Network connection automatic tuning system and device based on SRv6 protocol

Abstract

The invention relates to the technical field of communication routing, in particular to a network connection automatic tuning system and device based on an SRv6 protocol. According to the method, the state data of network equipment is collected in real time, key network indexes such as delay, bandwidth, packet loss rate and hardware performance are combined, the health condition of network connection is dynamically evaluated, and path optimization is carried out in combination with an SRv6 protocol. By automatically calculating a path scoring matrix, the system can automatically select an optimal network path according to the priority of network transmission tasks, and intelligent adjustment and optimization are realized under the conditions of dynamic traffic and complex tasks; the invention provides a path calculation and scheduling mechanism based on an SRv6 protocol, the priority of a network path is automatically adjusted according to different types of data transmission tasks, and the selection of an optimal path is ensured. Meanwhile, the system has an automatic switching mechanism, and can automatically switch to an alternative path when the network path has a fault, thereby improving the reliability and fault-tolerant capability of the network.

Description

Technical Field

[0001] The present invention relates to the technical field of communication routing, and in particular to a network connection automatic tuning system and device based on the SRv6 protocol. Background Art

[0002] With the continuous development of Internet applications, the scale and complexity of network traffic are increasing, and traditional network routing technology faces many challenges. Especially in large-scale enterprise networks and service provider networks, how to efficiently and flexibly perform traffic scheduling, path selection and real-time monitoring has become an urgent problem to be solved. Although the traditional network routing method based on the IP protocol has achieved relatively stable transmission, with the increasing complexity of the network environment, traditional routing technology has been unable to cope with the diverse traffic demands and dynamic network environment.

[0003] In view of the above problems, it is necessary to propose a network connection automatic tuning system and device based on the SRv6 protocol. Summary of the invention

[0004] The purpose of the present invention is to solve the problems existing in the background technology and to propose a network connection automatic tuning system and device based on the SRv6 protocol.

[0005] The purpose of the present invention can be achieved through the following technical solutions: In a first aspect, a network connection automatic tuning system based on the SRv6 protocol includes a status monitoring module, a path calculation module, a traffic optimization module, a network scheduling module, and a visualization and alarm module.

[0006] The status monitoring module collects the status data of network devices in real time, collects key network indicators of each network connection including latency, bandwidth utilization, packet loss rate and hardware performance, and evaluates the health of the current network connection through key network indicators.

[0007] All network devices of the target network are numbered with the symbol i, i=1, 2, ..., n; where n is the total number of network devices. If there is a direct communication link between two network devices, the communication link is recorded as a network connection, and all network connections are numbered with the symbol (i1, i2); where i1 and i2 are the network device number symbols at both ends of the network connection, where i1∈i, i2∈i and i1≠i2.

[0008] As a preferred embodiment of the present invention, key network indicators of all network devices are collected in real time.

[0009] Get the difference between the timestamp and the actual arrival time of all packets of all network devices i, Delayi, and extract its maximum value MaxDelayi. Get the queue length of the received packets , queue bandwidth , Packet processing capacity and packet processing capabilities . Through the preset formula Calculate the delay characteristic score Di of each network device i; wherein k1, k2 and k3 are preset weight factors, k1, k2 and k3 are all positive numbers ranging from 0 to 1 and k1+k2+k3=1.

[0010] Get the network bandwidth upper limit Ci of all network devices i, and monitor the total flow Ti of all network device interfaces in real time, through the preset formula Calculate the bandwidth utilization characteristic score value Bi of each network device.

[0011] Get the number of packets lost by all network devices i and the total number of packets actually received , through the preset formula Calculate the packet loss feature score Li of each network device.

[0012] Get the total CPU resources of all network devices i , the CPU resources currently used , Total memory and used memory . Through the preset formula Calculate the hardware comprehensive feature score Hi; k4 and k5 are both preset weight factors, k4 and k5 are both positive numbers ranging from 0 to 1, and k4+k5=1.

[0013] As a preferred embodiment of the present invention, key indicators of all network connections are analyzed in real time in combination with key data of network devices.

