A computer system and method for optimizing service

By integrating traffic scheduling, resource management, fault-tolerant recovery and security protection modules in the service system, the problems of insufficient service quality, resource utilization, fault recovery and security in traditional systems are solved, and efficient resource optimization and rapid fault recovery are achieved. and all-round security protection, improving user experience and system availability.

CN118540275BActive Publication Date: 2024-11-19BEIJING TIANYUN NETWORK SECURITY TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410760053.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-13
Publication Date
2024-11-19
Estimated Expiration
2044-06-13

AI Technical Summary

Technical Problem

Traditional service systems face problems such as unstable service quality, low resource utilization efficiency, weak fault recovery capabilities, insufficient system scalability and flexibility, and insufficient security protection and privacy protection, especially in high concurrency and large traffic environments. Availability.

Method used

A service-optimized computer system was designed, including traffic scheduling unit, resource management unit, fault-tolerant recovery unit, microservice management unit, business decision-making unit and security protection unit, through real-time monitoring, machine learning, distributed tracking and defense in-depth, etc. Technology to realize load balancing, resource optimization, automatic failure recovery, full life cycle management, data-driven decision-making and multi-level security protection.

Benefits of technology

It improves the overall performance and user satisfaction of the system, dynamically adjusts resource allocation, achieves rapid failure recovery and all-round security protection, reduces system failure time and security risks, adapts to business changes and optimizes service quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118540275B_ABST
    Figure CN118540275B_ABST
Patent Text Reader

Abstract

The present invention provides a computer system and method for service optimization, including request reception and parsing, service discovery and request routing, resource management and optimization, fault tolerance and exception handling, service monitoring and governance, business decision support, security protection measures, and continuous optimization and feedback. Through dynamic routing and load balancing technology, efficient request processing is achieved; through real-time resource management and automatic adjustment, resource utilization is improved; through fault monitoring and automatic recovery, high system availability is ensured; through data-driven decision support, business processes and user experience are optimized; through multi-level security measures, system and data security is guaranteed. The present invention is suitable for service environments that require high reliability and rapid response, such as cloud computing and big data processing, and provides an efficient, secure, and scalable service optimization solution.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of telecommunication technology, and in particular to a service optimization computer system and method thereof. Background Art

[0002] With the rapid development of information technology and Internet services, the service quality, resource allocation, system stability and security of data centers and cloud services have become important directions for research and improvement. Traditional service systems often face many challenges, including unstable service quality, low resource utilization efficiency, weak system fault recovery capabilities, and insufficient security protection capabilities.

[0003] In a high-concurrency and high-traffic operating environment, performance indicators such as the response time and throughput of the service system are often affected, resulting in a decline in user experience. In addition, traditional resource management strategies often cannot effectively cope with dynamically changing load demands, resulting in some servers being overloaded while others are idle, causing a waste of resources. At the same time, when the service system encounters a failure, the lack of effective fault detection and automatic recovery mechanisms may lead to extended service interruption time and affect the overall service availability. In addition, many service systems fail to fully consider flexibility and scalability when they are designed, resulting in the system being difficult to quickly adapt when business needs change, increasing the complexity and cost of system upgrades and expansions. In terms of security, service systems often face threats from network attacks, such as distributed denial of service (DDoS) attacks, data leaks, etc., and there are also deficiencies in protecting user data privacy.

[0004] Therefore, there is an urgent need for a new service optimization computer system and method thereof, which can intelligently handle multiple issues such as service quality, resource management, fault recovery and security protection, so as to improve the overall performance of the system and user satisfaction. Summary of the invention

[0005] The present invention proposes a computer system and method for service optimization. The system solves the problems of unstable service quality, low resource utilization, weak fault recovery capability, insufficient system scalability and flexibility, and insufficient security and privacy protection in the prior art. To achieve these purposes, the present invention adopts the following technical solutions, and the system is composed of:

[0006] 1. Traffic Scheduling Unit

[0007] The traffic scheduling unit is responsible for receiving client requests and dynamically routing the requests to the optimal service instance based on the characteristics of the request and the real-time performance of the service to achieve load balancing and traffic optimization. The unit dynamically adjusts the routing strategy of the request by monitoring the health status and load level of the service instance in real time to ensure that the request is processed promptly and reliably. At the same time, the unit also has traffic control and service protection functions, which can limit and shape the inlet traffic, and perform fuse and downgrade processing when the service is abnormal.

