Active network performance detection method and system

Abstract

The invention discloses an active network performance detection method and system, and the method comprises the steps: generating different types of detection packets based on monitoring demands and historical data; based on network topology, real-time flow and monitoring requirements, selecting an optimal routing path and a sending strategy, and sending the probe packet to a target network; deploying a distributed response data collection module at a key node of the target network, capturing response data of the probe packet in real time, and performing primary processing; and based on deep learning and big data analysis technologies, carrying out deep mining and intelligent analysis on the response data after primary processing, extracting key performance indexes and abnormal modes, and generating a performance optimization scheme and a resource allocation strategy. Through integration of artificial intelligence, big data analysis and an advanced routing strategy, deep, real-time and intelligent detection of network performance can be realized, and comprehensive support is provided for network optimization, fault prediction, dynamic resource allocation and network security.

Description

Technical Field

[0001] The present invention relates to the field of network security technology, and in particular to an active network performance detection method and system. Background Art Background technology:

[0003] With the rapid rise and deep integration of cutting-edge technologies such as cloud computing, big data, the Internet of Things, and artificial intelligence, the network environment has entered a new era of unprecedented complexity and dynamism. These technologies have not only greatly promoted the pace of digital transformation, but also posed more stringent challenges to network performance monitoring. Traditional network performance monitoring methods, limited by their passive data traffic analysis mode, are gradually exposing many limitations and are unable to meet the urgent needs of modern networks for efficiency, real-time, and intelligence. These methods can often only perform post-analysis after the problem occurs, with significant monitoring lags, and due to the limitations of data sampling frequency and coverage, the monitoring results are neither comprehensive nor accurate.

[0004] The prior art has the following problems:

[0005] Monitoring lag and lack of accuracy: Traditional passive monitoring methods, such as log analysis or regular network scanning, have inherent data collection and analysis delays, making it difficult to capture the real-time status of network performance immediately. This lag not only affects the timely discovery of problems, but also hinders the efficiency of rapid response and troubleshooting, resulting in a significant reduction in the quality of network services.

[0006] Coverage blind spots and data sparsity: Due to the complexity of network architecture and uneven distribution of traffic, data in some network areas or specific periods (such as peak hours) may be missing due to insufficient deployment of monitoring equipment or improper strategies, resulting in one-sided and incomplete monitoring results. This data sparsity seriously affects the overall evaluation of network performance and the formulation of optimization decisions.

[0007] Insufficient intelligence and automation: Facing massive and ever-changing network data, traditional methods lack efficient intelligent analysis tools and automated processing mechanisms. They cannot automatically identify abnormal behaviors in the network, predict potential failure points, or dynamically optimize resource configuration, resulting in high operation and maintenance costs and low efficiency.

[0008] Poor adaptability to complex network environments: In the complex network ecosystem built by emerging technologies such as multi-cloud, hybrid cloud, edge computing, and 5G, the effectiveness of traditional monitoring methods has significantly decreased. These environments not only require the monitoring system to have high flexibility and scalability, but also require the ability to conduct cross-platform and cross-domain collaborative monitoring and intelligent analysis, which is difficult to achieve with traditional methods.

[0009] To sum up, in order to solve the above problems, a more advanced, intelligent and comprehensive network performance monitoring solution is urgently needed to adapt to the development needs of future network environments and ensure the stability, efficiency and sustainable development of network services. Summary of the invention

[0010] In order to solve the above problems, the present invention proposes an active network performance detection method and system, which can realize in-depth, real-time and intelligent detection of network performance by integrating artificial intelligence, big data analysis and advanced routing strategies, and provide comprehensive support for network optimization, fault prediction, dynamic resource allocation and network security.

[0011] The technical solution adopted by the present invention is as follows:

[0012] An active network performance detection method, comprising:

[0013] Generate different types of detection packages based on monitoring requirements and historical data;

[0014] Based on network topology, real-time traffic and monitoring requirements, the best routing path and sending strategy are selected to send the detection packet to the target network;

[0015] Deploy distributed response data collection modules at key nodes of the target network to capture the response data of the detection package in real time and perform preliminary processing;

[0016] Based on deep learning and big data analysis technology, the response data after preliminary processing is deeply mined and intelligently analyzed to extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies.

[0017] Furthermore, based on monitoring requirements and historical data, different types of detection packages are generated, including:

[0018] Setting the detection packet format and verification algorithm, wherein the packet header content of the detection packet includes a type field, a priority field, a timestamp field and a verification code field;

[0019] Use machine learning algorithms to analyze historical data and real-time monitoring needs, generate different types of detection packets, and dynamically adjust the generation frequency of detection packets according to network conditions;

[0020] The encryption algorithm is dynamically selected according to the security level of the network and the sensitivity of the detection packet; a dynamic key management mechanism is adopted to ensure that the encryption key of each detection packet is different.