[0014] For each network connection (i1, i2), generate a connection score matrix ; Where d(i1, i2) is the distance between network devices i1 and i2, and E(i1) and E(i2) are the type characteristic values ​​of network devices i1 and i2 respectively. Set the value rules of type characteristic values: the value range of type characteristic values ​​is 0 to 1; a type characteristic value of 0 means that the device is very unfavorable to network transmission and greatly affects network performance; a type characteristic value of 1 means that the device greatly supports network transmission and has no adverse effect on network performance. Specifically: Let the specific value of the type characteristic value of the high-performance router be 1; Let the specific value of the type characteristic value of the network switch be 1; Let the specific value of the gateway type characteristic value be 1; Let the specific value of the type characteristic value of the wireless network access point be 0.8; Let the specific value of the type characteristic value of the low-performance router be 0.8; Let the specific value of the type characteristic value of the virtual router be 0.6; Let the specific value of the type characteristic value of the high-performance firewall be 0.6; Let the specific value of the characteristic value of the type of low-performance firewall be 0.4; Let the specific value of the proxy server type characteristic value be 0.4 Let the specific value of the characteristic value of the type of high-performance terminal equipment including servers and workstations be 0.2; Let the specific numerical value of the category feature value including camera, mobile phone, and personal computer be 0.

[0015] As a preferred embodiment of the present invention, the path scoring matrix is ​​dynamically updated, key network indicators of network devices and network connections are recollected at preset time intervals, the path scoring matrix of each network connection is recalculated, and the path scoring matrix of each network connection (i1, i2) obtained at each preset time interval is sent to the path calculation module.

[0016] The path calculation module obtains the path scoring matrix sent by the status monitoring module. The starting point i-strat and the end point i-end of all network transmission tasks in the target network are obtained, and the network path is defined: the set of all available network paths from the network transmission starting point i-strat to the end point i-end is called a network path, and each network path passes through multiple network devices and consists of multiple network connections.

[0017] The set of all network device IDs in the same network path is denoted as I.

[0018] Define a preset optimal network path optimization strategy, where the optimal transmission path includes the shortest path, the lowest delay path, the maximum bandwidth path and the most reliable path.

[0019] The goal of the shortest path is to find the shortest physical distance or network hop count between devices, and its objective function is: ; α1 is the target value of the path distance, which represents the total physical distance of the network path and reflects the degree of optimization of the shortest path; The goal of the lowest latency path is to minimize the delay experienced by the data packet from the source device to the destination device, and its objective function is: ; α2 is the delay path target value, which represents the sum of the delay characteristic score values ​​of the network path and reflects the degree of optimization of the minimum delay; The goal of the maximum bandwidth path is to select the path that can provide the maximum bandwidth, and its objective function is: ; Where α3 is the bandwidth path target value, which represents the inverse of the sum of the bandwidth characteristic score values ​​of the network path and reflects the degree of optimization of the maximum bandwidth.

[0020] Among them, the goal of the most reliable path is to maximize the reliability of the path, which is usually determined by the comprehensive feature score of the hardware on the path. Its objective function is: . Where α4 is the reliable path target value, which represents the sum of the hardware comprehensive feature score values ​​and reflects the optimization degree of the most reliable path.

[0021] Set the final optimization objective function of the network path for each network transmission task: ; where λ1, λ2, λ3 and λ4 are the priority weights of the objective function, all of which are positive values ​​ranging from 0 to 1, and the sum of λ1, λ2, λ3 and λ4 is 1.

[0022] Set basic constraints for the shortest path, lowest latency path, maximum bandwidth path, and most reliable path, including: No loop constraint: Each network device and network connection can only appear once on the path, that is, in the shortest path, for any two network connections (i1, i2) and (i3, i4), the norm of their connection score matrices is not equal, satisfying the condition .in and are the norms of the connection score matrices.

[0023] Network connection reachability constraint: d(i1, i2) has a value only when the distance condition between network devices i1 and i2 and the network device type condition meet the minimum requirements of network communication, that is, Where dmax is the preset maximum distance threshold, and Emin is the preset minimum category feature value threshold.

[0024] The traffic optimization module obtains the data type and task type of the network transmission task, and determines the priority weights λ1, λ2, λ3 and λ4 of the objective function in the final optimization objective function of the network path.

[0025] When the network transmission task contains regular types of data including text, code, and pictures, obtain the path length tolerance Dmax of data transmission and the maximum path distance target value maxα1 among all available network paths; otherwise, directly set the value of λ1 to 0.25; When the network transmission task contains data of the type of voice and video information, obtain the delay tolerance Tmax of data transmission and the maximum delay path target value maxα2 of all available network paths; otherwise, directly set the value of λ2 to 0.25 When the network transmission task contains large files and high-definition video streams whose data size is larger than the preset threshold, the bandwidth requirement tolerance Bmax of data transmission and the maximum bandwidth path target value maxα3 of all available network paths are obtained; otherwise, the value of λ3 is directly set to 0.25 When the network transmission task includes the network transmission task of encrypted data, obtain the preset task reliability requirement Hmax and the maximum reliable path target value α4 of all available network paths. Otherwise, directly set the value of λ4 to 0.25 As a preferred embodiment of the present invention, by presetting the formula Calculate the specific values ​​of the unassigned priority weights in λ1, λ2, λ3, and λ4.