[0008] 2. Resource Management Unit

[0009] The resource management unit is responsible for resource allocation and scheduling optimization. It formulates a global resource allocation strategy based on the resource demand characteristics and priorities of the service, and continuously optimizes the efficiency of resource use during operation. Specifically, the responsibilities of this unit include: resource demand prediction, predicting the resource consumption trend of services at different time granularities through machine learning algorithms; resource capacity planning, determining the overall resource configuration plan based on resource demand prediction and SLA requirements; resource scheduling optimization, dynamically adjusting the resource allocation ratio between services to minimize resource waste while meeting service performance; resource elastic scaling, automatically adjusting resource configuration according to load changes and fault conditions to achieve elastic adaptation of services.

[0010] 3. Fault-tolerant recovery unit

[0011] The fault-tolerant recovery unit is responsible for improving the reliability and recoverability of services, ensuring that services can quickly self-heal and recover when local failures occur. The unit uses distributed tracking and monitoring technology to collect the health status and key indicators of services in real time, and quickly discovers and locates faults based on anomaly detection and root cause analysis algorithms. After a failure occurs, the unit automatically triggers the failover and recovery process, takes the failed instance offline and migrates traffic to the healthy instance, while triggering the restart and repair of the failed instance.

[0012] 4. Microservice Management Unit

[0013] The microservice management unit provides one-stop microservice development, deployment, operation and maintenance, and governance capabilities. The unit is based on the Service Mesh and CloudNative architecture to achieve full lifecycle management of services. The main functions include: service registration and discovery, providing a unified service registration center to dynamically manage the address information of services; configuration management, providing a unified configuration center to manage various configuration parameters of services; API gateway, providing a unified service access portal to achieve protocol conversion, security authentication, flow control, etc.; service monitoring, providing a full range of service monitoring capabilities, including call chain tracking, indicator measurement, log collection, etc.; service governance, providing service flow limiting, degradation, authentication, authorization and other governance functions.

[0014] 5. Business Decision-Making Unit

[0015] The business decision-making unit uses big data analysis and machine learning technology to provide data support and intelligent suggestions for business decision-making and process optimization. The unit aggregates various business data (such as user behavior, transaction records, operational indicators, etc.), establishes a unified data warehouse and data mart, and deeply mines data value based on data mining, OLAP analysis, visualization and other technologies to discover business problems and optimization opportunities.

[0016] 6. Safety protection unit

[0017] The security protection unit provides comprehensive security protection capabilities for the entire service system to ensure the confidentiality, integrity and availability of the service. The unit adopts a defense-in-depth strategy to build security lines at different levels, including the network layer, host layer, application layer, and data layer. Typical security measures include: highly available DDoS protection to resist large-volume attacks; Web application firewall (WAF) to filter application layer attacks such as SQL injection and XSS; API authentication and authorization to control API access rights in a fine-grained manner; data encryption and desensitization to prevent sensitive data leakage; security audits to record and analyze various security incidents.

[0018] Further, the workflow of the traffic scheduling unit is as follows:

[0019] A. Request reception: The traffic scheduling unit receives HTTP / HTTPS requests from clients through the standard API interface.

[0020] B. Request parsing: The unit parses the request, extracts key attributes (method, path, header, parameters, etc.), and converts them into internal data structures.

[0021] C. Service discovery: Obtain a list of currently available service instances from the service registry, including the instance address, version number, metadata, etc.

[0022] D. Load balancing: Apply load balancing algorithm to select appropriate service instances.

[0023] E. Health check: Regularly perform heartbeat and status detection on the instance and update the instance status.

[0024] F. Flow control: Implement the token bucket algorithm to control the request flow. The formula is as follows:

[0025] B(t)=min(B(t-1)+r·Δt,N)

[0026] in:

[0027] B(t) is the number of tokens at time t, which represents the number of tokens at any given time t. This is the number of tokens in the bucket that can be used to process requests. If there are no tokens in the bucket when a request arrives (i.e. (B(t) < 1), the request will be delayed or rejected directly.

