Management system and method for polling output of network traffic access port
By designing a management system for polling output of network traffic access ports, the problem of high complexity of network data traffic management in the existing technology is solved, and automated management and efficient resource utilization are realized.
Patent Information
- Application Number
- CN202411825519.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-12
- Publication Date
- 2025-03-28
AI Technical Summary
The prior art requires professional knowledge and complex operations in the process of converging and diversion of network data traffic, resulting in high complexity in use and difficulty in achieving automation and efficient management.
Design a management system for polling output of network traffic access ports, and realize automatic calculation of data paths and data forwarding during polling cycles through static configuration modules, configuration of polling parameter modules and execution polling modules.
It realizes automatic operation without a manned position and automatically creates data connection channels, simplifies network management complexity and optimizes network resource usage efficiency.
Smart Images

Figure CN119728582A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of network data flow, and in particular to a management system and method for polling output of a network flow access port. Background Art
[0002] In the application field of network data traffic aggregation and diversion, it is usually necessary to classify, filter, aggregate and divert various complex access data traffic before sending them to the back-end equipment for further processing. These application scenarios usually require specific configuration of various types of diverters, which require personnel with professional knowledge and familiarity with the equipment to handle, and have a certain degree of usage complexity. Summary of the invention
[0003] In order to help solve the above technical problems, the present application provides a management system and method for polling output of network traffic access ports.
[0004] In the first aspect, the present application proposes a management system for polling output of network traffic access ports, which adopts the following technical solutions:
[0005] A management system for polling output of network traffic access ports, wherein the management system is used for a traffic forwarding path, and the traffic forwarding path includes:
[0006] A first-level aggregation device, wherein the first-level aggregation device is used to receive access traffic;
[0007] A secondary flow distribution device, wherein the secondary flow distribution device is connected to the primary aggregation device via a cascade port,
[0008] A final processing device, the final processing device is connected to the secondary flow distribution device via an output port, and the output port is used to poll the output processing model;
[0009] The management system comprises:
[0010] A static configuration module, the static configuration module comprising a cascade port configuration submodule, an output port configuration submodule, a 100G mapping configuration submodule and a device RPC configuration submodule;
[0011] A polling parameter configuration module, wherein the polling parameter configuration module includes a polling configuration submodule, a fixed polling port configuration submodule, and a flexible polling port configuration submodule;
[0012] The polling execution module includes a polling start submodule, a polling stop submodule and a polling path query submodule.
[0013] In the second aspect, the present application proposes a management method for a management system of a network traffic access port polling output, which adopts the following technical solution:
[0014] A management method for a management system for polling output of a network traffic access port as described in the first aspect, wherein the management method comprises:
[0015] Step S1: Parse the configuration file and extract configuration information;
[0016] Step S2: The system initializes the log module, loads the default configuration, and synchronizes the existing configuration;
[0017] Step S3: The administrator configures the device information port, the 100G port information port and the cascade port, the user configures the input port and the output port, and selects a polling mode, the polling mode includes fixed polling and flexible polling. When the user selects fixed polling, the user specifies a fixed polling port; when the user selects flexible polling, the user configures a flexible polling port;
[0018] Step S4: Start the polling function and automatically calculate the data path according to the device information and port settings. During the polling cycle, the management system continuously determines whether it is still within a valid polling cycle. If it is within a valid cycle, the management system will call the interface according to the calculated path to forward data and prepare for the next polling cycle; if it is not within a valid cycle, the system will stop polling.
[0019] Preferably, the management method includes a mutually exclusive forwarding management method, and the mutually exclusive forwarding management method is used for mutually exclusive forwarding traffic forwarding paths.
[0020] The mutually exclusive forwarding traffic forwarding path includes two first-level aggregation devices equipped with 10G ports and one second-level flow distribution device, wherein the first-level aggregation device has four receiving ports and is connected to the second-level flow distribution device through two cascade ports, and finally outputs through three output ports;
[0021] The mutually exclusive forwarding management method comprises:
[0022] In each polling cycle, a device can forward at most two links at the same time. If there is a third link forwarding, it will cause cascade port multiplexing.
[0023] Preferably, the management method includes a multiple-input-one-output management method, and the multiple-input-one-output management method is used for a multiple-input-one-output traffic forwarding path.