[0026] The network scheduling module defines and collects network environment data, assigns priority to the traffic path adjustment plan formulated by the traffic optimization module, and schedules different traffic based on priority to ensure that high-priority traffic is given priority.

[0027] According to the path calculation results, formulate a traffic path adjustment plan and retrieve the specific values ​​of priority weights λ1, λ2, λ3, and λ4.

[0028] By presetting the objective function The selection priority Pro of each network transmission task on each network path is calculated, and each network path is arranged in descending order according to the numerical value of the selection priority.

[0029] As a preferred embodiment of the present invention, the priority ranking of all network paths is converted into an SRv6SID list. An SRv6SID list is created for each specific network transmission task, and each row in the list corresponds to an available network path. The top of the list is the network path with the maximum selection priority maxPro, followed by the network path with the second largest selection priority, and so on. The end of the SRv6SID list is the network path with the minimum selection priority minPro.

[0030] As a preferred embodiment of the present invention, network paths are arranged for all network transmission tasks according to the arrangement order in the SRv6 SID list, and a backup path is selected through a preset automatic switching mechanism.

[0031] Among them, the automatic switching mechanism is specifically as follows: when the network path with the maximum selection priority maxPro is unavailable, the system should select the path with the second largest selection priority as the new transmission path based on the list of other candidate paths provided by the path calculation module. The selection of candidate paths should be dynamically adjusted according to the specific values ​​of the priority weights λ1, λ2, λ3 and λ4 of the traffic optimization module. The performance of the second best path is evaluated according to the specific values ​​of the priority weights λ1, λ2, λ3 and λ4 to ensure that the selection priority Pro of the path is greater than the preset threshold, otherwise the second path is determined to be unavailable. If the second best path is also unavailable, continue to try the third best path until an available path is found.

[0032] As a preferred embodiment of the present invention, when it is identified that the automatic switching mechanism is triggered in the SRv6 SID list, a path switching event signal is sent to the visualization and alarm module.

[0033] The visualization and alarm module displays and outputs a list of all SRv6 SIDs and assigns a unique corresponding coder to each network transmission task and available network path.

[0034] As a preferred method of the present invention, after the path switching event signal is identified, the corresponding SRv6SID list, as well as the network transmission task and the specific network path code are located and highlighted, so that the administrator or relevant personnel are notified of the changes in the current network path in a timely manner, and an alarm message is issued for the unavailable path, so as to facilitate further processing by the staff.

[0035] As a preferred method of the present invention, the delay feature score value, bandwidth feature score value, packet loss feature score value and hardware comprehensive feature score value of each network device are recorded, the path switching time signal and the change record of all SRv6SID lists are recorded, and a detailed event log is generated to help administrators trace back the cause of the failure.

[0036] In a second aspect, the present invention provides a network connection automatic tuning device based on the SRv6 protocol, comprising an SRv6 server group and a target network, wherein the SRv6 server group comprises a plurality of servers, high-performance routers, switches, firewalls, gateways and wireless access points.

[0037] Among them, all servers jointly run the status monitoring module, path calculation module, traffic optimization module, network scheduling module and visualization and alarm module, and all servers are connected through a high-speed data bus.

[0038] The data collected by the status monitoring module will be transmitted to the path calculation module in real time for path optimization calculation; the calculation results will be transmitted to the traffic optimization module to adjust the path selection weight, and finally to the network scheduling module for priority scheduling. The visualization and alarm module obtains relevant data from each module and displays it on the monitoring panel.

[0039] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention collects the status data of network devices in real time, combines key network indicators such as latency, bandwidth, packet loss rate and hardware performance, dynamically evaluates the health of network connections, and optimizes paths in combination with the SRv6 protocol. By automatically calculating the path scoring matrix, the system can automatically select the optimal network path according to the priority of network transmission tasks, ensuring efficient and stable network performance, especially realizing intelligent tuning in dynamic traffic and complex tasks; 2. The present invention provides a path calculation and scheduling mechanism based on the SRv6 protocol, which can automatically adjust the priority of the network path according to different types of data transmission tasks and ensure the selection of the optimal path. At the same time, the system has an automatic switching mechanism, which can automatically switch to an alternative path when a network path fails, thereby improving the reliability and fault tolerance of the network; 3. The present invention monitors the health of network connections in real time and promptly identifies abnormal changes in network paths through the collaborative work of the status monitoring module and the visualization and alarm module. When a path switching event is triggered, the system will notify the administrator through the alarm module and record detailed logs of the path switching to help the administrator quickly locate the cause of the fault and handle it. This mechanism enhances the flexibility and response speed of network management. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings: Figure 1 is a system block diagram of the present invention; Figure 2 This is a schematic diagram of the topology structure of a network connection automatic tuning device based on the SRv6 protocol of the present invention. DETAILED DESCRIPTION

[0041] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] See also Figure 1As shown, a network connection automatic tuning system based on the SRv6 protocol includes a status monitoring module, a path calculation module, a traffic optimization module, a network scheduling module and a visualization and alarm module.