[0028] r: The token issuance rate, usually in tokens / second. This is the rate at which the system adds tokens to the bucket at a fixed rate. This rate is pre-set and reflects the maximum request processing rate allowed by the system.

[0029] N: Bucket capacity, that is, the maximum number of tokens that the bucket can store. This parameter limits the maximum number of request bursts allowed at an instant, that is, the ability to process requests at a rate that exceeds the average rate in a short period of time.

[0030] Δt: time interval, which refers to the time granularity of token issuance, or the time difference between two requests. In practical applications, this parameter can be the actual time interval between requests, or the interval at which the system updates the number of tokens in the bucket at a fixed time interval.

[0031] G. Service protection: Use the fuse mechanism to prevent fault propagation. The parameters in circuit breaker mode are as follows:

[0032] T: Circuit breaker time window (in seconds). During this time period, the system will count the number of failures (such as timeouts, abnormal returns, etc.).

[0033] N: The maximum number of faults allowed within the time window T. If the number of faults exceeds N within T seconds, the circuit breaker will change from the closed state to the open state.

[0034] M: The maximum number of test requests allowed in the half-open state. These requests are used to detect whether the service instance has returned to normal.

[0035] Circuit breaker states and their transition rules:

[0036] Closed state: In this state, all requests are passed to the service instance normally. The circuit breaker continuously monitors the number of request failures. If the number of failures exceeds N within the set time window T, the circuit breaker state switches to the open state.

[0037] Open state: When the circuit breaker is in the open state, all requests are automatically rejected, usually with a fast fail response. This is done to prevent further requests to the faulty service, thereby preventing the fault from spreading. After a preset period of time T, the circuit breaker state automatically transitions to the half-open state.

[0038] Half-open state: In this state, the circuit breaker allows a limited number of requests (no more than M) to be passed to the service instance. If all these requests are successful, indicating that the service is stable enough, the circuit breaker will switch back to the closed state and resume the normal request flow. If any request fails during this period, the circuit breaker switches back to the open state, rejects all requests again, and waits for the next evaluation cycle.

[0039] H. Request forwarding and response return: The instance selected by the load balancing server processes the request and returns the result to the client.

[0040] Further, the workflow of the resource management unit is as follows:

[0041] A. Resource demand forecasting: By analyzing the historical resource usage data of the service (such as CPU usage, memory usage, network traffic, etc.), the resource demand of the service in the future is forecasted. The forecasting process can use a variety of time series forecasting algorithms, such as ARIMA, Prophet, LSTM, etc.

[0042] B. Resource capacity planning: Determine the overall configuration scale and elastic scaling range of different resources (such as CPU, memory, disk, network, etc.) based on the resource demand forecast results and the service SLA (service level agreement) requirements. Resource capacity planning can be expressed as a multi-objective optimization problem:

[0043] The constraints are:

[0044] This constraint ensures that the total configuration of all resource types at least meets the overall demand D, ensuring that the system operation will not be affected by insufficient resources.

[0045] Ensure that the amount configured for each resource type is no less than its predicted demand. i , to avoid service interruption or performance degradation due to insufficient resources.

[0046] Ensure that the allocation of any resource does not exceed its maximum supply capacity C i , to accommodate physical or budgetary constraints.

[0047] in:

[0048] f 1 (x) The objective function attempts to minimize the sum of squared deviations between resource allocation and predicted demand. The goal is to get as close as possible to predicted demand, thereby minimizing the risk of over- or under-allocation, which helps optimize resource utilization efficiency and cost control;

[0049] f 2(x) The objective function attempts to minimize the total cost, which is calculated by the sum of the product of the unit cost of the resource and the allocation amount. It encourages decision makers to consider cost-effectiveness and meet resource requirements at the lowest cost possible;

[0050] x=(x 1 ,x 2 ,...x m ) This is a vector where each element x i represents the configuration amount of the i-th resource. These variables are the decision variables of the optimization problem and need to be determined during the optimization process;

[0051] d i Represents the predicted demand c i Indicates the unit cost of resources;

[0052] D represents the total demand;

[0053] C i Indicates the maximum supply capacity.