[0024] The multi-input and one-output traffic forwarding path includes two first-level aggregation devices equipped with 10G ports and one second-level distribution device. The first-level aggregation device has four receiving ports and is connected to the second-level distribution device through two cascade ports, and finally outputs through one output port;
[0025] The multiple-input-one-output management method comprises:
[0026] Cancel the previous round of copy path configuration, select an input port from the first-level aggregation device, and copy the input port to the cascade port of the first-level aggregation device, and copy the output from the cascade port of the second-level diversion device to the final output port.
[0027] Preferably, the management method includes a multi-input multi-output management method, and the multi-input multi-output management method is used for a multi-input multi-output traffic forwarding path.
[0028] The traffic forwarding path includes a multi-input and multi-output traffic forwarding path, which includes two first-level aggregation devices using 10G ports and one second-level flow distribution device, wherein the first-level aggregation device has four receiving ports and is connected to the second-level flow distribution device through two cascade ports, and finally outputs through two output ports;
[0029] The multiple-input multiple-output management method comprises:
[0030] The same device outputs two flows each time.
[0031] Preferably, the management method includes a 100G inbound splitting and outbound management method, and the 100G inbound splitting and outbound management method is used for a 100G inbound splitting and outbound traffic forwarding path.
[0032] The traffic forwarding path includes a 100G inbound, split and outbound traffic forwarding path, which is composed of a first-level aggregation device and a second-level flow distribution device. The first-level aggregation device accesses traffic through a 100G port, and the 100G inbound, split and outbound traffic forwarding path is also provided with a 10G cascade port and a 10G output port.
[0033] The 100G in / out management method includes:
[0034] In the polling cycle, the polling cycle is divided into ten parts, and one port is polled each time, so that the traffic of the 100G port is evenly distributed to the ten 10G ports.
[0035] Preferably, in step S4, the data path is automatically calculated by a fixed port polling path algorithm, and the fixed port polling path algorithm includes:
[0036] If it is currently in the polling state, stop polling and check whether it is still in the polling cycle. If not, clean up the resources;
[0037] Get all output ports: Collect and record all output ports in the network, and get all fixed forwarding strategies and their corresponding input and output ports;
[0038] Determine whether the output port of the current fixed forwarding path is among the recorded output ports. If there is a conflict, adjust the configuration to avoid the conflict.
[0039] All fixed forwarding configurations are grouped according to the output ports, and within each group, they are sorted according to the sequence number of the output ports;
[0040] Starting from the first forwarding strategy, the forwarding paths are combined one by one. Each output port is combined with a link and a cascade port to determine whether the current link will cause a cascade port conflict. If there is a conflict, the next link under the same output group is selected for configuration.
[0041] When all forwarding paths are configured and there are no conflicts, the configuration phase ends.
[0042] Preferably, the fixed port polling path algorithm further includes:
[0043] Determine whether the input port is a 100G port. If it is a 100G port, split a complete policy cycle into 10 small cycles, and use a 10G table position for polling in each small cycle.
[0044] Preferably, in step S4, the data path is automatically calculated by a flexible port polling path algorithm, and the flexible port polling path algorithm includes:
[0045] Determine whether it is still within the polling cycle. If not, stop polling and clean up resources. If it is within the polling cycle, obtain all output ports and fixed forwarding input and output ports;
[0046] Determine whether the output port of the polling forwarding path is included in the obtained output port. If included, sort all the polling forwarding entries, group and sort them according to the input port, and further sort them according to the number of input ports;
[0047] Combining forwarding links from the first cycle, determining the number of cascade ports required according to the number of output ports, allocating input and output ports according to the number of cascade ports, and allocating corresponding cascade ports and output ports to each input and output port.
[0048] Preferably, the flexible port polling path algorithm further includes:
[0049] Start polling and determine whether the input port is a 100G port. If it is a 100G port, split one cycle into 10 small cycles, and use the 10G port for polling in each small cycle.