[0021] Furthermore, the method of selecting the best routing path and sending strategy based on network topology, real-time traffic and monitoring requirements, and sending the detection packet to the target network includes:

[0022] Intelligent routing selection: According to network topology, real-time traffic and monitoring requirements, the best routing path is selected through routing algorithms; traffic changes in the target network are monitored in real time, the sending path of the detection packet is dynamically adjusted, and the routing table is updated regularly;

[0023] Dynamic scheduling and load balancing: A load balancing algorithm is used to evenly distribute the detection packets in the target network, and the sending order and frequency of the detection packets are dynamically adjusted according to the network load. High-priority detection packets are sent first when the network is congested.

[0024] Adaptive sending frequency adjustment: Dynamically adjust the sending frequency of the detection packet according to the network status and monitoring results, reduce the frequency when the load is high, and increase the frequency when the load is low.

[0025] Furthermore, the distributed response data collection module is deployed at the key nodes of the target network to capture the response data of the detection packet in real time and perform preliminary processing, including:

[0026] Distributed data collection: Deploy distributed response data collection modules at each key node of the target network, covering all key paths of the target network, and capturing the response data of the detection packet in real time;

[0027] Data cleaning and preprocessing: Use big data processing technology to clean, deduplicate, aggregate and format the collected response data, and format the cleaned data into a unified format;

[0028] Real-time transmission and distributed storage: Use message queue technology to achieve real-time data transmission, and use distributed database to store response data.

[0029] Furthermore, based on deep learning and big data analysis technology, the response data after preliminary processing is deeply mined and intelligently analyzed to extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies, including:

[0030] Deep learning and big data analysis: Use deep learning algorithms to conduct in-depth mining and intelligent analysis of the processed response data, including delay analysis, bandwidth analysis, and packet loss rate analysis; use big data analysis technology to conduct real-time analysis of massive response data and extract key performance indicators;

[0031] Real-time evaluation and fault warning: Real-time evaluation of network performance and generation of performance reports; identification of potential performance bottlenecks, fault points, and security risks through machine learning models, and issuance of warnings;

[0032] Performance optimization suggestions: Automatically generate optimization suggestions based on the analysis results. When it is detected that the delay of a certain path is higher than the threshold, it is recommended to adjust the routing policy; when it is detected that the bandwidth is insufficient, it is recommended to increase the bandwidth; when it is detected that the resource allocation is unbalanced, it is recommended to reallocate resources.

[0033] An active network performance detection system, comprising:

[0034] An intelligent detection packet generation module is configured to generate different types of detection packets based on monitoring requirements and historical data;

[0035] The dynamic routing and sending module is configured to select the best routing path and sending strategy based on the network topology, real-time traffic and monitoring requirements, and send the detection packet to the target network;

[0036] The distributed response data collection module is configured at the key nodes of the target network to capture the response data of the detection packet in real time and perform preliminary processing;

[0037] The intelligent performance analysis module is configured to perform deep mining and intelligent analysis on the response data after preliminary processing based on deep learning and big data analysis technology, extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies.

[0038] Furthermore, in the intelligent detection package generation module, different types of detection packages are generated based on monitoring requirements and historical data, including:

[0039] Setting the detection packet format and verification algorithm, wherein the packet header content of the detection packet includes a type field, a priority field, a timestamp field and a verification code field;

[0040] Use machine learning algorithms to analyze historical data and real-time monitoring needs, generate different types of detection packets, and dynamically adjust the generation frequency of detection packets according to network conditions;

[0041] The encryption algorithm is dynamically selected according to the security level of the network and the sensitivity of the detection packet; a dynamic key management mechanism is adopted to ensure that the encryption key of each detection packet is different.

[0042] Furthermore, in the dynamic routing and sending module, based on the network topology, real-time traffic and monitoring requirements, the best routing path and sending strategy are selected to send the detection packet to the target network, including:

[0043] Intelligent routing selection: According to network topology, real-time traffic and monitoring requirements, the best routing path is selected through routing algorithms; traffic changes in the target network are monitored in real time, the sending path of the detection packet is dynamically adjusted, and the routing table is updated regularly;

[0044] Dynamic scheduling and load balancing: A load balancing algorithm is used to evenly distribute the detection packets in the target network, and the sending order and frequency of the detection packets are dynamically adjusted according to the network load. High-priority detection packets are sent first when the network is congested.

[0045] Adaptive sending frequency adjustment: Dynamically adjust the sending frequency of the detection packet according to the network status and monitoring results, reduce the frequency when the load is high, and increase the frequency when the load is low.