[0054] C. Resource scheduling optimization: Under the constraint of total resource volume, further optimize the allocation ratio of resources in different services and different time periods. The optimization goal is to minimize resource fragmentation and achieve load balancing:

[0055] The constraints are:

[0056] Ensure that the total amount of the jth resource allocated to all services does not exceed the total available amount C of the resource j .

[0057]

[0058] Ensure that the total amount of resources allocated to each service i from different resources at least meets its minimum requirement d i .

[0059] Indicates that the amount of resources allocated to the service must be an integer. This is usually because resources such as CPU cores or GB of memory cannot be divided into non-integer values.

[0060] in:

[0061] x ij represents the amount of the j-th resource allocated to the i-th service; w ij is the resource preference weight; C j and d i They are the total resource and demand constraints respectively.

[0062] D. Resource elasticity: Dynamically adjust resource quotas based on the real-time load of the service and resource utilization. When the service load exceeds the threshold (such as CPU utilization exceeding 80%), resource expansion is triggered; when the load is below the threshold (such as below 20%), resource reduction is triggered. Resource adjustment can be achieved through a proportional integral controller:

[0063]

[0064] in:

[0065] R(t) is the resource quantity; e(t) is the load deviation; K p and K i is the control coefficient.

[0066] Further, the workflow of the fault-tolerant recovery unit is as follows:

[0067] A. Fault monitoring: By deploying Agent or Sidecar in the service instance, key indicator data of the service process is collected, such as resource indicators such as CPU usage, memory usage, disk I / O, network traffic, and performance indicators such as average response time, error rate, and timeout rate of service calls. These indicator data are sent to the monitoring system in real time to evaluate the health status of the service. The frequency of monitoring data collection is dynamically adjusted according to the importance and change frequency of the service to reflect the service status in real time while controlling the impact on service performance.

[0068] B. Fault detection: Timely discovery of abnormal behavior of services from massive monitoring data is a key task of the fault-tolerant recovery unit. Common anomaly detection methods include threshold method, statistical method and machine learning method. For example, the threshold method sets the warning threshold of key indicators, the statistical method establishes the normal range based on historical data, and the machine learning method establishes the normal behavior model of the service by learning historical data. Here, the statistical anomaly detection algorithm is used:

[0069] Assume that the historical data of service indicator X is {x 1 ,x 2 ,...,x n}, calculate the mean μ and standard

[0070] T=μ±z α / 2 ·σ

[0071] in:

[0072] z α / 2 It is the α / 2 quantile of the standard normal distribution. If the current index value x exceeds the interval [μ-T,μ+T], it is judged as abnormal.

[0073] C. Fault diagnosis: When a service anomaly is detected, the fault-tolerant recovery unit needs to further determine the root cause of the anomaly to guide subsequent recovery operations. Fault diagnosis usually requires comprehensive analysis of multiple data sources, such as monitoring indicators, service logs, call chain tracing, etc., to find out the time, location, impact range and possible causes of the anomaly. Here, a decision tree is used for fault diagnosis:

[0074] Assume there are m service indicators {X 1 ,X 2 ,...X m} and n types of faults {Y 1 ,Y 2 ,...Y m}, and establish a decision tree model T based on historical fault data. When a service anomaly is detected, the current indicator value {x 1 ,x 2 ,...x m}, input the model to obtain the probability distribution P(Y|X) of fault type y:

[0075]

[0076] Select the fault type with the largest posterior probability as the diagnosis result:

[0077] y * =argmaxP(Y=y i |X=x)

[0078] D. Fault recovery: Based on the fault diagnosis results, select an appropriate recovery strategy, such as retry, restart, current limiting, isolation, degradation, etc., to restore the service to normal as soon as possible.

[0079] Further, the workflow of the microservice management unit is as follows:

[0080] A. Service registration: When a microservice instance starts, it sends a registration request to the service registry (such as Eureka, Consul, etc.), providing its own meta information such as service name, IP address, port number, health status, etc. After receiving the request, the registry stores the service instance information in the registry and regularly checks the health status of the instance. If the instance is unavailable, it is removed from the registry.

[0081] B. Service discovery: When a service consumer needs to call a service, it sends a service discovery request to the service registry to query the list of available instances of the service. The service registry searches for qualified instances from the registry and returns the instance list to the consumer. The consumer selects an instance from the instance list (through a load balancing algorithm) and initiates a service call.

