received requests are processed according to preset business rules, ensuring the compliance and consistency of business operations.

[0130] Accuracy of data processing: The data accounting unit is responsible for performing data accounting, including calculating, verifying and recording business data, ensuring the accuracy and completeness of data processing.

[0131] Response efficiency: The response generation unit quickly generates a response message based on the processing results and returns it to the service front-end module, improving the system's response efficiency and customer satisfaction.

[0132] Accounting automation: Automating the data accounting process through algorithms reduces human intervention, reduces operational errors and improves processing speed.

[0133] Dynamic parameter adjustment: The dynamic adjustment capability of the accounting parameter P enables the system to flexibly adapt to business needs and market changes, improving the adaptability and flexibility of the system.

[0134] Business continuity: The old system module 3 continues to handle business while the new system module 4 is running, ensuring business continuity and reducing the impact of system upgrades on business.

[0135] Risk management: When problems occur in the new system module 4, the old system module 3 can be used as a backup plan to reduce the risk of system upgrades.

[0136] The new system module 4 includes

[0137] A call receiving unit, used for receiving a call request of a service front-end accompanying module;

[0138] The accounting logic unit is used to execute new accounting logic, including but not limited to calculating, verifying and recording business data;

[0139] A result generating unit is used to generate a response message according to the processing result of the accounting logic unit and return it to the service front-end accompanying module;

[0140] The accounting logic unit uses the following algorithm to calculate business data:

[0141] V new =g(D;Q), where Vnew represents the accounting result of the new accounting system, D represents the input data, Q represents the new accounting parameter, and g represents the new accounting function, which is defined according to the business rules of the new accounting system; the new accounting parameter Q is dynamically adjusted according to the following formula: Q new =Q old +γ×(HQold ), Qnew represents the new accounting parameter, Qold represents the old accounting parameter, H represents the target parameter value, and γ is the adjustment coefficient, which is used to dynamically adjust the accounting parameters according to business needs and market changes.

[0142] This solution has the following advantages through the design and implementation of the new system module 4:

[0143] Efficient call processing: The call receiving unit can efficiently receive call requests from the service front-end running module, ensuring that the new system module can respond to external requests quickly.

[0144] Modernization of accounting logic: The accounting logic unit executes new accounting logic, enabling the system to adapt to new business rules and market changes, and improve accounting accuracy and efficiency.

[0145] Response automation: The result generation unit automatically generates response messages based on the processing results of the accounting logic unit, improving the system's processing speed and response capabilities.

[0146] Dynamic adjustment of accounting parameters: The dynamic adjustment capability of the new accounting parameter Q enables the system to flexibly adapt according to business needs and market changes, improving the adaptability and flexibility of the system.

[0147] Business innovation support: The new system module 4 can support new business processes and innovations, providing financial institutions with more business development opportunities.

[0148] The difference recording and solving module 5 further comprises:

[0149] Detection unit, used to detect the difference in results between the new and old accounting systems;

[0150] A recording unit, for storing the detected difference records in a database;

[0151] Analytical unit, used to analyze the causes of differences and provide solutions;

[0152] The detection unit uses the following algorithm to perform difference detection:

[0153] D i =|R new (i)-R old (i)|; wherein Di represents the difference of the i-th calculation result, Rnew(i) and Rold(i) represent the results of the new and old calculation systems at the i-th calculation point respectively; the detection unit also includes a difference threshold judgment, which is judged by the following formula:

[0154] Among them, Ti represents the mark of whether the i-th calculation point exceeds the difference threshold θ;

[0155] The recording unit stores the difference information in the following structured format:

[0156] Record = {D i , T i , R new (i), R old (i), Timestamp}; where Record represents a difference record, including the difference value Di, the threshold judgment result Ti, the new calculation result Rnew(i), the old calculation result Rold(i), and the timestamp Timestamp;

[0157] The analysis unit uses the following algorithm to perform difference analysis:

[0158] C = A (D, H), where C represents the cause of the difference, D represents the difference data set, H represents the historical difference case library, and A represents the analysis function, which performs pattern matching and cause analysis based on historical cases and current difference data; the analysis function A uses a machine learning method, specifically: in, represents the predicted difference cause, and P(C=c|D,H;θ) represents the posterior probability of the difference cause c given the difference data set D, the historical difference case library H and the model parameter θ.

[0159] The design and implementation of the difference recording and resolution module 5 has the following advantages:

[0160] Accurate difference detection: The detection unit can accurately detect the difference in results between the new and old accounting systems, ensuring the accuracy of the difference analysis.

[0161] Data persistence: The recording unit stores the detected difference records in the database, realizing persistent storage of data, which is convenient for subsequent analysis and auditing.

[0162] Structured storage: Using a structured format to store difference information improves the readability and queryability of data, making it easier to manage and retrieve.