[0046] Furthermore, in the distributed response data collection module, a distributed response data collection module is deployed at a key node of the target network to capture the response data of the detection packet in real time and perform preliminary processing, including:

[0047] Distributed data collection: Deploy distributed response data collection modules at each key node of the target network, covering all key paths of the target network, and capturing the response data of the detection packet in real time;

[0048] Data cleaning and preprocessing: Use big data processing technology to clean, deduplicate, aggregate and format the collected response data, and format the cleaned data into a unified format;

[0049] Real-time transmission and distributed storage: Use message queue technology to achieve real-time data transmission, and use distributed database to store response data.

[0050] Furthermore, in the intelligent performance analysis module, based on deep learning and big data analysis technology, the response data after preliminary processing is deeply mined and intelligently analyzed to extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies, including:

[0051] Deep learning and big data analysis: Use deep learning algorithms to conduct in-depth mining and intelligent analysis of the processed response data, including delay analysis, bandwidth analysis, and packet loss rate analysis; use big data analysis technology to conduct real-time analysis of massive response data and extract key performance indicators;

[0052] Real-time evaluation and fault warning: Real-time evaluation of network performance and generation of performance reports; identification of potential performance bottlenecks, fault points, and security risks through machine learning models, and issuance of warnings;

[0053] Performance optimization suggestions: Automatically generate optimization suggestions based on the analysis results. When it is detected that the delay of a certain path is higher than the threshold, it is recommended to adjust the routing policy; when it is detected that the bandwidth is insufficient, it is recommended to increase the bandwidth; when it is detected that the resource allocation is unbalanced, it is recommended to reallocate resources.

[0054] The beneficial effects of the present invention are:

[0055] Deep monitoring: Through intelligent detection packages and deep learning algorithms, we can gain in-depth insights into subtle changes in network performance, and achieve in-depth monitoring and intelligent analysis of network performance. These intelligent detection packages can actively detect network status, combined with the powerful analysis capabilities of deep learning algorithms, to accurately identify potential problems and performance bottlenecks in the network.

[0056] Real-time: Advanced dynamic routing technology and real-time data processing technology are used to ensure that monitoring data can instantly feedback the latest status of network performance, thereby greatly improving the real-time and accuracy of monitoring results. This real-time performance is crucial for quickly responding to network failures and ensuring business continuity.

[0057] Comprehensiveness: The in-depth monitoring solution covers all key paths and nodes of the network, and can achieve comprehensive and complete monitoring of both core devices and edge nodes. This comprehensiveness ensures all-round control of network performance and effectively avoids potential risks caused by monitoring blind spots.

[0058] Intelligence: The in-depth monitoring solution integrates artificial intelligence and big data analysis technology, which can automatically analyze network performance data and provide intelligent performance evaluation, prediction and optimization suggestions. Through machine learning algorithms, the system can continuously learn and optimize monitoring strategies to achieve more accurate and efficient performance management. This intelligence not only improves operation and maintenance efficiency, but also reduces the risk of human intervention, providing a strong guarantee for the stable operation of the network.

[0059] In summary, the present invention realizes in-depth, real-time, and intelligent detection of network performance by integrating artificial intelligence, big data analysis, and advanced routing strategies, which not only improves the accuracy and real-time performance of network performance monitoring, but also provides network administrators with intelligent decision-making basis and optimization solutions, which helps to improve the overall performance, user experience, and security of the network. Therefore, the present invention has significant technical advantages and broad application prospects, and is an indispensable network performance monitoring and analysis tool in the modern network environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a flow chart of an active network performance detection method according to Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0061] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, the specific implementation methods of the present invention are now described. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the embodiments described are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0062] Example 1

[0063] like Figure 1 As shown, this embodiment provides an active network performance detection method, including:

[0064] Generate different types of detection packages based on monitoring requirements and historical data;

[0065] Based on network topology, real-time traffic and monitoring requirements, the best routing path and sending strategy are selected to send the detection packet to the target network;

[0066] Deploy distributed response data collection modules at key nodes of the target network to capture the response data of the detection package in real time and perform preliminary processing;

[0067] Based on deep learning and big data analysis technology, the response data after preliminary processing is deeply mined and intelligently analyzed to extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies.

[0068] Preferably, different types of detection packages are generated based on monitoring requirements and historical data, which can be achieved by the following steps:

[0069] Step 1.1: Design a flexible probe packet format

[0070] (1) Packet header design: The header of the detection packet contains the following fields:

[0071] Type field: identifies the type of detection packet (such as delay detection, bandwidth detection, packet loss rate detection, etc.).

[0072] Priority field: identifies the priority of the detection packet. High-priority detection packets are sent first when the network is congested.

