Traffic distribution method and system based on label routing
Through the traffic allocation method based on tag routing, the problem of insufficient traffic routing security and isolation in full-link pressure measurement is solved, and the isolation and performance improvement of pressure measurement data is achieved, and massive requests in the production environment can be simulated more realistically.
Patent Information
- Application Number
- CN202411370102.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-29
- Publication Date
- 2024-12-17
AI Technical Summary
During the full-link pressure measurement process, the security and isolation of traffic routes cannot be guaranteed, which can easily lead to interference or leakage risks of pressure measurement data to actual services. The performance of single-machine pressure measurement is limited, making it difficult to simulate the massive production environment ask.
Using the traffic allocation method based on tag routing, by establishing a database to store real traffic, building a pressure measurement engine to integrate distributed batch running model, asynchronous framework and buffer mechanism, adding data sign-up and parameter offset capabilities, deploying environment isolation, and configuring Tag routing rules and filters to achieve traffic isolation and allocation.
It effectively avoids the risk of interference or leakage of pressure measurement data on actual business, reduces the workload and risks of development and transformation, and at the same time breaks through the performance limitations of stand-alone pressure measurement, and can more realistically simulate massive requests in the production environment.
Smart Images
Figure CN119155226A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a flow distribution method, in particular to a flow distribution method based on label routing, and belongs to the technical field of communications. Background Art
[0002] Label routing: A routing strategy that configures routing rules by labeling service providers. This strategy constrains traffic to flow only in specified groups, thereby achieving the purpose of traffic isolation. Label routing can serve as the capability basis for scenarios such as blue-green release and grayscale release.
[0003] Full-link stress testing: Full-link stress testing is a comprehensive performance testing method that aims to simulate high-concurrency access in real scenarios. It mainly performs stress testing on the entire business chain by simulating massive user requests and data, and continuously optimizes the process. Full-link stress testing covers the entire processing flow of user requests, including front-end display, back-end services, database operations, middleware processing, and external dependency services, to ensure that each link can withstand the expected pressure.
[0004] With the development of Internet technology and the expansion of business scale, the complexity of the system continues to increase. In order to ensure the stability and reliability of the business, large-scale stress testing is required to simulate actual user requests and verify the scalability and stability of the system. As an effective stress testing method, full-link stress testing can stress test the entire business chain, discover system bottlenecks and potential problems, and provide a basis for system optimization and upgrades. However, when performing full-link stress testing, the security and isolation of traffic routing cannot be guaranteed, which can easily cause stress test data to interfere with actual business or have the risk of leakage. In addition, the performance of single-machine stress testing is limited, and it is difficult to simulate the massive requests of the production environment.
[0005] To this end, a traffic distribution method and system based on label routing are proposed. Summary of the invention
[0006] In view of this, the present invention provides a traffic distribution method and system based on label routing to solve or alleviate the technical problems existing in the prior art and at least provide a beneficial choice.
[0007] The technical solution of the embodiment of the present invention is implemented as follows: a traffic distribution method based on label routing, comprising the following steps:
[0008] S1. Establish a database to store real traffic for data playback;
[0009] S2. Build a stress testing engine based on label routing - integrate distributed batch running model, asynchronous framework and buffer mechanism;
[0010] S3, add data signing and parameter offset capabilities;
[0011] S4, deployment environment isolation;
[0012] S5. Configure label routing rules;
[0013] S6, insert label routing filter;
[0014] S7. Receive traffic requests through the stress testing engine and complete traffic distribution in conjunction with label routing.
[0015] Further preferably, in S1, the database is used to store the real traffic of the production environment analyzed through the log service.
[0016] Further preferably, in S2, the distributed batch running model is a lightweight distributed batch running model of SchedulerX, which is used for task splitting and horizontal expansion of services.
[0017] Further preferably, in S2, the asynchronous framework is a Spring WebFlux non-blocking asynchronous framework, which is used for separation of I / O read and write events.
[0018] Further preferably, in S2, the buffer mechanism is a RingBuffer ring buffer mechanism, which is used for ring multiplexing memory.