[0050] In summary, the present application can achieve one-time configuration without maintenance, unattended automatic operation, automatic creation of data paths for data communication, automatic switching according to configuration, and regular polling to switch data traffic paths, which greatly simplifies the complexity of network management and optimizes the efficiency of network resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] Figure 1 It is a schematic block diagram of a manually operated port polling operation in the prior art;
[0052] Figure 2 A schematic block diagram of an embodiment of a traffic forwarding path of the present application;
[0053] Figure 3 A flowchart of an embodiment of a management method for a management system of a network traffic access port polling output of the present application;
[0054] Figure 4 A schematic diagram of the principle of the management method of the present application for mutually exclusive forwarding of traffic forwarding paths;
[0055] Figure 5 This is a schematic diagram of the principle of the management method of the present application for device 1 in a multi-input and one-output traffic forwarding path;
[0056] Figure 6 This is a schematic diagram of the principle of the management method of the present application for device 2 in a multi-input and one-output traffic forwarding path;
[0057] Figure 7 It is a schematic diagram of the principle of the same device output model when the management method of the present application is used for a multi-input and multi-output traffic forwarding path;
[0058] Figure 8 It is a schematic diagram of the principle of a multi-device average output model when the management method of the present application is used for a multi-input and multi-output traffic forwarding path;
[0059] Fig. 9 This is a schematic diagram of the principle of the management method of the present application for 100G inbound, outbound and outbound traffic forwarding paths;
[0060] Fig.10 A flowchart of an embodiment of a fixed port polling path algorithm of the present application;
[0061] Fig.11 The figure is a flowchart of an embodiment of the flexible port polling path algorithm of the present application. DETAILED DESCRIPTION
[0062] The present application is further described below in conjunction with the accompanying drawings, and the structure and principle of the present application are very clear to people in the field. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0063] Figure 1 FIG. 1 is a schematic block diagram of a manual port polling operation in the prior art. Figure 1 As can be seen in the figure, the manual operation of port polling operation involves multiple devices and a large number of ports. Manual operation is tedious and complicated. In order to improve efficiency, the configuration and management of automated splitters are required to ensure accurate replication and real-time analysis of traffic. In the logic of manual operation, each round of port polling operation requires users to log in to different devices regularly to cancel old link configurations and configure new links. When the number of links is large, users must confirm the polling order, and when the number of devices is large, they must ensure that they are logged in to the corresponding device to avoid misoperation. If the cancellation and configuration of links are distributed on different devices, users need to log in multiple times, and manual operation is very tedious.
[0064] Figure 2 A schematic block diagram of an embodiment of a traffic forwarding path of the present application.
[0065] After the existing equipment is connected to the system, in order to ensure the stability and security of the network, it is impossible to cut off the network to detect the quality or content information of the corresponding port access traffic separately. It is necessary to copy and forward the access traffic through a technical means, and copy and forward the access traffic to the corresponding analysis port by formulating a copy forwarding strategy. Usually there is only one analysis port, on which the back-end device will implement quality analysis, content filtering or fault detection of the access traffic.
[0066] This application defines the input device of the access traffic as a first-level aggregation device, which is responsible for receiving all system traffic. Since the sources of the existing network traffic access links are diverse and the expected traffic is large, the number of devices at this level is set to N1, so multiple first-level aggregation devices will be configured. The data traffic aggregated by the first-level aggregation device will be transmitted to the second-level diversion device, which is referred to as the second-level diversion device in this application, and is responsible for distributing the first-level aggregated traffic to other devices. The second-level diversion device receives the traffic filtered and processed by the first-level aggregation device and performs a certain degree of convergence, so its number (N2) will be less than the first-level aggregation device.
[0067] The primary and secondary distribution devices are connected through data ports (called cascade ports) to forward data between devices. The connection port between the secondary distribution device and the final processing device is called the output port. The number of final processing devices is usually one, two or four. The specific configuration depends on the existing network traffic and customer needs, and is set to N3.
[0068] In this embodiment, the typical port polling device model is N1:N2:N3=2:1:1, that is, two first-level aggregation devices, one second-level diversion device and one final processing device, and the port polling output processing model. The implementation details of the technical solution described in this application are all explained based on this typical model. In other optional embodiments, N1, N2 and N3 can be different proportions.
[0069] The present application proposes a management system for polling output of network traffic access ports, the management system is used for traffic forwarding paths, and the traffic forwarding paths include a primary aggregation device, a secondary diversion device, and a final processing device. The primary aggregation device is used to receive access traffic, the secondary diversion device is connected to the primary aggregation device through a cascade port, and the final processing device is connected to the secondary diversion device through an output port, and the output port is used for polling output processing models.
[0070] The structure of the management system is as follows:
[0071]
[0072]
[0073] The management system may include a static configuration module, a polling parameter configuration module and a polling execution module. The static configuration module includes a cascade port configuration submodule, an output port configuration submodule, a 100G mapping configuration submodule and a device RPC configuration submodule; the polling parameter configuration module includes a polling configuration submodule, a fixed polling port configuration submodule and a flexible polling port configuration submodule; the polling execution module includes a polling start submodule, a polling stop submodule and a polling channel query submodule.