[0043] The status monitoring module collects the status data of network devices in real time, collects key network indicators of each network connection including latency, bandwidth utilization, packet loss rate and hardware performance, and evaluates the health of the current network connection through key network indicators.

[0044] All network devices of the target network are numbered with the symbol i, i=1, 2, ..., n; where n is the total number of network devices. If there is a direct communication link between two network devices, the communication link is recorded as a network connection, and all network connections are numbered with the symbol (i1, i2); where i1 and i2 are the network device number symbols at both ends of the network connection, where i1∈i, i2∈i and i1≠i2.

[0045] Furthermore, key network indicators of all network devices are collected in real time.

[0046] Get the difference between the timestamp and the actual arrival time of all packets of all network devices i, Delayi, and extract its maximum value MaxDelayi. Get the queue length of the received packets , queue bandwidth , Packet processing capacity and packet processing capabilities . Through the preset formula Calculate the delay characteristic score Di of each network device i; wherein k1, k2 and k3 are preset weight factors, k1, k2 and k3 are all positive numbers ranging from 0 to 1 and k1+k2+k3=1.

[0047] Get the network bandwidth upper limit Ci of all network devices i, and monitor the total flow Ti of all network device interfaces in real time, through the preset formula Calculate the bandwidth utilization characteristic score value Bi of each network device.

[0048] Get the number of packets lost by all network devices i and the total number of packets actually received , through the preset formula Calculate the packet loss feature score Li of each network device.

[0049] Get the total CPU resources of all network devices i , the CPU resources currently used , Total memory and used memory . Through the preset formula Calculate the hardware comprehensive feature score Hi; k4 and k5 are both preset weight factors, k4 and k5 are both positive numbers ranging from 0 to 1, and k4+k5=1.

[0050] Furthermore, the key indicators of all network connections are analyzed in real time in combination with the key data of network devices.

[0051] For each network connection (i1, i2), generate a connection score matrix ; Where d(i1, i2) is the distance between network devices i1 and i2, and E(i1) and E(i2) are the type characteristic values ​​of network devices i1 and i2 respectively. Set the value rules of type characteristic values: the value range of type characteristic values ​​is 0 to 1; a type characteristic value of 0 means that the device is very unfavorable to network transmission and greatly affects network performance; a type characteristic value of 1 means that the device greatly supports network transmission and has no adverse effect on network performance. Specifically: Let the specific value of the type characteristic value of the high-performance router be 1; Let the specific value of the type characteristic value of the network switch be 1; Let the specific value of the gateway type characteristic value be 1; Let the specific value of the type characteristic value of the wireless network access point be 0.8; Let the specific value of the type characteristic value of the low-performance router be 0.8; Let the specific value of the type characteristic value of the virtual router be 0.6; Let the specific value of the type characteristic value of the high-performance firewall be 0.6; Let the specific value of the characteristic value of the type of low-performance firewall be 0.4; Let the specific value of the proxy server type characteristic value be 0.4 Let the specific value of the characteristic value of the type of high-performance terminal equipment including servers and workstations be 0.2; Let the specific numerical value of the category feature value including camera, mobile phone, and personal computer be 0.

[0052] It should be noted that in the network connection automatic tuning system based on the SRv6 protocol, the category characteristic value is a numerical value assigned to each device after classifying different network devices, which is used to measure the impact of the device on the network connection performance.

[0053] Furthermore, the path scoring matrix is ​​dynamically updated, and the key network indicators of network devices and network connections are recollected at preset time intervals, the path scoring matrix of each network connection is recalculated, and the path scoring matrix of each network connection (i1, i2) obtained at each preset time interval is sent to the path calculation module.

[0054] The path calculation module obtains the path scoring matrix sent by the status monitoring module. The starting point i-strat and the end point i-end of all network transmission tasks in the target network are obtained, and the network path is defined: the set of all available network paths from the network transmission starting point i-strat to the end point i-end is called a network path, and each network path passes through multiple network devices and consists of multiple network connections.

[0055] The set of all network device IDs in the same network path is denoted as I.