[0019] Further preferably, in S3, the data signing capability is used to tag the data; the parameter offset capability is used to perform business isolation and desensitization.
[0020] Further preferably, in S4, the environment isolation is based on cloud control, and the corresponding machine grouping is configured for the Tag label to represent the isolation of the stress testing instance and the deployment environment.
[0021] Further preferably, in S5, the label routing rule is based on cloud control, and the corresponding routing rule is configured for the Tag label. The label routing rule is loaded by parsing the traffic gateway layer to route the stress testing traffic to the stress testing instance bound to the Tag label for traffic isolation and / or traffic shielding.
[0022] Further preferably, in S6, the tag routing filter is a TagRouteFilter tag routing filter inserted into the filter link of the traffic gateway layer for performing tag filtering of the service instance.
[0023] A method for distributing traffic based on label routing, the system comprises a processor and a register coupled to the processor, the register stores program instructions, and when the program instructions are executed by the processor, the processor executes the above-mentioned method for distributing traffic based on label routing.
[0024] Since the embodiment of the present invention adopts the above technical solution, it has the following advantages: the present invention provides the data playback capability of real online traffic by establishing a database, ensures the isolation of stress test data by label routing technology, and realizes full-link stress test data isolation by dynamic rule labeling and parameter offset, thereby avoiding the risk of interference or leakage of stress test data to actual business, and reducing the workload and risk of development and transformation.
[0025] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present invention will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0027] Figure 1 is a flow chart of the steps of the present invention;
[0028] Figure 2 It is a structural schematic diagram of the present invention;
[0029] Figure 3 It is a schematic diagram of the process of the label routing technology of the present invention. DETAILED DESCRIPTION
[0030] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and descriptions are considered to be exemplary and non-restrictive in nature.
[0031] It should be noted that the terms "first", "second", "symmetrical", "array", etc. are only used to distinguish between description and position description purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the definition of "first", "symmetrical", etc. can explicitly or implicitly include one or more of these features; similarly, when the number of certain features is not limited in the form of words such as "two" or "three", it should be noted that this feature also explicitly or implicitly includes one or more feature quantities.
[0032] Related terms:
[0033] Shadow library: It is a data isolation solution for full-link stress testing. It is to establish a corresponding shadow library on the same instance. If it is shadow traffic, the data generated by the stress testing traffic will be bypassed to the corresponding shadow library, so as to achieve the effect of isolating the data from the production library and avoid the pollution of the production library by the stress testing data.
[0034] Offset field: In full-link stress testing, ensure that stress test data is isolated from actual data to avoid the risk of stress test data interfering with actual business or leaking.
[0035] Traffic isolation: Traffic isolation means dividing network traffic into multiple isolated channels. The traffic in each channel can only flow within it and cannot be mixed with the traffic in other channels. This isolation can effectively prevent network attacks, such as DDoS (distributed denial of service) attacks, because the attack traffic is isolated in a specific channel and will not affect the normal traffic in other channels. In addition, traffic isolation can also be used to protect sensitive data and ensure that data cannot be accessed by unauthorized users during transmission.
[0036] Traffic shielding: Traffic shielding refers to detecting and filtering out malicious or inappropriate traffic in the network to protect the security of network resources and user devices. This shielding can be achieved through various means, such as using firewalls, intrusion detection systems (IDS), intrusion prevention systems (IPS) and other security devices or software. Traffic shielding can prevent the spread of malware, prevent network attacks, and protect user privacy.
[0037] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.
[0038] Embodiment 1
[0039] like Figure 1-3 As shown, an embodiment of the present invention provides a traffic distribution method based on label routing, comprising the following steps:
[0040] S1. Establish a database to store real traffic for data playback;
[0041] S2. Build a stress testing engine based on label routing - integrate distributed batch running model, asynchronous framework and buffer mechanism;
[0042] S3, add data signing and parameter offset capabilities;
[0043] S4, deployment environment isolation;
[0044] S5. Configure label routing rules;
[0045] S6, insert label routing filter;
[0046] S7. Receive traffic requests through the stress testing engine and complete traffic distribution in conjunction with label routing.