[0073] Timestamp field: records the time when the detection packet is generated and is used to calculate network delay.

[0074] Checksum field: used to ensure the integrity and accuracy of the detection packet during transmission.

[0075] (2) Intelligent payload design: The payload is dynamically generated based on the type of the probe packet. For example, the payload of a latency probe packet can contain timestamp information, and the payload of a bandwidth probe packet can contain large-sized data blocks to test the bandwidth.

[0076] (3) Checksum design: Use checksum algorithms such as CRC32 or MD5 to ensure that the detection packet is not tampered with or damaged during transmission.

[0077] Step 1.2: Intelligently generate a detection packet

[0078] (1) Machine learning algorithms: Machine learning algorithms such as decision trees and random forests are used to analyze historical data and real-time monitoring needs, and intelligently generate different types of detection packages.

[0079] Delay detection packet: When the system detects an increase in network delay, it automatically generates a delay detection packet to measure the delay in the network.

[0080] Bandwidth detection packet: When the system detects a bandwidth bottleneck, it automatically generates a bandwidth detection packet to measure the bandwidth utilization in the network.

[0081] Packet loss rate detection packet: When the system detects an abnormal packet loss rate, it automatically generates a packet loss rate detection packet to measure the packet loss situation in the network.

[0082] (2) Dynamically adjust the generation frequency: Dynamically adjust the generation frequency of the detection packet according to the network conditions. For example, when the network load is high, the system will automatically reduce the generation frequency of the detection packet to reduce the impact on network performance.

[0083] Step 1.3: Adaptive Encryption Techniques

[0084] (1) Encryption algorithm selection: Dynamically select an encryption algorithm based on the network security level and the sensitivity of the detection packet. For example, for a high-security network, use the AES-256 encryption algorithm; for a common network, use the RSA encryption algorithm.

[0085] (2) Key management: A dynamic key management mechanism is used to ensure that the encryption key of each detection packet is different, preventing the security of the entire system from being affected if the key is cracked.

[0086] Preferably, based on the network topology, real-time traffic and monitoring requirements, the best routing path and sending strategy are selected to send the detection packet to the target network, which can be achieved by the following steps:

[0087] Step 2.1: Intelligent routing

[0088] (1) Routing algorithm: Advanced routing algorithms such as the Dijkstra algorithm and the A* algorithm are used to intelligently select the best routing path based on network topology, real-time traffic, and monitoring requirements.

[0089] Real-time traffic monitoring: The system monitors traffic changes in the network in real time and dynamically adjusts the sending path of the detection packet to avoid network congestion.

[0090] Multi-path selection: When a path is detected to be congested, the system will automatically select other available paths to ensure that the detection packet can successfully reach the target node.

[0091] (2) Routing table update: The system regularly updates the routing table to ensure the timeliness and accuracy of routing information.

[0092] Step 2.2: Dynamic Scheduling and Load Balancing

[0093] (1) Load balancing algorithm: Use load balancing algorithms such as round-robin and weighted round-robin to ensure that the detection packets are evenly distributed in the network and avoid overloading certain nodes.

[0094] Dynamic scheduling: Dynamically adjust the order and frequency of sending probe packets according to the network load. For example, when a node has a high load, the system will automatically reduce the frequency of sending probe packets to the node.

[0095] (2) Priority scheduling: High-priority detection packets are sent first when the network is congested to ensure the real-time performance of key monitoring tasks.

[0096] Step 2.3: Adaptive transmission frequency adjustment

[0097] Frequency adjustment mechanism: Dynamically adjust the frequency of sending detection packets according to network conditions and monitoring results. When the network load is high, the system will automatically reduce the frequency of sending detection packets to reduce the impact on network performance; when the network load is low, the system will increase the frequency of sending detection packets to improve the accuracy of monitoring.

[0098] Preferably, a distributed response data collection module is deployed at key nodes of the target network to capture the response data of the detection packet in real time and perform preliminary processing, which can be achieved by the following steps:

[0099] Step 3.1: Distributed Data Collection

[0100] (1) Data collection module deployment: Deploy distributed response data collection modules at each key node of the network to capture the response data of the detection packet in real time.

[0101] Node selection: Select key nodes in the network (such as routers, switches, etc.) to deploy data collection modules to ensure that all key paths of the network are covered.

[0102] (2) Real-time capture: The data collection module captures the response data of the detection packet in real time and sends the data to the central processing system.

[0103] Step 3.2: Data cleaning and preprocessing

[0104] (1) Data cleaning: Use big data processing technologies such as Hadoop and Spark to clean, deduplicate, aggregate, and format the collected response data.

[0105] Deduplication: Remove duplicate response data to ensure data uniqueness.

[0106] Filling missing values: For missing data, interpolation or mean method is used to fill it.