[0074] Figure 3 A flow chart of an embodiment of a management method for a management system of a network traffic access port polling output of the present application. During the construction of the management system, the device information, 100G port information and the wiring of the cascade port are predetermined by the administrator. The user only needs to configure the input port, output port and polling information (such as the start time and the time of each polling) according to the needs. After the software is started, the configuration file is first parsed to extract the configuration information, the system initializes the log module, records the information generated during the operation, and loads the default configuration to synchronize the existing settings to ensure the consistency of the management system configuration.
[0075] Users need to configure input ports and output ports according to their needs. Then select the polling mode (divided into fixed polling and flexible polling). When choosing fixed polling, users need to specify the fixed polling port; if choosing flexible polling, users need to configure the corresponding flexible polling port.
[0076] After all configurations are completed, the system will start the polling function and automatically calculate the data path based on the device information and port settings. During the polling cycle, the system will continue to determine whether it is still in a valid polling cycle. If it is in a valid cycle, the system will call the interface to forward data based on the calculated path and prepare for the next polling cycle; if it is not in a valid cycle, the system will stop polling and clean up related resources.
[0077] Through this series of operations, the system ensures automatic calculation of data paths and avoids data link conflicts, thereby achieving efficient data transmission and management. Calculating data paths and ensuring their mutual exclusion is the core function of the system.
[0078] In summary, management methods may include:
[0079] Step S1: Parse the configuration file and extract configuration information;
[0080] Step S2: The system initializes the log module, loads the default configuration, and synchronizes the existing configuration;
[0081] Step S3: The administrator configures the device information port, 100G port information port and cascade port, the user configures the input port and output port, and selects the polling mode. The polling mode includes fixed polling and flexible polling. When the user selects fixed polling, the user specifies the fixed polling port; when the user selects flexible polling, the user configures the flexible polling port;
[0082] Step S4: Start the polling function and automatically calculate the data path according to the device information and port settings. During the polling cycle, the management system continuously determines whether it is still within a valid polling cycle. If it is within a valid cycle, the management system will call the interface according to the calculated path to forward data and prepare for the next polling cycle; if it is not within a valid cycle, the system will stop polling.
[0083] To help understand the technical solution of the present application, traffic forwarding paths may include mutually exclusive forwarding traffic forwarding paths, multi-input and one-output traffic forwarding paths, multi-input and multi-output traffic forwarding paths, and 100G in-split and out-put traffic forwarding paths. For the aforementioned traffic forwarding paths, the present application proposes a mutually exclusive forwarding management method, a multi-input and one-output management method, a multi-input and multi-output management method, and a 100G in-split and out-put management method.
[0084] Figure 4 The management method of the present application is used for the principle diagram of mutually exclusive forwarding traffic forwarding path. The management method includes a mutually exclusive forwarding management method, which is used for mutually exclusive forwarding traffic forwarding path.
[0085] The mutually exclusive forwarding traffic forwarding path includes two first-level aggregation devices equipped with 10G ports and one second-level distribution device. The first-level aggregation device has four receiving ports and is connected to the second-level distribution device through two cascade ports, and finally outputs through three output ports; the mutually exclusive forwarding management method includes: in each polling cycle, the same device has a maximum of two links forwarding at the same time. If there is a third link forwarding, it will cause the cascade port to be reused.
[0086] Specifically, there are two first-level aggregation devices and one second-level distribution device, all of which are equipped with 10G ports. The first-level aggregation device has four receiving ports, and is connected to the second-level distribution device through two cascade ports, with three final output ports. Because there are three final output ports and two cascade ports, the same device can forward at most two links at the same time in each polling cycle. If there is a third link, it will cause the cascade port to be reused, which will cause data link abnormalities. Figure 4 In the port mutual exclusion scheme diagram, link 3 and link 2 in device 1 use the same cascade port, which is an error link. The mutual exclusion of the ports can be ensured by using link 1 in device 2 and links 1 and 2 in device 1.
[0087] Figure 5 This is a schematic diagram of the principle of the management method of this application for device 1 in a multi-input and one-output traffic forwarding path. Figure 6 This is a schematic diagram showing the principle of the management method of the present application being used for device 2 in a multi-input and one-output traffic forwarding path.