[0082] C. Configuration distribution: When a microservice starts, it will pull the latest configuration information from the configuration center (such as Spring Cloud Config, Apollo, etc.). The configuration center stores the configuration data of all microservices, including application parameters, environment variables, data sources, etc. When the configuration changes, the configuration center will actively push or the service will regularly pull the new configuration to achieve hot update of the configuration.

[0083] D. Request routing: External requests enter the microservice system through the API gateway (such as Zuul, Kong, etc.). The API gateway forwards the request to the specific service instance of the backend based on the request URL, parameters, headers and other information, combined with the routing rules in the service registry. During the forwarding process, the API gateway can perform additional functions such as protocol conversion, security authentication, and traffic control.

[0084] E. Link tracing: When a request passes through multiple microservices, the link tracing system (such as Zipkin, Jaeger, etc.) will insert a globally unique trace ID (Trace ID) in the request header and generate a range-unique segment ID (Span ID) for each service call. When the service processes the request, it will record the Trace ID and Span ID in the log and send it to the link tracing system. The link tracing system collects the logs of all services, connects the logs with the same Trace ID in series, and reconstructs the complete call link of the request.

[0085] F. Indicator monitoring: Service instances will collect some key performance indicators (such as the number of requests, response time, number of errors, etc.) regularly (such as every minute) and send them to the indicator monitoring system (such as Prometheus). The indicator monitoring system stores indicator data in a time series database and provides query, aggregation, visualization and other functions to help operation and maintenance personnel observe the operating status of the service and set alarm rules. When the indicator is abnormal, the monitoring system will trigger an alarm and notify relevant personnel to handle it.

[0086] G. Log collection: Service instances will print logs (debug logs, error logs, access logs, etc.) to the console or write them to files. Log collection systems (such as ELK) collect service instance logs in real time through agents (such as Filebeat, Fluentd, etc.) and send them to centralized log storage (such as Elasticsearch). Log storage provides powerful indexing, retrieval, and analysis functions, allowing operation and maintenance personnel to easily query and analyze massive log data and quickly locate problems.

[0087] H. Service governance: Service governance is a series of management and control activities throughout the life cycle of microservices, including service registration and discovery, configuration management, security authentication, current limiting and circuit breaking, degradation and fault tolerance, etc. Service governance is implemented through Sidecar agents (such as Envoy). Sidecar guards service instances, intercepts communications between services, and processes requests according to governance rules. Sidecar obtains governance rules from the control plane of the service mesh (such as Istio), dynamically updates Sidecar configuration, and decouples service governance logic from business logic.

[0088] Furthermore, the workflow of the business decision unit is as follows:

[0089] A. Data Integration and Management

[0090] Collect raw data from various business systems (such as CRM, ERP, etc.), including user behavior data, transaction data, etc.; after data cleaning and conversion, the data is stored in distributed systems such as HDFS to form structured and semi-structured data warehouses.

[0091] B. Data Analysis

[0092] Use big data technologies such as Spark or Flink for batch and stream processing of data; apply various algorithms such as clustering, classification, prediction, etc. to extract business insights from data.

[0093] C. Decision support

[0094] The analysis results are converted into business knowledge, such as user portraits and optimization suggestions, and provided to decision makers through APIs or visualization tools. Based on the knowledge provided, the business team formulates strategies and implements decisions such as marketing and product adjustments.

[0095] D. Optimization and feedback

[0096] The business decision-making unit needs to adjust resource allocation to meet data analysis needs, and rely on the service governance functions provided by the microservice management unit to ensure data security and service quality; monitor the decision-making effects, and continue to optimize data models and business strategies based on feedback.

[0097] Further, the working process of the safety protection unit is as follows:

[0098] A. Threat identification: Continuously collect data from multiple sources such as network traffic, user behavior, application logs, etc.; use machine learning and pattern recognition technology to analyze the collected data to identify potential threat patterns and abnormal behaviors.

[0099] B. Protection strategy formulation: Define protection rules and strategies based on known security threats and vulnerability libraries; deploy corresponding protection measures at the appropriate system level, such as WAF, IDS / IPS, etc.