[0107] Correcting erroneous data: For obviously erroneous data, the rule engine is used to correct it.

[0108] (2) Data formatting: Format the cleaned data into a unified format to facilitate subsequent analysis.

[0109] Step 3.3: Real-time transmission and distributed storage

[0110] (1) Real-time transmission: Use message queue technologies such as Kafka and RabbitMQ to achieve real-time data transmission and ensure the timeliness of data.

[0111] (2) Distributed storage: Use distributed databases such as Cassandra and MngDB to store response data to ensure high availability and fault tolerance of data.

[0112] Preferably, based on deep learning and big data analysis technology, the response data after preliminary processing is deeply mined and intelligently analyzed to extract key performance indicators and abnormal patterns, generate performance optimization solutions and resource allocation strategies, which can be achieved by the following steps:

[0113] Step 4.1: Deep Learning and Big Data Analysis

[0114] (1) Deep learning algorithm: Deep learning algorithms such as convolutional neural network (CNN) and recurrent neural network (RNN) are used to perform in-depth mining and intelligent analysis of the processed response data.

[0115] Delay analysis: By analyzing the response data of the delay detection packet, the delay distribution in the network is extracted.

[0116] Bandwidth analysis: Extract bandwidth utilization in the network by analyzing the response data of bandwidth detection packets.

[0117] Packet loss rate analysis: By analyzing the response data of the packet loss rate detection packet, the packet loss rate in the network can be extracted.

[0118] (2) Big data analysis: Use big data analysis technologies such as Spark and Flink to perform real-time analysis on massive response data and extract key performance indicators.

[0119] Step 4.2: Real-time evaluation and fault warning

[0120] (1) Real-time evaluation: The system evaluates network performance in real time and generates performance reports.

[0121] Latency assessment: When the network latency is detected to exceed the threshold, the system will automatically issue an alert.

[0122] Bandwidth assessment: When the bandwidth utilization is detected to exceed the threshold, the system will automatically issue an alert.

[0123] Packet loss rate assessment: When the packet loss rate is detected to exceed the threshold, the system will automatically issue an early warning.

[0124] (2) Fault warning: Through machine learning models (such as support vector machines (SVMs), random forests, etc.), the system can identify potential performance bottlenecks, failure points, and security risks, and issue warnings.

[0125] Step 4.3: Performance optimization suggestions

[0126] Automatically generate optimization suggestions based on the analysis results.

[0127] (1) Route optimization: When a path is detected to have high latency, the system will recommend adjusting the routing strategy.

[0128] (2) Bandwidth optimization: When insufficient bandwidth is detected, the system will recommend increasing the bandwidth.

[0129] (3) Resource allocation optimization: When an imbalance in resource allocation is detected, the system will recommend reallocating resources.

[0130] Preferably, the active network performance detection method of this implementation also includes visual reporting and decision support, which can be implemented by the following steps:

[0131] Step 5.1: Generate a visualization report

[0132] (1) Report content: Generates a detailed network performance report, including real-time performance indicators, historical data comparison, abnormal event list, optimization suggestions, etc.

[0133] Real-time performance indicators: Display real-time performance indicators such as network latency, bandwidth utilization, and packet loss rate in the form of graphs.

[0134] Historical data comparison: Displays the comparison between historical data and real-time data to help administrators understand the changing trend of network performance.

[0135] Abnormal event list: lists all detected abnormal events and provides detailed analysis results.

[0136] Optimization suggestions: List the optimization suggestions generated by the system and provide specific implementation plans.

[0137] (2) Graphical display: Use visualization tools such as ECharts and D3.js to display report contents in an intuitive and easy-to-understand graphical way.

[0138] Step 5.2: Decision Support System

[0139] Combining artificial intelligence and big data analysis technology, it provides network administrators with intelligent decision-making basis and optimization solutions.

[0140] (1) Automated decision-making: The system can automatically generate decision plans based on the analysis results. For example, when a high network delay is detected, the system will automatically adjust the routing strategy.

[0141] (2) Manual intervention: Administrators can conduct manual intervention based on the decision-making plan provided by the system to ensure the accuracy and effectiveness of the decision.

[0142] Step 5.3: Customized service

[0143] Provide personalized network performance monitoring and analysis solutions based on user needs and preferences.

[0144] (1) Report content customization: Users can select the content and format of the report to ensure that the report meets their specific management needs.

[0145] (2) Report frequency customization: Users can select the frequency of report generation, such as real-time reports, daily reports, weekly reports, etc.

[0146] In summary, the active network performance detection method of this embodiment has the following characteristics:

[0147] 1. The design and generation method of intelligent detection packets, including the design of dynamically configurable packet headers, intelligent payloads and efficient checksums.