[0088] The management method includes a multi-input and one-output management method, which is used for a multi-input and one-output traffic forwarding path. The multi-input and one-output traffic forwarding path includes two first-level aggregation devices and one second-level diversion device equipped with 10G ports. The first-level aggregation device has four receiving ports and is connected to the second-level diversion device through two cascade ports, and finally outputs through one output port; the multi-input and one-output management method includes: canceling the previous round of copy path configuration, selecting an input port from the first-level aggregation device, and copying the input port to the cascade port of the first-level aggregation device, and copying the output from the cascade port of the second-level diversion device to the final output port.
[0089] Specifically, there are two first-level aggregation devices and one second-level distribution device, all of which are equipped with 10G ports. The first-level aggregation device has four receiving ports and is connected to the second-level distribution device through two cascade ports. The final output is only through one port. Since there is only one final output port, each round of port polling can only be output by one port.
[0090] The following is an example of the specific operation of the periodic channel of port polling, which has eight rounds in total. After the polling is completed, it will restart from the first round.
[0091] The first round of port polling: select port F / 1 from the first-level aggregation device (device 1), copy it and output it to the cascade port B / 1 of device 1, and then copy it from the cascade port F / 2 of device 3 to the final output port B / 6.
[0092] Second round of port polling: cancel the first round of copy path configuration, select port F / 2 from the first level aggregation device (device 1), copy it and output it to the cascade port B / 1 of device 1, and then copy it from the cascade port F / 2 of device 3 to the final output port B / 6.
[0093] The third round of port polling: cancel the second round of copy path configuration, select port F / 3 from the first level aggregation device (device 1), copy it and output it to the cascade port B / 1 of device 1, and then copy it from the cascade port F / 2 of device 3 to the final output port B / 6.
[0094] Fourth round of port polling: cancel the third round of copy path configuration, select port F / 4 from the first level aggregation device (device 1), copy it and output it to the cascade port B / 1 of device 1, and then copy it from the cascade port F / 2 of device 3 to the final output port B / 6.
[0095] The fifth round of port polling: cancel the fourth round of copy path configuration, select port F / 1 from the second first-level aggregation device (device 2), copy it and output it to the cascade port B / 1 of device 2, and then copy it from the cascade port F / 6 of device 3 to the final output port B / 6.
[0096] The sixth round of port polling: cancel the fifth round of copy path configuration, select port F / 2 from the second first-level aggregation device (device 2), copy it and output it to the cascade port B / 1 of device 2, and then copy it from the cascade port F / 6 of device 3 to the final output port B / 6.
[0097] Seventh round of port polling: Cancel the sixth round of copy path configuration, select port F / 3 from the second first-level aggregation device (device 2), copy it and output it to the cascade port B / 1 of device 2, and then copy it from the cascade port F / 6 of device 3 to the final output port B / 6.
[0098] The eighth round of port polling: cancel the seventh round of copy path configuration, select port F / 4 from the second first-level aggregation device (device 2), copy it and output it to the cascade port B / 1 of device 2, and then copy it from the cascade port F / 6 of device 3 to the final output port B / 6.
[0099] Figure 7 This is a schematic diagram of the principle of the same device output model when the management method of this application is used for a multi-input and multi-output traffic forwarding path. Figure 8The schematic diagram of the principle of the multi-device average output model when the management method of the present application is used for a multi-input and multi-output traffic forwarding path. The management method includes a multi-input and multi-output management method, and the multi-input and multi-output management method is used for a multi-input and multi-output traffic forwarding path.
[0100] The traffic forwarding path includes a multi-input and multi-output traffic forwarding path, which includes two first-level aggregation devices and one second-level diversion device using 10G ports. The first-level aggregation device has four receiving ports and is connected to the second-level diversion device through two cascade ports, and finally outputs through two output ports; the multi-input and multi-output management method includes: the same device outputs two copies of traffic each time.
[0101] Specifically, there are two first-level aggregation devices and one second-level distribution device, all of which use 10G ports. The first-level aggregation device has four receiving ports and is connected to the second-level distribution device through two cascade ports, and finally has two output ports. In the current model, the cascade port and the output port are both two ports, which allows two copies of traffic to be output each time, that is, the same device can output two copies at the same time. The same device can output the model as much as possible, or two devices can output one copy each, and different devices can output the model on average.
[0102] Fig. 9 This is a schematic diagram showing the principle of the management method of the present application for 100G inbound, outbound and outbound traffic forwarding paths.