[0163] Real-time: The detection unit can detect differences in real time and find problems in time, reducing the delay in problem discovery.

[0164] Threshold judgment: Through difference threshold judgment, the system can identify important differences and avoid overreacting to unimportant fluctuations.

[0165] Historical case study: The analysis unit uses the historical difference case library to perform pattern matching and cause analysis, which improves the depth and breadth of problem analysis.

[0166] Intelligent analysis: The use of machine learning methods for differential analysis improves the intelligence level of analysis and can predict and identify the causes of complex problems.

[0167] The present invention also relates to a running-accompanying method based on the lightweight flow switching and running-accompanying system, comprising the following steps:

[0168] Traffic interception: As the first contact point of the system, the service front-end module intercepts the traffic entering the system to ensure that all requests are first processed by the service front-end module;

[0169] Traffic distribution: Based on the current status and configuration of the system, the service front-end module distributes the intercepted traffic to different processing paths, including old system modules and new system modules;

[0170] Dynamically adjust traffic distribution: The service front-end module dynamically adjusts the traffic distribution ratio between the old system modules and the new system modules according to the real-time operating status of the system and preset rules to achieve smooth transition and load balancing.

[0171] The accompanying running method based on the lightweight flow switching and accompanying running system of the present invention has the following advantages:

[0172] Seamless business transition: Through traffic interception and distribution of the service front-end module, this method can achieve a smooth transition between the old and new systems without interrupting business, ensuring business continuity.

[0173] Load balancing: Dynamically adjusting the traffic distribution steps enables the system to balance the load according to the real-time operating status, optimize resource usage, and improve the overall performance and response speed of the system.

[0174] Reduced upgrade risk: By gradually transferring traffic from the old system to the new system, this approach reduces the risk of a direct switch, allowing problems with the new system to be discovered and resolved without affecting overall business.

[0175] Improve system availability: During the period when the new and old systems run in parallel, if there is a problem with the new system, the old system can continue to process business, thereby improving the availability of the system.

[0176] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A lightweight flow switching and accompanying running system, characterized in that: include: The service front-end module is used for traffic interception and distribution, and dynamically adjusts traffic distribution according to the system status; The service front-end running module is used to consume the MQ messages pushed by the service front-end, call the new system module to run with it, and compare the differences between the new system module and the old system module; The old system module is used to take over the functions of the old accounting system; The new system module is used as the target system for running with the service provider, and receives the call of the service front-end running module to trigger the new accounting system. The difference recording and resolution module is used to record the accounting differences found during the running process, store the difference information, and provide difference analysis when the marking comparison fails.

2. The lightweight flow switching and accompanying running system according to claim 1 is characterized in that: The service front-end module further includes: Status monitoring unit, used to monitor the operating status of the system in real time, including but not limited to system performance indicators, response time and error rate; The decision engine unit dynamically adjusts the traffic distribution strategy based on the data provided by the status monitoring unit and the preset strategy to optimize system performance and user experience; Configuration management unit, used to manage and update the configuration information of traffic distribution, ensuring that the service front-end module can perform traffic interception and distribution according to the latest configuration; The specific allocation steps of the service front-end module are as follows: 1.1 Traffic interception: As the first contact point of the system, the service front-end module intercepts the traffic entering the system to ensure that all requests are first processed by the service front-end module; 1.2 Traffic distribution: Based on the current status and configuration of the system, the service front-end module distributes the intercepted traffic to different processing paths, including the old core system and the new core system; 1.3 Dynamically adjust traffic distribution: The service front-end module dynamically adjusts the traffic distribution ratio between the old core system and the new core system according to the real-time operating status of the system and preset rules to achieve smooth transition and load balancing; The decision engine in the service front-end module uses the following algorithm to dynamically adjust the traffic distribution: new =T old +α×(P new -P old ), where Tnew represents the traffic allocation ratio of the new core system, Told represents the traffic allocation ratio of the old core system, Pnew and Pold represent the performance indicators of the new and old core systems respectively, and α is the adjustment coefficient, which is used to dynamically adjust the traffic allocation according to the performance difference; The adjustment factor α is calculated according to the following formula: Where W represents the weight vector, X represents the state feature vector, and b represents the bias term. This formula is a logistic regression model, which is used to dynamically calculate the adjustment coefficient according to the system state characteristics.