[0148] 2. Implementation of dynamic routing and sending strategies, including advanced routing algorithms, dynamic scheduling and load balancing technologies.

[0149] 3. Distributed response data collection and processing process, including real-time capture, storage, cleaning, deduplication, aggregation and formatting of data.

[0150] 4. Intelligent performance analysis and evaluation algorithms, including deep learning algorithms, big data analysis techniques, and generation of performance optimization recommendations.

[0151] It can be seen that the active network performance detection method of this embodiment has the following advantages:

[0152] 1. Compared with traditional methods, the present invention has higher monitoring accuracy and real-time performance, and can promptly discover and solve network performance problems.

[0153] 2. Through intelligent detection and dynamic routing technology, in-depth monitoring and comprehensive coverage of network performance are achieved, and accurate evaluation can be obtained regardless of complex network environments or specific time periods.

[0154] 3. Integrate artificial intelligence and big data analysis technologies to provide intelligent performance optimization suggestions and resource allocation strategies, which help improve the overall performance of the network and user experience.

[0155] 4. The visualization report and decision support system provided by the present invention provides network administrators with intuitive and easy-to-use decision-making basis and optimization solutions, reducing management difficulty and cost.

[0156] Example 2

[0157] This embodiment provides an active network performance detection system, including:

[0158] An intelligent detection packet generation module is configured to generate different types of detection packets based on monitoring requirements and historical data;

[0159] The dynamic routing and sending module is configured to select the best routing path and sending strategy based on the network topology, real-time traffic and monitoring requirements, and send the detection packet to the target network;

[0160] The distributed response data collection module is configured at the key nodes of the target network to capture the response data of the detection packet in real time and perform preliminary processing;

[0161] The intelligent performance analysis module is configured to perform deep mining and intelligent analysis on the response data after preliminary processing based on deep learning and big data analysis technology, extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies.

[0162] Preferably, in the intelligent detection package generation module, different types of detection packages are generated based on monitoring requirements and historical data, including:

[0163] Set the detection packet format and verification algorithm. The packet header of the detection packet includes a type field, a priority field, a timestamp field, and a verification code field.

[0164] Use machine learning algorithms to analyze historical data and real-time monitoring needs, generate different types of detection packets, and dynamically adjust the generation frequency of detection packets according to network conditions;

[0165] The encryption algorithm is dynamically selected according to the security level of the network and the sensitivity of the detection packet; a dynamic key management mechanism is adopted to ensure that the encryption key of each detection packet is different.

[0166] Preferably, in the dynamic routing and sending module, based on the network topology, real-time traffic and monitoring requirements, the best routing path and sending strategy are selected to send the detection packet to the target network, including:

[0167] Intelligent routing selection: According to network topology, real-time traffic and monitoring requirements, the best routing path is selected through routing algorithms; traffic changes in the target network are monitored in real time, the sending path of the detection packet is dynamically adjusted, and the routing table is updated regularly;

[0168] Dynamic scheduling and load balancing: A load balancing algorithm is used to evenly distribute the detection packets in the target network, and the sending order and frequency of the detection packets are dynamically adjusted according to the network load. High-priority detection packets are sent first when the network is congested.

[0169] Adaptive sending frequency adjustment: Dynamically adjust the sending frequency of the detection packet according to the network status and monitoring results, reduce the frequency when the load is high, and increase the frequency when the load is low.

[0170] Preferably, in the distributed response data collection module, the distributed response data collection module is deployed at the key nodes of the target network to capture the response data of the detection packet in real time and perform preliminary processing, including:

[0171] Distributed data collection: Deploy distributed response data collection modules at each key node of the target network, covering all key paths of the target network, and capturing the response data of the detection packet in real time;

[0172] Data cleaning and preprocessing: Use big data processing technology to clean, deduplicate, aggregate and format the collected response data, and format the cleaned data into a unified format;

[0173] Real-time transmission and distributed storage: Use message queue technology to achieve real-time data transmission, and use distributed database to store response data.

[0174] Preferably, in the intelligent performance analysis module, based on deep learning and big data analysis technology, the response data after preliminary processing is deeply mined and intelligently analyzed to extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies, including:

[0175] Deep learning and big data analysis: Use deep learning algorithms to conduct in-depth mining and intelligent analysis of the processed response data, including delay analysis, bandwidth analysis, and packet loss rate analysis; use big data analysis technology to conduct real-time analysis of massive response data and extract key performance indicators;

[0176] Real-time evaluation and fault warning: Real-time evaluation of network performance and generation of performance reports; identification of potential performance bottlenecks, fault points, and security risks through machine learning models, and issuance of warnings;

[0177] Performance optimization suggestions: Automatically generate optimization suggestions based on the analysis results. When it is detected that the delay of a certain path is higher than the threshold, it is recommended to adjust the routing policy; when it is detected that the bandwidth is insufficient, it is recommended to increase the bandwidth; when it is detected that the resource allocation is unbalanced, it is recommended to reallocate resources.