[0100] C. Real-time monitoring and defense: Monitor the traffic in and out of the network in real time, use predefined thresholds and behavior patterns to detect anomalies; deploy DDoS protection mechanisms, analyze and clean malicious traffic, and prevent service interruptions.

[0101] D. Intrusion Detection and Response: Use various intrusion detection systems (IDS) to automatically detect unauthorized access and other security incidents; when potential threats are detected, automatically execute predefined response measures, such as isolating the attack source, notifying the administrator, etc.

[0102] E. Web Application Firewall (WAF): Analyzes all traffic entering the application, blocks SQL injection, cross-site scripting attacks, etc.; implements defense measures against the attack signature library to protect Web applications from attacks.

[0103] F. API security: Ensure that all API requests are strictly authenticated and use technologies such as OAuth and JWT for security checks; ensure that requesters can only access resources they are authorized to use, and use RBAC or other access control models to manage permissions.

[0104] G. Data protection: Encrypt sensitive data for storage and transmission, and use desensitizing technology to protect privacy when displaying data externally; ensure that data can only be accessed by authorized users, and apply the principle of least privilege to limit data access.

[0105] H. Security Audit and Compliance: Record and manage security-related log information for post-analysis and evidence retention; regularly check the effectiveness of security measures and policies to ensure compliance with relevant regulations and standards.

[0106] Beneficial effects of the present invention:

[0107] 1. Dynamic resource management: Through prediction and real-time monitoring, the system can dynamically adjust resource allocation, optimize computing resource usage, reduce costs and improve efficiency.

[0108] 2. Intelligent traffic scheduling: Integrates advanced load balancing algorithms to automatically route user requests to the best service instance to ensure service response speed and system load balance.

[0109] 3. Efficient fault-tolerant recovery: Implement automatic fault detection and rapid recovery mechanisms to reduce system failure time and improve service availability and business continuity.

[0110] 4. Comprehensive security protection: Implement security strategies from multiple levels to comprehensively protect system and data security and defend against external attacks and internal leakage risks.

[0111] 5. Data-driven business decisions: Use big data and machine learning technologies to extract insights from business data, support efficient decision making, and optimize business processes and user experience.

[0112] 6. Continuous optimization of service quality: The system's built-in continuous monitoring and optimization mechanisms ensure that services are performed to the highest standards while quickly adapting to business needs and market changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0113] Figure 1 The system architecture diagram of the present invention is shown;

[0114] Figure 2 The system working flow chart of the present invention is shown. DETAILED DESCRIPTION

[0115] Exemplary embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0116] Combination Figure 1 The present invention provides a service optimization computer system and method thereof, wherein the system is composed as follows:

[0117] 1. Traffic Scheduling Unit

[0118] The traffic scheduling unit is responsible for receiving client requests and dynamically routing the requests to the optimal service instance based on the characteristics of the request and the real-time performance of the service to achieve load balancing and traffic optimization. The unit dynamically adjusts the routing strategy of the request by monitoring the health status and load level of the service instance in real time to ensure that the request is processed promptly and reliably. At the same time, the unit also has traffic control and service protection functions, which can limit and shape the inlet traffic, and perform fuse and downgrade processing when the service is abnormal.

[0119] 2. Resource Management Unit

[0120] The resource management unit is responsible for resource allocation and scheduling optimization. It formulates a global resource allocation strategy based on the resource demand characteristics and priorities of the service, and continuously optimizes the efficiency of resource use during operation. Specifically, the responsibilities of this unit include: resource demand prediction, predicting the resource consumption trend of services at different time granularities through machine learning algorithms; resource capacity planning, determining the overall resource configuration plan based on resource demand prediction and SLA requirements; resource scheduling optimization, dynamically adjusting the resource allocation ratio between services to minimize resource waste while meeting service performance; resource elastic scaling, automatically adjusting resource configuration according to load changes and fault conditions to achieve elastic adaptation of services.

[0121] 3. Fault-tolerant recovery unit

[0122] The fault-tolerant recovery unit is responsible for improving the reliability and recoverability of services, ensuring that services can quickly self-heal and recover when local failures occur. The unit uses distributed tracking and monitoring technology to collect the health status and key indicators of services in real time, and quickly discovers and locates faults based on anomaly detection and root cause analysis algorithms. After a failure occurs, the unit automatically triggers the failover and recovery process, takes the failed instance offline and migrates traffic to the healthy instance, while triggering the restart and repair of the failed instance.