[0103] The management method includes a 100G in / out split management method, which is used for a 100G in / out split traffic forwarding path, and the traffic forwarding path includes a 100G in / out split traffic forwarding path, and the 100G in / out split traffic forwarding path is composed of a first-level aggregation device and a second-level diversion device, and the first-level aggregation device accesses the traffic through a 100G port, and the 100G in / out split traffic forwarding path is also provided with a 10G cascade port and a 10G output port, and the 100G in / out split management method includes: in a polling cycle, dividing the polling cycle into ten parts, polling one port each time, and evenly distributing the traffic of the 100G port to ten 10G ports.
[0104] Specifically, there is a first-level aggregation device and a second-level distribution device, where ports F / 1 and F / 9 of the first-level and second-level distribution devices are 100G ports, and the rest of the ports are 10G ports. The first-level aggregation device accesses traffic through the 100G port, and has a 10G cascade port and a 10G output port. Since the rate of the input port is much different from the rate of the output port, it cannot be forwarded directly. If the output is directly copied, a large number of packets will be lost on the first-level 10G output port. Therefore, we configured ten 10G load balancing ports on the first-level aggregation device to achieve load balancing of the 100G port inlet traffic.
[0105] In the polling cycle, the cycle is divided into ten parts, and one port is polled each time, so that the traffic of the 100G port is evenly distributed to the ten 10G ports. This ensures that only one part of the traffic is processed at a time, and the 100G traffic is accessed and parsed as efficiently as possible.
[0106] According to the above data path model, the software will automatically calculate all available data paths when configuring input ports, cascade ports and output ports. In step S4, the data paths are automatically calculated by the fixed port polling path algorithm and the flexible port polling path algorithm.
[0107] Fig.10 This is a flow chart of an embodiment of the fixed port polling path algorithm of the present application. The fixed port polling path algorithm may include the following steps:
[0108] If it is currently in the polling state, stop polling and check whether it is still in the polling cycle. If not, clean up the resources;
[0109] Get all output ports: Collect and record all output ports in the network, and get all fixed forwarding strategies and their corresponding input and output ports;
[0110] Determine whether the output port of the current fixed forwarding path is among the recorded output ports. If there is a conflict, adjust the configuration to avoid the conflict.
[0111] All fixed forwarding configurations are grouped according to the output ports, and within each group, they are sorted according to the sequence number of the output ports;
[0112] Starting from the first forwarding strategy, the forwarding paths are combined one by one. Each output port is combined with a link and a cascade port to determine whether the current link will cause a cascade port conflict. If there is a conflict, the next link under the same output group is selected for configuration.
[0113] When all forwarding paths are configured and there is no conflict, the configuration phase ends;
[0114] Determine whether the input port is a 100G port. If so, split a complete policy cycle into 10 small cycles, and use a 10G table position for polling in each small cycle.
[0115] Fixed port polling configuration is configured by the user specifying the input port of the first-level aggregation device and the output port of the second-level distribution device. The input and output are associated and bound, and the order from input to output must be followed. The software needs to find the cascade port that can connect these two ports based on the specified input and output ports. First, it is necessary to ensure that the output port is within the range specified by the system, and then maximize the use of the output port resources to ensure that the resources of the output port are fully utilized in each polling cycle. During the port link pairing process, it must be ensured that the ports of each cycle are mutually exclusive and cannot be called multiple times in the same cycle.
[0116] Fig.11 The figure is a flowchart of an embodiment of the flexible port polling path algorithm of the present application.
[0117] The flexible port round-robin algorithm may include the following steps:
[0118] Determine whether it is still within the polling cycle. If not, stop polling and clean up resources. If it is within the polling cycle, obtain all output ports and fixed forwarding input and output ports;
[0119] Determine whether the output port of the polling forwarding path is included in the obtained output port. If included, sort all the polling forwarding entries, group and sort them according to the input port, and further sort them according to the number of input ports;
[0120] Combining forwarding links from the first cycle, determining the number of cascade ports required according to the number of output ports, allocating input and output ports according to the number of cascade ports, and allocating corresponding cascade ports and output ports to each input and output port;
[0121] Start polling and determine whether the input port is a 100G port. If it is a 100G port, split one cycle into 10 small cycles, and use the 10G port for polling in each small cycle.