[0178] Therefore, the active network performance detection method of this embodiment has the following characteristics:

[0179] 1. The specific format and generation method of the intelligent detection package, including the implementation details of dynamic configuration and adaptive encryption technology.

[0180] 2. The core algorithms and implementation details of dynamic routing and sending strategies, including routing path selection, sending frequency adjustment and load balancing technology.

[0181] 3. The specific deployment and data processing flow of the distributed response data collection module, including real-time data transmission, distributed storage and security protection mechanisms.

[0182] 4. The deep learning algorithms, big data analysis techniques, and methods for generating performance optimization recommendations used in the intelligent performance analysis module.

[0183] In summary, the present invention realizes in-depth, real-time, and intelligent detection of network performance by integrating artificial intelligence, big data analysis, and advanced routing strategies. The system not only improves the accuracy and real-time performance of network performance monitoring, but also provides network administrators with intelligent decision-making basis and optimization solutions, which helps to improve the overall performance, user experience, and security of the network. Therefore, the present invention has significant technical advantages and broad application prospects, and is an indispensable network performance monitoring and analysis tool in the modern network environment.

[0184] Example 3

[0185] This embodiment is based on embodiment 1:

[0186] This embodiment provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the active network performance detection method of Embodiment 1 is implemented. The computer program may be in source code form, object code form, executable file, or some intermediate form.

[0187] Example 4

[0188] This embodiment is based on embodiment 1:

[0189] This embodiment provides a computer-readable storage medium storing a computer program, which implements the active network performance detection method of Embodiment 1 when executed by a processor. The computer program may be in source code form, object code form, executable file, or some intermediate form. The storage medium includes: any entity or device capable of carrying computer program code, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content contained in the storage medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the storage medium does not include electric carrier signals and telecommunication signals.

[0190] It should be noted that, for the above method embodiments, for the sake of simplicity of description, they are expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

Claims

1. An active network performance detection method, characterized in that: include: Generate different types of detection packages based on monitoring requirements and historical data; Based on network topology, real-time traffic and monitoring requirements, the best routing path and sending strategy are selected to send the detection packet to the target network; Deploy distributed response data collection modules at key nodes of the target network to capture the response data of the detection package in real time and perform preliminary processing; Based on deep learning and big data analysis technology, the response data after preliminary processing is deeply mined and intelligently analyzed to extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies.

2. The active network performance detection method according to claim 1, characterized in that: Based on monitoring requirements and historical data, different types of detection packages are generated, including: Setting the detection packet format and verification algorithm, wherein the packet header content of the detection packet includes a type field, a priority field, a timestamp field and a verification code field; Use machine learning algorithms to analyze historical data and real-time monitoring needs, generate different types of detection packets, and dynamically adjust the generation frequency of detection packets according to network conditions; The encryption algorithm is dynamically selected according to the security level of the network and the sensitivity of the detection packet; a dynamic key management mechanism is adopted to ensure that the encryption key of each detection packet is different.

3. The active network performance detection method according to claim 1, characterized in that: The method of selecting the best routing path and sending strategy based on network topology, real-time traffic and monitoring requirements, and sending the detection packet to the target network includes: Intelligent routing selection: According to network topology, real-time traffic and monitoring requirements, the best routing path is selected through routing algorithms; traffic changes in the target network are monitored in real time, the sending path of the detection packet is dynamically adjusted, and the routing table is updated regularly; Dynamic scheduling and load balancing: A load balancing algorithm is used to evenly distribute the detection packets in the target network, and the sending order and frequency of the detection packets are dynamically adjusted according to the network load. High-priority detection packets are sent first when the network is congested. Adaptive sending frequency adjustment: Dynamically adjust the sending frequency of the detection packet according to the network status and monitoring results, reduce the frequency when the load is high, and increase the frequency when the load is low.

4. The active network performance detection method according to claim 1, characterized in that: The distributed response data collection module is deployed at the key nodes of the target network to capture the response data of the detection packet in real time and perform preliminary processing, including: Distributed data collection: Deploy distributed response data collection modules at each key node of the target network, covering all key paths of the target network, and capturing the response data of the detection packet in real time; Data cleaning and preprocessing: Use big data processing technology to clean, deduplicate, aggregate and format the collected response data, and format the cleaned data into a unified format; Real-time transmission and distributed storage: Use message queue technology to achieve real-time data transmission, and use distributed database to store response data.