[0056] Define a preset optimal network path optimization strategy, where the optimal transmission path includes the shortest path, the lowest delay path, the maximum bandwidth path and the most reliable path.

[0057] The goal of the shortest path is to find the shortest physical distance or network hop count between devices, and its objective function is: ; α1 is the target value of the path distance, which represents the total physical distance of the network path and reflects the degree of optimization of the shortest path; The goal of the lowest latency path is to minimize the delay experienced by the data packet from the source device to the destination device, and its objective function is: ; α2 is the delay path target value, which represents the sum of the delay characteristic score values ​​of the network path and reflects the degree of optimization of the minimum delay; The goal of the maximum bandwidth path is to select the path that can provide the maximum bandwidth, and its objective function is: ; Where α3 is the bandwidth path target value, which represents the inverse of the sum of the bandwidth characteristic score values ​​of the network path and reflects the degree of optimization of the maximum bandwidth.

[0058] Among them, the goal of the most reliable path is to maximize the reliability of the path, which is usually determined by the comprehensive feature score of the hardware on the path. Its objective function is: . Where α4 is the reliable path target value, which represents the sum of the hardware comprehensive feature score values ​​and reflects the optimization degree of the most reliable path.

[0059] Set the final optimization objective function of the network path for each network transmission task: ; where λ1, λ2, λ3 and λ4 are the priority weights of the objective function, all of which are positive values ​​ranging from 0 to 1, and the sum of λ1, λ2, λ3 and λ4 is 1.

[0060] Set basic constraints for the shortest path, lowest latency path, maximum bandwidth path, and most reliable path, including: No loop constraint: Each network device and network connection can only appear once on the path, that is, in the shortest path, for any two network connections (i1, i2) and (i3, i4), the norm of their connection score matrices is not equal, satisfying the condition .in and are the norms of the connection score matrices.

[0061] Network connection reachability constraint: d(i1, i2) has a value only when the distance condition between network devices i1 and i2 and the network device type condition meet the minimum requirements of network communication, that is, Where dmax is the preset maximum distance threshold, and Emin is the preset minimum category feature value threshold.

[0062] The traffic optimization module obtains the data type and task type of the network transmission task, and determines the priority weights λ1, λ2, λ3 and λ4 of the objective function in the final optimization objective function of the network path.

[0063] When the network transmission task contains regular types of data such as text, code, and pictures, obtain the path length tolerance Dmax of data transmission and the maximum path distance target value maxα1 among all available network paths; otherwise, directly set the value of λ1 to 0.25; When the network transmission task contains data of the type of voice and video information, obtain the delay tolerance Tmax of data transmission and the maximum delay path target value maxα2 of all available network paths; otherwise, directly set the value of λ2 to 0.25 When the network transmission task contains large files and high-definition video streams whose data size is larger than the preset threshold, the bandwidth requirement tolerance Bmax of data transmission and the maximum bandwidth path target value maxα3 of all available network paths are obtained; otherwise, the value of λ3 is directly set to 0.25 When the network transmission tasks include those that require high reliability, such as financial transactions and medical data, the preset task reliability requirement Hmax and the maximum reliable path target value α4 of all available network paths are obtained. Otherwise, the value of λ4 is directly set to 0.25. Furthermore, by pre-setting the formula Calculate the specific values ​​of the unassigned priority weights in λ1, λ2, λ3, and λ4.

[0064] It should be noted that in actual deployment, the weights need to be adjusted dynamically according to the specific types and changes of network transmission tasks. The ratios of λ1, λ2, λ3, and λ4 can be adjusted by monitoring the real-time situation of network load and the real-time feedback of network transmission tasks to better cope with the changing network environment and task requirements. For example, for network transmission tasks containing large data files, by adjusting the specific value of λ3, network paths with matching bandwidth requirements can be preferentially selected to meet the actual needs of network transmission tasks.

[0065] The network scheduling module defines and collects network environment data, assigns priority to the traffic path adjustment plan formulated by the traffic optimization module, and schedules different traffic based on priority to ensure that high-priority traffic is given priority.

[0066] According to the path calculation results, formulate a traffic path adjustment plan and retrieve the specific values ​​of priority weights λ1, λ2, λ3, and λ4.

[0067] By presetting the objective function The selection priority Pro of each network transmission task on each network path is calculated, and each network path is arranged in descending order according to the numerical value of the selection priority.

[0068] Furthermore, the priority ranking of all network paths is converted into an SRv6SID list. An SRv6SID list is created for each specific network transmission task, and each row in the list corresponds to an available network path. The top of the list is the network path with the maximum selection priority maxPro, followed by the network path with the second largest selection priority, and so on. The end of the SRv6SID list is the network path with the minimum selection priority minPro.