[0047] In one embodiment, in S1, the database is used to store the real traffic of the production environment parsed through the log service;
[0048] The stress testing engine is equipped with log storage capabilities through the database, and the log service is Alibaba Cloud's SLS log service;
[0049] Use Alibaba Cloud's SLS log service to parse and store the real traffic of the production environment in the database for data playback;
[0050] SLS Log Service is a serverless log service provided by Alibaba Cloud. It is a managed service for collecting, storing, querying, and analyzing log data in applications. SLS supports multiple programming languages and frameworks, which can help developers process log data more efficiently in a cloud environment.
[0051] In one embodiment, in S2, the distributed batch running model is a lightweight distributed batch running model of SchedulerX, which is used for task splitting and horizontal expansion of services;
[0052] The stress testing engine uses SchedulerX's lightweight distributed batch model to split tasks, achieve horizontal expansion of services, and break through the bottleneck limitations of single-machine stress testing.
[0053] SchedulerX is a lightweight distributed task scheduling platform that supports multi-node distributed operation of tasks and can help users efficiently manage and execute scheduled tasks. In the distributed batch model, SchedulerX can help users split large-scale tasks and perform distributed processing to improve task execution efficiency and system stability.
[0054] In S2, the asynchronous framework is the Spring WebFlux non-blocking asynchronous framework, which is used to separate read and write events of I / O;
[0055] The Spring WebFlux non-blocking asynchronous framework is used through the stress testing engine to separate the read and write events of I / O.
[0056] Spring WebFlux is a module of the Spring Framework that provides support for building non-blocking, asynchronous applications. It was introduced in Spring 5.0 to support the responsive programming model, especially in Web application development. Spring WebFlux uses Reactor as its underlying library and provides an event-driven Web framework based on non-blocking I / O and asynchronous processing.
[0057] In S2, the buffer mechanism is a RingBuffer ring buffer mechanism, which is used for ring multiplexing memory;
[0058] The stress testing engine uses the RingBuffer mechanism to reuse memory in a loop, reducing frequent garbage collection caused by massive data.
[0059] Ring Buffer is a data structure that stores data in a loop in a set of fixed-size buffers. This buffer is designed to efficiently process continuous data streams and is often used for data buffering in real-time systems.
[0060] The characteristics of the ring buffer are as follows:
[0061] Data Circular Storage: When the buffer is full, new data will overwrite the old data. The data in the buffer is processed in the order of First In First Out (FIFO).
[0062] Fixed size: When a buffer is created, its capacity is determined. This capacity is fixed and does not change with the increase or decrease of data.
[0063] Efficient reading and writing: Since data is stored continuously in memory, read and write operations can be very efficient, especially when processing continuous data streams.
[0064] No need to move data: Unlike a linear buffer, when the buffer is full, data does not need to be moved or copied to make room for new data.
[0065] Multiple producers single consumer model: Ring buffers are often used in situations where multiple producers provide data to one consumer.
[0066] Synchronization mechanism: Since multiple producers and consumers are involved, a synchronization mechanism is usually required to ensure data consistency and integrity.
[0067] In one embodiment, in S3, the data signing capability is used to tag data; the parameter offset capability is used to perform business isolation and desensitization;
[0068] The data signing capability can be used to tag data for subsequent tag routing. The parameter offset capability can be used to solve problems such as business isolation and desensitization of data.
[0069] Tag is a category tag that identifies the stress testing instance.
[0070] In one embodiment, in S4, environment isolation is based on cloud control, and corresponding machine groups are configured for Tag labels to indicate the isolation of stress testing instances and deployment environments;
[0071] When deploying a service, the service instance is bound to the machine group corresponding to the tag through environmental isolation. This achieves the isolation of the deployment environment. At the same time, after the service is successfully deployed, a tag will be added to identify the stress test instance.