[0122] Flexible port polling configuration allows users to specify the input port of the first-level aggregation device and the output port of the second-level distribution device. There is no fixed binding relationship between the inlet and the outlet, and they can be freely combined. The software will use the cascaded ports as much as possible to achieve the best match between the inlet and the outlet. During the port link pairing process, it must be ensured that the ports of each cycle cannot be called repeatedly in the same cycle (mutual exclusivity).
[0123] In both link calculation algorithms, the software ensures that each cycle can maximize the use of cascade port resources while ensuring mutual exclusion of port resources to avoid conflicts. By automatically configuring the port data forwarding path in the link, the management software can complete the configuration of the port data path within the polling cycle with extremely low operation latency.
[0124] This software supports two management methods: Web management and command line management. As the basic operation method, command line management provides all the functions of the software and is suitable for professional users. Although the command line configuration has high operation efficiency, its operation is relatively complex and not intuitive enough for novice users. To facilitate the use of novice users, the system has designed a more intuitive Web management method, where users can view information and perform operation configuration through a visual web page interface. The two management methods complement each other and can manage port polling more effectively.
[0125] With the software designed by this solution, users only need to configure the corresponding polling forwarding port solution according to the specific application scenario. The system can automatically calculate the port link and automatically perform the port polling operation according to the specified polling time and date range. Users can also view the current polling link information and polling records of other cycles at any time. In addition, the operation record log function enables users to check the index information of historical operations. The system automatically handles the polling port, which simplifies manual operation and reduces the risk of errors. Intelligent link information creation and allocation allows users to focus on business without worrying about the details of internal links.
[0126] By implementing automated polling operations and flexible interval execution mechanisms, the complexity of network management is greatly simplified. The management system can autonomously perform multiple tasks, reducing the need for manual intervention and improving work efficiency. The system can intelligently identify and optimize network paths to ensure smooth and efficient data transmission. The introduction of dual management methods allows users to choose the most suitable operation method according to their own needs. This flexibility not only improves the availability of the system, but also greatly improves the user experience and meets the needs of different user groups. Combined with low-latency data forwarding technology, the overall solution optimizes the efficiency of network resource use while meeting high bandwidth requirements. The overall solution is not only technologically advanced, but also has high practical value. It can optimize resource use to reduce costs while meeting actual application requirements and improve the efficiency and effectiveness of overall network management. The interval execution mechanism simplifies network management. The interval execution mechanism is for network management. The overall solution optimizes the efficiency of network resource use while meeting high bandwidth requirements. The overall solution is not only technologically advanced, but also has high practical value.
[0127] In summary, the advantages of the present invention are:
[0128] Simplified operation process: By realizing automatic timing switching of port links, the present invention significantly reduces the complexity of manual operation and improves the efficiency and stability of system operation.
[0129] Dual management mode: It provides two management modes: Web interface control and command line control. It has a beautiful interface and excellent visualization effect, supports refined management, and improves user experience.
[0130] Automatic control function: The system supports automatic control start and stop functions, which can automatically meet specific needs according to user configuration, reducing dependence on manual intervention.
[0131] Dynamic data path generation: The present invention can automatically associate input and output ports and generate a complete data path based on the configured static first and second level forwarding relationships. This function has a port and link error and anomaly detection mechanism to ensure the uniqueness of the path and avoid duplication or conflict.
[0132] The management system and method described in this application have been improved for the current scheme (manual operation port polling operation), and a system has been designed and developed to realize the function of copying the traffic mirror in the existing network and outputting it to the back end. The management system is simple to operate and has clear configuration. It can dynamically calculate the mirror forwarding and path of the traffic to ensure the consistency and efficiency of the forwarding path. It has flexible polling settings, and users can customize the start time, duration and pause time of the polling, and support fixed mode and flexible mode at the same time to meet the diverse traffic monitoring needs. By automatically switching the port link, the user's cumbersome manual operation is effectively reduced, and low-latency data link switching is achieved through the corresponding interface, reducing the packet loss rate in the data path switching.
Claims
1. A management system for polling output of network traffic access ports, characterized in that: The management system is used for a traffic forwarding path, and the traffic forwarding path includes: A first-level aggregation device, the first-level aggregation device is used to receive access traffic, A secondary flow distribution device, wherein the secondary flow distribution device is connected to the primary aggregation device via a cascade port, A final processing device, the final processing device is connected to the secondary flow distribution device via an output port, and the output port is used to poll the output processing model; The management system comprises: A static configuration module, the static configuration module comprising a cascade port configuration submodule, an output port configuration submodule, a 100G mapping configuration submodule and a device RPC configuration submodule; A polling parameter configuration module, wherein the polling parameter configuration module includes a polling configuration submodule, a fixed polling port configuration submodule, and a flexible polling port configuration submodule; The polling execution module includes a polling start submodule, a polling stop submodule and a polling path query submodule.