3. The lightweight flow switching and accompanying running system according to claim 1 or 2, characterized in that: The service pre-running module further includes: A message parsing unit, used to parse the MQ message pushed by the service front-end module and extract the request message and the response message; A calling unit, used for calling the new system module and the old system module according to the parsed request message; A comparison unit is used to compare the response messages of the new system module and the old system module, and record and analyze the differences; The accompanying steps of the service front-end accompanying module are as follows: 2.1 Message consumption: The service front-end running module receives the MQ message pushed by the service front-end module; 2.2 Calling the new system module: The service front-end module parses the MQ message and calls the new system module for processing; 2.3 Comparison of differences: The service pre-running module compares the differences in processing results between the new system module and the old system module; The comparison unit uses the following algorithm to compare differences: Where D represents the difference metric, Rnew(i) and Rold(i) represent the response values ​​of the new and old system modules on the i-th field, respectively, and n is the total number of fields in the response message; The service pre-running module uses the following formula to calculate the difference weight at the field level: Among them, W(i) represents the difference weight of the i-th field, Wbase represents the basic weight vector, F(i) represents the feature vector of the i-th field, and b represents the bias term. This formula is a logistic regression model, which is used to dynamically calculate the difference weight according to the field characteristics; The comparison unit adjusts the calculation method of the difference metric D according to the difference weight W(i): Where D′ represents the weighted difference measure.

4. The lightweight flow switching and accompanying running system according to claim 3 is characterized in that: The old system module further includes: A request processing unit, used to receive requests and process them according to preset business rules; Data accounting unit: used to perform data accounting, including but not limited to calculating, verifying and recording business data; Response generation unit: used to generate a response message according to the processing result and return it to the service front-end module; The data calculation unit uses the following algorithm to perform data calculation: V=f(D;P), where V represents the calculation result, D represents the input data, P represents the calculation parameter, and f represents the calculation function, which is defined according to the business rules of the old accounting system; the calculation parameter P is dynamically adjusted according to the following formula: P new =P old +β×(GP old ), where Pnew represents the new accounting parameter, Pold represents the old accounting parameter, G represents the target parameter value, and β is the adjustment coefficient, which is used to dynamically adjust the accounting parameter according to business needs and market changes.

5. The lightweight flow switching and accompanying running system according to claim 4 is characterized in that: The new system modules include A call receiving unit, used for receiving a call request of a service front-end accompanying module; The accounting logic unit is used to execute new accounting logic, including but not limited to calculating, verifying and recording business data; A result generating unit is used to generate a response message according to the processing result of the accounting logic unit and return it to the service front-end accompanying module; The accounting logic unit uses the following algorithm to calculate business data: V new =g(D;Q), where Vnew represents the accounting result of the new accounting system, D represents the input data, Q represents the new accounting parameter, and g represents the new accounting function, which is defined according to the business rules of the new accounting system; the new accounting parameter Q is dynamically adjusted according to the following formula: Q new =Q old +γ×(HQ old ), Qnew represents the new accounting parameter, Qold represents the old accounting parameter, H represents the target parameter value, and γ is the adjustment coefficient, which is used to dynamically adjust the accounting parameters according to business needs and market changes.

6. The lightweight flow switching and accompanying running system according to claim 5 is characterized in that: The difference recording and solving module further comprises: Detection unit, used to detect the difference in results between the new and old accounting systems; A recording unit, for storing the detected difference records in a database; Analytical unit, used to analyze the causes of differences and provide solutions; The detection unit uses the following algorithm to perform difference detection: D i =|R new (i)-R old (i)|; wherein Di represents the difference of the i-th calculation result, Rnew(i) and Rold(i) represent the results of the new and old calculation systems at the i-th calculation point respectively; the detection unit also includes a difference threshold judgment, which is judged by the following formula: Among them, Ti represents the mark of whether the i-th calculation point exceeds the difference threshold θ; The recording unit stores the difference information in the following structured format: Record = {D i , T i , R new (i), R old (i), Timestamp}; where Record represents a difference record, including the difference value Di, the threshold judgment result Ti, the new calculation result Rnew(i), the old calculation result Rold(i), and the timestamp Timestamp; The analysis unit uses the following algorithm to perform difference analysis: C = A (D, H), where C represents the cause of the difference, D represents the difference data set, H represents the historical difference case library, and A represents the analysis function, which performs pattern matching and cause analysis based on historical cases and current difference data; the analysis function A uses a machine learning method, specifically: in, represents the predicted difference cause, and P(C=c|D,H;θ) represents the posterior probability of the difference cause c given the difference data set D, the historical difference case library H and the model parameter θ.

7. A running-accompanying method based on the lightweight flow switching and running-accompanying system according to any one of claims 1 to 6, characterized in that: The following steps are involved: Traffic interception: As the first contact point of the system, the service front-end module intercepts the traffic entering the system to ensure that all requests are first processed by the service front-end module; Traffic distribution: Based on the current status and configuration of the system, the service front-end module distributes the intercepted traffic to different processing paths, including old system modules and new system modules; Dynamically adjust traffic distribution: The service front-end module dynamically adjusts the traffic distribution ratio between the old system modules and the new system modules according to the real-time operating status of the system and preset rules to achieve smooth transition and load balancing.

Images