[0072] In one embodiment, in S5, the label routing rule is based on cloud control, and the corresponding routing rule is configured for the Tag label. The label routing rule is loaded by parsing the traffic gateway layer, and the stress test traffic is routed to the stress test instance bound to the Tag label for traffic isolation.
[0073] Label routing rules provide a flexible and dynamic way to manage traffic in cloud environments. It allows users to quickly adjust routing policies based on actual needs without making large-scale modifications to the existing network architecture. This mechanism is very valuable for users of cloud-native applications and services that want to achieve automation, high availability, and scalability.
[0074] In one embodiment, in S6, the tag routing filter is a TagRouteFilter tag routing filter inserted into the filter link of the traffic gateway layer to perform tag filtering of the service instance;
[0075] Insert TagRouteFilter tag routing filter in the filter link of the traffic gateway layer to implement tag filtering of service instances. For example: stress test data Tag = tag1, registration center service instance list = [tag1, tag2, tag3]. After interception, stress test requests can only be routed within the instance range corresponding to tag1;
[0076] "TagRouteFilter" generally refers to a mechanism used for label-based routing decisions in network architecture. This filter can decide how to handle requests based on labels in the request (such as service identifier, version number, environment, etc.), including routing the request to a specific service instance or performing a specific operation.
[0077] Embodiment 2
[0078] The embodiment of the present invention also provides a method for distributing traffic based on label routing, comprising the following steps:
[0079] S1. Establish a database to store real traffic for data playback;
[0080] S2. Build a stress testing engine based on label routing - integrate distributed batch running model, asynchronous framework and buffer mechanism;
[0081] S3, add data signing and parameter offset capabilities;
[0082] S4, deployment environment isolation;
[0083] S5. Configure label routing rules;
[0084] S6, insert label routing filter;
[0085] S7. Receive traffic requests through the stress testing engine and complete traffic distribution in conjunction with label routing.
[0086] In one embodiment, in S1, the database is used to store the real traffic of the production environment through the log service; the stress testing engine is equipped with log storage capability through the database, wherein the log service is the CLS log service of Tencent Cloud;
[0087] Through Tencent Cloud's CLS log service, parse and store the real traffic of the production environment in the database for data playback;
[0088] Tencent Cloud Log Service CLS provides log collection, storage, query, analysis and other functions, and supports access to multiple data sources, such as cloud products, custom applications, etc.
[0089] It also provides data visualization and alarm notification functions to help users monitor and manage log data in real time.
[0090] In one embodiment, in S2, the distributed batch running model is a lightweight distributed batch running model of SchedulerX, which is used for task splitting and horizontal expansion of services;
[0091] The stress testing engine uses SchedulerX's lightweight distributed batch model to split tasks, achieve horizontal expansion of services, and break through the bottleneck limitations of single-machine stress testing.
[0092] SchedulerX is a lightweight distributed task scheduling platform that supports multi-node distributed operation of tasks and can help users efficiently manage and execute scheduled tasks. In the distributed batch model, SchedulerX can help users split large-scale tasks and perform distributed processing to improve task execution efficiency and system stability;
[0093] The characteristics of SchedulerX's lightweight distributed batch model are:
[0094] High availability: SchedulerX supports multi-node operation. When a node fails, other nodes can continue to execute tasks to ensure the continuous operation of tasks.
[0095] Horizontal expansion: By adding nodes, tasks can be expanded horizontally, improving task processing capabilities and system throughput;
[0096] Task splitting: Users can split large-scale tasks into smaller task units that can be processed in parallel according to the nature and needs of the tasks;
[0097] Distributed execution: SchedulerX can automatically distribute tasks to different nodes for execution, improving the efficiency of task execution;
[0098] Flexible task scheduling: SchedulerX provides a variety of task scheduling strategies, supporting various task scheduling methods such as timing, trigger, and periodicity;
[0099] Task monitoring and management: SchedulerX provides task monitoring and management functions, allowing users to view the execution status and operation of tasks in real time, facilitating troubleshooting and performance optimization;
[0100] Integration and compatibility: SchedulerX can be integrated with other cloud services and applications, supports multiple programming languages and data formats, and has good compatibility.