[0123] 4. Microservice Management Unit

[0124] The microservice management unit provides one-stop microservice development, deployment, operation and maintenance, and governance capabilities. The unit is based on the Service Mesh and Cloud Native architecture to achieve full lifecycle management of services. The main functions include: service registration and discovery, providing a unified service registration center to dynamically manage the address information of services; configuration management, providing a unified configuration center to manage various configuration parameters of services; API gateway, providing a unified service access entry to achieve protocol conversion, security authentication, flow control, etc.; service monitoring, providing a full range of service monitoring capabilities, including call chain tracking, indicator measurement, log collection, etc.; service governance, providing service flow limiting, degradation, authentication, authorization and other governance functions.

[0125] 5. Business Decision-Making Unit

[0126] The business decision-making unit uses big data analysis and machine learning technology to provide data support and intelligent suggestions for business decision-making and process optimization. The unit aggregates various business data (such as user behavior, transaction records, operational indicators, etc.), establishes a unified data warehouse and data mart, and deeply mines data value based on data mining, OLAP analysis, visualization and other technologies to discover business problems and optimization opportunities.

[0127] 6. Safety protection unit

[0128] The security protection unit provides comprehensive security protection capabilities for the entire service system to ensure the confidentiality, integrity and availability of the service. The unit adopts a defense-in-depth strategy to build security lines at different levels, including the network layer, host layer, application layer, and data layer. Typical security measures include: highly available DDoS protection to resist large-volume attacks; Web application firewall (WAF) to filter application layer attacks such as SQL injection and XSS; API authentication and authorization to control API access rights in a fine-grained manner; data encryption and desensitization to prevent sensitive data leakage; security audits to record and analyze various security incidents.

[0129] Combination Figure 2 , the workflow of the present invention is as follows:

[0130] S1. Request reception and analysis

[0131] The system receives HTTP / HTTPS requests from clients; parses the request content, including method, path, parameters, etc., and prepares for subsequent routing and processing.

[0132] S2. Service discovery and request routing

[0133] Query the currently available service instances through the service registration center; use the load balancing strategy to select the appropriate service instance for request processing based on the performance indicators and load conditions of the service instance.

[0134] S3. Resource management and optimization

[0135] Monitor current resource usage, such as CPU and memory usage; automatically adjust resource allocation based on system load to ensure service performance and responsiveness.

[0136] S4. Fault tolerance and exception handling

[0137] Continuously monitor the operational status and key performance indicators of services; automatically take recovery measures when potential service failures are detected, such as rerouting traffic or restarting service instances.

[0138] S5. Service monitoring and governance

[0139] Implement system-level monitoring to collect service operation data, such as response time and system load; perform service governance based on monitoring data and adjust system configuration to optimize performance.

[0140] S6.Business decision support

[0141] Analyze collected business data to identify inefficient processes and potential improvement points; provide data-based business decision support to help optimize operations and enhance user experience.

[0142] S7. Safety precautions

[0143] Implement multi-layered security measures, including access control, data encryption, and intrusion detection; regularly update security policies to respond to new security threats and protect systems and data from infringement.

[0144] S8.Continuous optimization and feedback

[0145] Continuously optimize system configuration and resource allocation based on real-time performance data and user feedback; adjust and improve service processes to adapt to changing business needs and market conditions.

[0146] In summary, the present invention provides a computer system and method for service optimization, which effectively improves the reliability, efficiency and security of services by integrating multiple modules such as advanced traffic scheduling, resource management, exception handling, and security protection. By implementing real-time monitoring, dynamic resource optimization, intelligent fault recovery and precise safety measures, this system not only optimizes the operating process, but also enhances the ability to respond to potential risks, significantly improving the overall service management level. The implementation of this invention can provide a reliable and efficient solution for business areas that require highly reliable and adaptive service optimization.

[0147] It should be noted that although some basic principles and conventional methods of load balancing and resource scheduling may not be elaborated in detail in this document, this is because these aspects follow the common practices and established standards in the field. For these common parts, we believe that they are well known in the technical field, so some descriptions are omitted.