[0069] Furthermore, network paths are arranged for all network transmission tasks according to the arrangement order in the SRv6SID list, and a backup path is selected through a preset automatic switching mechanism.

[0070] Among them, the automatic switching mechanism is specifically as follows: when the network path with the maximum selection priority maxPro is unavailable, the system should select the path with the second largest selection priority as the new transmission path based on the list of other candidate paths provided by the path calculation module. The selection of candidate paths should be dynamically adjusted according to the specific values ​​of the priority weights λ1, λ2, λ3 and λ4 of the traffic optimization module. The performance of the second best path is evaluated according to the specific values ​​of the priority weights λ1, λ2, λ3 and λ4 to ensure that the selection priority Pro of the path is greater than the preset threshold, otherwise the second path is determined to be unavailable. If the second best path is also unavailable, continue to try the third best path until an available path is found.

[0071] Furthermore, when it is identified that the automatic switching mechanism is triggered in the SRv6 SID list, a path switching event signal is sent to the visualization and alarm module.

[0072] The visualization and alarm module displays and outputs a list of all SRv6 SIDs and assigns a unique corresponding coder to each network transmission task and available network path.

[0073] Furthermore, after identifying the path switching event signal, the corresponding SRv6SID list, as well as the network transmission task and the specific network path code are located and highlighted, so that the administrator or relevant personnel are notified of the changes in the current network path in a timely manner, and an alarm message is issued for the unavailable path to facilitate further processing by the staff.

[0074] Furthermore, the delay feature score, bandwidth feature score, packet loss feature score and hardware comprehensive feature score of each network device are recorded, the path switching time signal and the change records of all SRv6SID lists are recorded, and a detailed event log is generated to help administrators trace the cause of the failure.

[0075] See also Figure 2 As shown, a network connection automatic tuning device based on the SRv6 protocol includes an SRv6 server group and a target network. The SRv6 server group includes several servers, high-performance routers, switches, firewalls, gateways and wireless access points.

[0076] Among them, all servers jointly run the status monitoring module, path calculation module, traffic optimization module, network scheduling module and visualization and alarm module, and all servers are connected through a high-speed data bus.

[0077] The data collected by the status monitoring module will be transmitted to the path calculation module in real time for path optimization calculation; the calculation results will be transmitted to the traffic optimization module to adjust the path selection weight, and finally to the network scheduling module for priority scheduling. The visualization and alarm module obtains relevant data from each module and displays it on the monitoring panel.

[0078] It should be understood that the terms “include” and “comprising” used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0079] It should also be understood that the terms used in this disclosure are only for the purpose of describing specific embodiments and are not intended to limit the disclosure. As used in this disclosure and the claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and the claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations; The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A network connection automatic tuning system based on the SRv6 protocol, comprising a status monitoring module, a path calculation module, a traffic optimization module and a network scheduling module, characterized in that: The status monitoring module collects the status data of network devices in real time, collects key network indicators of each network connection including latency, bandwidth utilization, packet loss rate and hardware performance, and evaluates the health of the current network connection through key network indicators; The path calculation module defines the network path and defines the preset optimal network path optimization strategy. The optimal transmission path includes the shortest path, the lowest delay path, the maximum bandwidth path and the most reliable path; the final optimization objective function of the network path including the priority weight of each network transmission task is set: The traffic optimization module obtains the data type and task type of the network transmission task, and determines the specific value of the priority weight of the objective function in the final optimization objective function of the network path; The network scheduling module defines and collects network environment data, assigns priority to the traffic path adjustment plan formulated by the traffic optimization module, schedules different traffic based on priority, creates an SRv6SID list, and ensures that high-priority traffic is given priority.

2. According to the SRv6 protocol-based network connection automatic tuning system of claim 1, it is characterized in that: It also includes visualization and alarm modules; The visualization and alarm module displays and outputs a list of all SRv6 SIDs and assigns a unique corresponding code to each network transmission task and available network path; After identifying the path switching event signal, locate the corresponding SRv6SID list, the network transmission task and the specific network path code, and highlight them, promptly notify the administrator or relevant personnel of the change of the current network path, and issue an alarm message for the unavailable path to facilitate further processing by the staff; Record the delay feature score, bandwidth feature score, packet loss feature score, and hardware comprehensive feature score of each network device, record the path switching time signal and the change records of all SRv6SID lists, and generate detailed event logs to help administrators trace the cause of the failure.