[0101] In S2, the asynchronous framework is the Spring WebFlux non-blocking asynchronous framework, which is used to separate read and write events of I / O;
[0102] The Spring WebFlux non-blocking asynchronous framework is used through the stress testing engine to separate the read and write events of I / O.
[0103] Spring WebFlux is a module of the Spring framework that provides support for building non-blocking, asynchronous applications. It was introduced in Spring 5.0 to support the responsive programming model, especially in web application development. Spring WebFlux uses Reactor as its underlying library and provides an event-driven web framework based on non-blocking I / O and asynchronous processing;
[0104] The characteristics of Spring WebFlux non-blocking asynchronous framework are:
[0105] Responsive programming model: Spring WebFlux uses the Reactor library as its reactive programming foundation, supporting non-blocking I / O operations and asynchronous processing. This allows applications to handle high concurrency and latency-sensitive requests more efficiently;
[0106] Non-blocking I / O: Unlike traditional synchronous blocking frameworks, Spring WebFlux uses non-blocking I / O, which means that when processing a request, the thread can perform other tasks while waiting for a response or data transfer, thereby improving resource utilization and application performance;
[0107] Asynchronous processing: Spring WebFlux supports asynchronous processing mode, allowing operations to be performed in background threads without blocking the main thread. This helps reduce waiting time and improve the responsiveness of the application;
[0108] WebFlux annotation-based programming model: Spring WebFlux provides a set of annotations, such as @RestController, @RequestMapping, etc., which allows developers to easily create reactive web controllers;
[0109] Compatibility: Although Spring WebFlux is reactive, it is compatible with Spring's traditional synchronous web applications. This means that you can use synchronous and asynchronous processing in the same project, choosing the appropriate mode as needed;
[0110] Feature-rich ecosystem: Spring WebFlux seamlessly integrates with other modules in the Spring ecosystem (such as Spring Data, Spring Security, etc.), providing the functionality required to build complex applications.
[0111] In S2, the buffer mechanism is a RingBuffer ring buffer mechanism, which is used for ring multiplexing memory;
[0112] The stress testing engine uses the RingBuffer mechanism to reuse memory in a loop, reducing frequent garbage collection caused by massive data.
[0113] Ring Buffer is a data structure that stores data in a loop in a set of fixed-size buffers. This buffer is designed to efficiently process continuous data streams and is often used for data buffering in real-time systems.
[0114] The characteristics of the ring buffer are as follows:
[0115] Data Circular Storage: When the buffer is full, new data will overwrite the old data. The data in the buffer is processed in the order of First In First Out (FIFO).
[0116] Fixed size: When a buffer is created, its capacity is determined. This capacity is fixed and does not change with the increase or decrease of data.
[0117] Efficient reading and writing: Since data is stored continuously in memory, read and write operations can be very efficient, especially when processing continuous data streams.
[0118] No need to move data: Unlike a linear buffer, when the buffer is full, data does not need to be moved or copied to make room for new data.
[0119] Multiple producers single consumer model: Ring buffers are often used in situations where multiple producers provide data to one consumer.
[0120] Synchronization mechanism: Since multiple producers and consumers are involved, a synchronization mechanism is usually required to ensure data consistency and integrity.
[0121] In one embodiment, in S3, the data signing capability is used to tag data; the parameter offset capability is used to perform business isolation and desensitization;
[0122] The data signing capability can be used to tag data for subsequent tag routing. The parameter offset capability can be used to solve problems such as business isolation and desensitization of data.
[0123] Tag is a category tag that identifies the stress testing instance.
[0124] In one embodiment, in S4, environment isolation is based on cloud control, and corresponding machine groups are configured for Tag labels to indicate the isolation of stress testing instances and deployment environments;
[0125] When deploying a service, the service instance is bound to the machine group corresponding to the tag through environmental isolation. This achieves the isolation of the deployment environment. At the same time, after the service is successfully deployed, a tag will be added to identify the stress test instance.