[0148] Although this document shows the specific application of the present invention through specific embodiments, these embodiments are only for illustrative purposes and are not intended to limit the scope of protection of the present invention. The scope of protection of the present invention is defined by the claims, and any appropriate modification, equivalent substitution or improvement based on the principle of the present invention should be deemed to fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be given the broadest interpretation to cover all such modifications, equivalent structures and functions.

Claims

1. A computer system for service optimization, comprising: The traffic scheduling unit is responsible for receiving client requests and dynamically routing the requests to the optimal service instance based on the characteristics of the requests and the real-time performance indicators of the services; The resource management unit formulates a global resource allocation strategy based on the resource demand characteristics and priorities of the service, and continuously optimizes the resource usage efficiency during runtime; The fault-tolerant recovery unit uses distributed tracing and monitoring technology to collect the health status and key performance indicators of services in real time, and quickly discovers and locates faults based on anomaly detection and root cause analysis algorithms; Microservice management unit, which provides one-stop microservice development, deployment, operation and maintenance, and governance capabilities, and implements full lifecycle management of services based on service mesh and cloud-native architecture; The business decision-making unit uses big data analysis and machine learning technology to provide data support and intelligent suggestions for business decision-making and process optimization; The security protection unit adopts a defense-in-depth strategy, builds security lines at different system levels, and provides comprehensive security protection capabilities.

2. The system as described in claim 1, wherein the traffic scheduling unit further includes a traffic control function and a service protection function, which can limit and shape the inlet traffic, and perform fuse and degradation processing when the service is abnormal.

3. The system as described in claim 1, wherein the resource management unit further includes resource demand prediction, resource capacity planning, resource scheduling optimization and resource elastic scaling, and can automatically adjust resource configuration according to load changes and fault conditions to achieve elastic adaptation of services.

4. The system as claimed in claim 1, wherein the fault-tolerant recovery unit further includes fault monitoring, fault detection, fault diagnosis and fault recovery processes to ensure that the service can quickly self-heal and recover when a local fault occurs.

5. The system as described in claim 1, wherein the microservice management unit further includes service registration and discovery, configuration management, API gateway, service monitoring and service governance functions to support effective management and optimized operation and maintenance of services.

6. The system as claimed in claim 1, wherein the business decision unit further utilizes data mining, OLAP analysis and visualization technology to deeply explore data value and discover business problems and optimization opportunities.

7. The system as described in claim 1, wherein the security protection unit further includes DDoS protection, Web application firewall, API authentication and authorization, data encryption and desensitization, and security audit functions to ensure the confidentiality, integrity and availability of the service.

8. A service optimization method applied to the computer system of claim 1, characterized in that: The following steps are involved: S1. Request reception and analysis: Receive HTTP / HTTPS requests from clients and analyze the request contents, including methods, paths, and parameters; S2. Service discovery and request routing: query the currently available service instances through the service registry, and use the load balancing strategy to select the appropriate service instance for request processing based on the performance indicators and load of the service instance; S3. Resource management and optimization: monitor current resource usage, automatically adjust resource allocation according to system load, ensure service performance and response speed, including using machine learning algorithms to predict resource demand, and plan and dynamically adjust resource capacity based on the prediction results; S4. Fault tolerance and exception handling: Continuously monitor the operating status and key performance indicators of the service, and automatically take recovery measures when potential service failures are detected, including using statistical methods for anomaly detection and decision tree algorithms for fault diagnosis; S5. Service monitoring and governance: Implement system-level monitoring, collect service operation data, conduct service governance based on monitoring data, and adjust system configuration to optimize performance; S6. Business decision support: Analyze the collected business data, identify inefficient processes and potential improvement points, provide data-based business decision support, adopt big data analysis and machine learning technology, and support decision making through data integration, data analysis, decision support, optimization and feedback processes; S7. Security protection measures: Implement multi-layered security measures, including DDoS protection, web application firewall, API authentication and authorization, data encryption and desensitization, and security audits; S8. Continuous optimization and feedback: Based on performance data and user feedback, continuously optimize system configuration and resource allocation, and adjust and improve service processes.