3. According to the SRv6 protocol-based network connection automatic tuning system of claim 1, it is characterized in that: The specific process of real-time collection of network device status data is as follows: All network devices of the target network are numbered with the symbol i, i=1, 2, ..., n; where n is the total number of network devices; if there is a direct communication link between two network devices, the communication link is recorded as a network connection, and all network connections are numbered with the symbol (i1, i2); where i1 and i2 are the network device number symbols at both ends of the network connection, where i1∈i, i2∈i and i1≠i2; Collect key network indicators of all network devices in real time; Get the difference between the timestamp and the actual arrival time of all packets of all network devices i, Delayi, and extract its maximum value MaxDelayi; get the queue length of the received packets , queue bandwidth , Packet processing capacity and packet processing capabilities ; Through the preset formula Calculate the delay characteristic score Di of each network device i; Wherein k1, k2 and k3 are preset weight factors, k1, k2 and k3 are all positive numbers ranging from 0 to 1 and k1+k2+k3=1; Get the network bandwidth upper limit Ci of all network devices i, and monitor the total flow Ti of all network device interfaces in real time, through the preset formula Calculate the bandwidth utilization characteristic score value Bi of each network device; Get the number of packets lost by all network devices i and the total number of packets actually received , through the preset formula Calculate the packet loss feature score Li of each network device; Get the total CPU resources of all network devices i , the CPU resources currently used , Total memory and used memory ; Through the preset formula Calculate the hardware comprehensive feature score Hi; k4 and k5 are both preset weight factors, k4 and k5 are both positive numbers ranging from 0 to 1, and k4+k5=1; Combine key data from network devices to analyze key indicators of all network connections in real time; For each network connection (i1, i2), generate a connection score matrix ; Where d(i1, i2) is the distance between network devices i1 and i2, where E(i1) and E(i2) are the type characteristic values ​​of network devices i1 and i2 respectively; Set the value rule of type characteristic value: the value range of type characteristic value is 0 to 1; the type characteristic value of 0 means that the device is very unfavorable to network transmission and greatly affects network performance; the type characteristic value of 1 means that the device greatly supports network transmission and has no adverse effect on network performance; The path scoring matrix is ​​dynamically updated. At preset time intervals, key network indicators of network devices and network connections are recollected, the path scoring matrix of each network connection is recalculated, and the path scoring matrix of each network connection (i1, i2) obtained at preset time intervals is sent to the path calculation module.

4. According to claim 3, a network connection automatic tuning system based on the SRv6 protocol is characterized in that: The specific rules for selecting the type characteristic value are as follows: Let the specific value of the type characteristic value of the high-performance router be 1; Let the specific value of the type characteristic value of the network switch be 1; Let the specific value of the gateway type characteristic value be 1; Let the specific value of the type characteristic value of the wireless network access point be 0.8; Let the specific value of the type characteristic value of the low-performance router be 0.8; Let the specific value of the type characteristic value of the virtual router be 0.6; Let the specific value of the type characteristic value of the high-performance firewall be 0.6; Let the specific value of the characteristic value of the type of low-performance firewall be 0.4; Let the specific value of the proxy server type characteristic value be 0.4 Let the specific value of the characteristic value of the type of high-performance terminal equipment including servers and workstations be 0.2; Let the specific numerical value of the category feature value including camera, mobile phone, and personal computer be 0.

5. According to the SRv6 protocol-based network connection automatic tuning system of claim 1, it is characterized in that: The specific process of defining a network path is as follows: Get the starting point i-strat and end point i-end of all network transmission tasks in the target network, and define the network path: the set of all available network paths from the network transmission starting point i-strat to the end point i-end is called a network path. Each network path passes through multiple network devices and consists of multiple network connections. The set of all network device IDs in the same network path is denoted as I.

6. The network connection automatic tuning system based on the SRv6 protocol according to claim 1 is characterized in that: The shortest path, lowest latency path, maximum bandwidth path, and most reliable path are: The goal of the shortest path is to find the shortest physical distance or network hop count between devices. Its objective function is: ; α1 is the target value of the path distance, which represents the total physical distance of the network path and reflects the degree of optimization of the shortest path; The goal of the minimum latency path is to minimize the latency experienced by the data packet from the source device to the destination device. Its objective function is: ; α2 is the delay path target value, which represents the sum of the delay characteristic score values ​​of the network path and reflects the degree of optimization of the minimum delay; The goal of the maximum bandwidth path is to select the path that can provide the maximum bandwidth. Its objective function is: ; α3 is the bandwidth path target value, which represents the inverse of the sum of the bandwidth characteristic score values ​​of the network path and reflects the degree of optimization of the maximum bandwidth; The goal of the most reliable path is to maximize the reliability of the path, which is usually determined by the comprehensive feature score of the hardware on the path. Its objective function is: ; α4 is the reliable path target value, which represents the sum of the hardware comprehensive feature score values ​​and reflects the optimization degree of the most reliable path; Set the final optimization objective function of the network path for each network transmission task: ; Where λ1, λ2, λ3 and λ4 are the priority weights of the objective function, all of which are positive values ​​ranging from 0 to 1, and the sum of λ1, λ2, λ3 and λ4 is 1; Set basic constraints for the shortest path, lowest latency path, maximum bandwidth path, and most reliable path.