5. The active network performance detection method according to claim 1, characterized in that: Based on deep learning and big data analysis technology, the response data after preliminary processing is deeply mined and intelligently analyzed to extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies, including: Deep learning and big data analysis: Use deep learning algorithms to conduct in-depth mining and intelligent analysis of the processed response data, including delay analysis, bandwidth analysis, and packet loss rate analysis; use big data analysis technology to conduct real-time analysis of massive response data and extract key performance indicators; Real-time evaluation and fault warning: Real-time evaluation of network performance and generation of performance reports; identification of potential performance bottlenecks, fault points, and security risks through machine learning models, and issuance of warnings; Performance optimization suggestions: Automatically generate optimization suggestions based on the analysis results. When it is detected that the delay of a certain path is higher than the threshold, it is recommended to adjust the routing policy; when it is detected that the bandwidth is insufficient, it is recommended to increase the bandwidth; when it is detected that the resource allocation is unbalanced, it is recommended to reallocate resources.

6. An active network performance detection system, characterized in that: include: An intelligent detection packet generation module is configured to generate different types of detection packets based on monitoring requirements and historical data; The dynamic routing and sending module is configured to select the best routing path and sending strategy based on the network topology, real-time traffic and monitoring requirements, and send the detection packet to the target network; The distributed response data collection module is configured at the key nodes of the target network to capture the response data of the detection packet in real time and perform preliminary processing; The intelligent performance analysis module is configured to perform deep mining and intelligent analysis on the response data after preliminary processing based on deep learning and big data analysis technology, extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies.

7. The active network performance detection system according to claim 6, characterized in that: In the intelligent detection package generation module, different types of detection packages are generated based on monitoring requirements and historical data, including: Setting the detection packet format and verification algorithm, wherein the packet header content of the detection packet includes a type field, a priority field, a timestamp field and a verification code field; Use machine learning algorithms to analyze historical data and real-time monitoring needs, generate different types of detection packets, and dynamically adjust the generation frequency of detection packets according to network conditions; The encryption algorithm is dynamically selected according to the security level of the network and the sensitivity of the detection packet; a dynamic key management mechanism is adopted to ensure that the encryption key of each detection packet is different.

8. The active network performance detection system according to claim 6, characterized in that: In the dynamic routing and sending module, based on network topology, real-time traffic and monitoring requirements, the best routing path and sending strategy are selected to send the detection packet to the target network, including: Intelligent routing selection: According to network topology, real-time traffic and monitoring requirements, the best routing path is selected through routing algorithms; traffic changes in the target network are monitored in real time, the sending path of the detection packet is dynamically adjusted, and the routing table is updated regularly; Dynamic scheduling and load balancing: A load balancing algorithm is used to evenly distribute the detection packets in the target network, and the sending order and frequency of the detection packets are dynamically adjusted according to the network load. High-priority detection packets are sent first when the network is congested. Adaptive sending frequency adjustment: Dynamically adjust the sending frequency of the detection packet according to the network status and monitoring results, reduce the frequency when the load is high, and increase the frequency when the load is low.

9. The active network performance detection system according to claim 6, characterized in that: In the distributed response data collection module, the distributed response data collection module is deployed at the key nodes of the target network to capture the response data of the detection packet in real time and perform preliminary processing, including: Distributed data collection: Deploy distributed response data collection modules at each key node of the target network, covering all key paths of the target network, and capturing the response data of the detection packet in real time; Data cleaning and preprocessing: Use big data processing technology to clean, deduplicate, aggregate and format the collected response data, and format the cleaned data into a unified format; Real-time transmission and distributed storage: Use message queue technology to achieve real-time data transmission, and use distributed database to store response data.

10. The active network performance detection system according to claim 6, characterized in that: In the intelligent performance analysis module, based on deep learning and big data analysis technology, the response data after preliminary processing is deeply mined and intelligently analyzed to extract key performance indicators and abnormal patterns, and generate performance optimization solutions and resource allocation strategies, including: Deep learning and big data analysis: Use deep learning algorithms to conduct in-depth mining and intelligent analysis of the processed response data, including delay analysis, bandwidth analysis, and packet loss rate analysis; use big data analysis technology to conduct real-time analysis of massive response data and extract key performance indicators; Real-time evaluation and fault warning: Real-time evaluation of network performance and generation of performance reports; identification of potential performance bottlenecks, fault points, and security risks through machine learning models, and issuance of warnings; Performance optimization suggestions: Automatically generate optimization suggestions based on the analysis results. When it is detected that the delay of a certain path is higher than the threshold, it is recommended to adjust the routing policy; when it is detected that the bandwidth is insufficient, it is recommended to increase the bandwidth; when it is detected that the resource allocation is unbalanced, it is recommended to reallocate resources.

Images