[0126] In one embodiment, in S5, the label routing rule is based on cloud control, and the corresponding routing rule is configured for the Tag label. The label routing rule is loaded by parsing the traffic gateway layer, and the stress test traffic is routed to the stress test instance bound to the Tag label for traffic isolation. When the system load is too high, the label routing rule is modified to achieve traffic shielding.
[0127] Label routing rules provide a flexible and dynamic way to manage traffic in cloud environments. It allows users to quickly adjust routing policies based on actual needs without making large-scale modifications to the existing network architecture. This mechanism is very valuable for users of cloud-native applications and services that want to achieve automation, high availability, and scalability.
[0128] In one embodiment, in S6, the tag routing filter is a TagRouteFilter tag routing filter inserted into the filter link of the traffic gateway layer to perform tag filtering of the service instance;
[0129] Insert the TagRouteFilter tag routing filter into the filter link of the traffic gateway layer to implement tag filtering of service instances.
[0130] For example: stress test data Tag = tag1, registration center service instance list = [tag1, tag2, tag3].
[0131] After interception, the stress test request can only be routed within the instance range corresponding to tag1;
[0132] "TagRouteFilter" generally refers to a mechanism used for label-based routing decisions in network architecture. This filter can decide how to handle requests based on labels in the request (such as service identifier, version number, environment, etc.), including routing the request to a specific service instance or performing a specific operation.
[0133] Embodiment 3
[0134] An embodiment of the present invention also provides a traffic distribution method based on label routing. The system includes a processor and a register coupled to the processor. The register stores program instructions. When the program instructions are executed by the processor, the processor executes the traffic distribution method based on label routing of embodiment one or embodiment two.
[0135] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of various changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A traffic distribution method based on label routing, characterized in that: The following steps are involved: S1. Establish a database to store real traffic; S2. Build a stress testing engine based on label routing - integrate distributed batch running model, asynchronous framework and buffer mechanism; S3, add data signing and parameter offset capabilities; S4, deployment environment isolation; S5. Configure label routing rules; S6, insert label routing filter; S7. Receive traffic requests through the stress testing engine and complete traffic distribution in conjunction with label routing.
2. The method for distributing traffic based on label routing according to claim 1, characterized in that: In S1, the database is used to store the real traffic of the production environment analyzed by the log service.
3. The method for distributing traffic based on label routing according to claim 1, characterized in that: In S2, the distributed batch running model is a lightweight distributed batch running model of SchedulerX, which is used for task splitting and horizontal expansion of services.
4. The method for distributing traffic based on label routing according to claim 3 is characterized in that: In S2, the asynchronous framework is the Spring WebFlux non-blocking asynchronous framework, which is used for separation of I / O read and write events.
5. The method for distributing traffic based on label routing according to claim 4, characterized in that: In S2, the buffer mechanism is a RingBuffer ring buffer mechanism, which is used for ring multiplexing memory.
6. The method for distributing traffic based on label routing according to claim 1, characterized in that: In S3, the data signing capability is used to tag data; the parameter offset capability is used to perform business isolation and desensitization.
7. The method for distributing traffic based on label routing according to claim 6, characterized in that: In S4, environment isolation is based on cloud control, and corresponding machine groups are configured for Tag labels to indicate the isolation of stress testing instances and deployment environments.
8. The method for distributing traffic based on label routing according to claim 6, characterized in that: In S5, the label routing rule is based on cloud control, and the corresponding routing rule is configured for the Tag label. The label routing rule is loaded by parsing the traffic gateway layer to route the stress testing traffic to the stress testing instance bound to the Tag label for traffic isolation and / or traffic shielding.
9. The method for distributing traffic based on label routing according to claim 6, characterized in that: In S6, the tag routing filter is inserted into the filter link of the traffic gateway layer, and is used for tag filtering of the service instance.
10. A traffic distribution method based on label routing, characterized in that: The system includes a processor and a register coupled to the processor, wherein the register stores program instructions, and when the program instructions are executed by the processor, the processor executes the label routing-based traffic distribution method described in any one of claims 1-9.