7. The automatic network connection tuning system based on the SRv6 protocol according to claim 6 is characterized in that: The basic constraints are as follows: No loop constraint: Each network device and network connection can only appear once on the path, that is, in the shortest path, for any two network connections (i1, i2) and (i3, i4), the norm of their connection score matrices is not equal, satisfying the condition ;in and are the norms of the connection score matrices; Network connection reachability constraint: d(i1, i2) has a value only when the distance condition between network devices i1 and i2 and the type of network devices meet the minimum requirements of network communication, that is, Where dmax is the preset maximum distance threshold, and Emin is the preset minimum category feature value threshold.

8. The automatic network connection tuning system based on the SRv6 protocol according to claim 1, characterized in that: The specific process of determining priority weights is: When the network transmission task contains regular types of data including text, code, and pictures, obtain the path length tolerance Dmax of data transmission and the maximum path distance target value maxα1 among all available network paths; otherwise, directly set the value of λ1 to 0.25; When the network transmission task contains data of the type of voice and video information, obtain the delay tolerance Tmax of data transmission and the maximum delay path target value maxα2 of all available network paths; otherwise, directly set the value of λ2 to 0.25 When the network transmission task contains large files and high-definition video streams whose data size is larger than the preset threshold, the bandwidth requirement tolerance Bmax of data transmission and the maximum bandwidth path target value maxα3 of all available network paths are obtained; otherwise, the value of λ3 is directly set to 0.25 When the network transmission task includes the network transmission task of encrypted data, the preset task reliability requirement Hmax and the maximum reliable path target value α4 of all available network paths are obtained; Otherwise, directly set the value of λ4 to 0.25 By preset formula Calculate the specific values ​​of the unassigned priority weights in λ1, λ2, λ3, and λ4.

9. The network connection automatic tuning system based on the SRv6 protocol according to claim 1, characterized in that: The specific process of scheduling different traffic flows is as follows: Retrieve specific values ​​of priority weights λ1, λ2, λ3, and λ4; By presetting the objective function Calculate the selection priority Pro of each network transmission task on each network path, and arrange each network path in descending order according to the numerical value of the selection priority; Convert the priority ranking of all network paths into an SRv6SID list; create an SRv6SID list for each specific network transmission task, where each row in the list corresponds to an available network path; the top of the list is the network path with the maximum selection priority maxPro, followed by the network path with the second largest selection priority, and so on, and the end of the SRv6SID list is the network path with the minimum selection priority minPro; Arrange network paths for all network transmission tasks according to the order of arrangement in the SRv6SID list, and select backup paths through the preset automatic switching mechanism; The automatic switching mechanism is specifically as follows: when the network path with the maximum selection priority maxPro is unavailable, the system shall select the path with the second largest selection priority as the new transmission path according to the list of other candidate paths provided by the path calculation module; the selection of candidate paths shall be dynamically adjusted according to the specific values ​​of the priority weights λ1, λ2, λ3 and λ4 of the traffic optimization module; the performance of the second best path shall be evaluated according to the specific values ​​of the priority weights λ1, λ2, λ3 and λ4 to ensure that the selection priority Pro of the path is greater than the preset threshold, otherwise the second path shall be determined to be unavailable; if the second best path is also unavailable, continue to try the third best path until an available path is found; When the automatic switching mechanism is triggered in the SRv6 SID list, a path switching event signal is sent to the visualization and alarm module.

10. A network connection automatic tuning device based on the SRv6 protocol, comprising an SRv6 server group and a target network, wherein the SRv6 server group comprises a number of servers, high-performance routers, switches, firewalls, gateways and wireless access points, characterized in that: The status monitoring module, path calculation module, traffic optimization module, network scheduling module and visualization and alarm module are operated together, and all servers are connected through a high-speed data bus; The data collected by the status monitoring module will be transmitted to the path calculation module in real time for path optimization calculation; The calculation results will be transmitted to the traffic optimization module to adjust the path selection weight, and finally handed over to the network scheduling module for priority scheduling; The visualization and alarm module obtains relevant data from each module and displays it on the monitoring panel.

Images