2. A management method for a management system for network traffic access port polling output as claimed in claim 1, characterized in that: The management method comprises: Step S1: Parse the configuration file and extract configuration information; Step S2: The system initializes the log module, loads the default configuration, and synchronizes the existing configuration; Step S3: The administrator configures the device information port, the 100G port information port and the cascade port, the user configures the input port and the output port, and selects a polling mode, the polling mode includes fixed polling and flexible polling. When the user selects fixed polling, the user specifies a fixed polling port; when the user selects flexible polling, the user configures a flexible polling port; Step S4: Start the polling function and automatically calculate the data path according to the device information and port settings. During the polling cycle, the management system continuously determines whether it is still within a valid polling cycle. If it is within a valid cycle, the management system will call the interface according to the calculated path to forward data and prepare for the next polling cycle; if it is not within a valid cycle, the system will stop polling.
3. The management method according to claim 2, characterized in that: The management method includes a mutually exclusive forwarding management method, wherein the mutually exclusive forwarding management method is used for mutually exclusive forwarding traffic forwarding paths. The mutually exclusive forwarding traffic forwarding path includes two first-level aggregation devices equipped with 10G ports and one second-level flow distribution device, wherein the first-level aggregation device has four receiving ports and is connected to the second-level flow distribution device through two cascade ports, and finally outputs through three output ports; The mutually exclusive forwarding management method comprises: In each polling cycle, a device can forward at most two links at the same time. If there is a third link forwarding, it will cause cascade port multiplexing.
4. The management method according to claim 2, characterized in that: The management method includes a multiple-input and one-output management method, and the multiple-input and one-output management method is used for a multiple-input and one-output traffic forwarding path. The multi-input and one-output traffic forwarding path includes two first-level aggregation devices equipped with 10G ports and one second-level distribution device. The first-level aggregation device has four receiving ports and is connected to the second-level distribution device through two cascade ports, and finally outputs through one output port; The multiple-input-one-output management method comprises: Cancel the previous round of copy path configuration, select an input port from the first-level aggregation device, and copy the input port to the cascade port of the first-level aggregation device, and copy the output from the cascade port of the second-level diversion device to the final output port.
5. The management method according to claim 2, characterized in that: The management method includes a multi-input multi-output management method, and the multi-input multi-output management method is used for a multi-input multi-output traffic forwarding path. The traffic forwarding path includes a multi-input and multi-output traffic forwarding path, which includes two first-level aggregation devices using 10G ports and one second-level flow distribution device, wherein the first-level aggregation device has four receiving ports and is connected to the second-level flow distribution device through two cascade ports, and finally outputs through two output ports; The multiple-input multiple-output management method comprises: The same device outputs two flows each time.
6. The management method according to claim 2, characterized in that: The management method includes a 100G inbound splitting and outbound management method, and the 100G inbound splitting and outbound management method is used for a 100G inbound splitting and outbound traffic forwarding path. The traffic forwarding path includes a 100G inbound, split and outbound traffic forwarding path, which is composed of a first-level aggregation device and a second-level flow distribution device. The first-level aggregation device accesses traffic through a 100G port, and the 100G inbound, split and outbound traffic forwarding path is also provided with a 10G cascade port and a 10G output port. The 100G in / out management method includes: In the polling cycle, the polling cycle is divided into ten parts, and one port is polled each time, so that the traffic of the 100G port is evenly distributed to the ten 10G ports.
7. The management method according to any one of claims 3 to 6, characterized in that: In step S4, the data path is automatically calculated by a fixed port round-robin path algorithm, and the fixed port round-robin path algorithm includes: If it is currently in the polling state, stop polling and check whether it is still in the polling cycle. If not, clean up the resources; Get all output ports: Collect and record all output ports in the network, and get all fixed forwarding strategies and their corresponding input and output ports; Determine whether the output port of the current fixed forwarding path is among the recorded output ports. If there is a conflict, adjust the configuration to avoid the conflict. All fixed forwarding configurations are grouped according to the output ports, and within each group, they are sorted according to the sequence number of the output ports; Starting from the first forwarding strategy, the forwarding paths are combined one by one. Each output port is combined with a link and a cascade port to determine whether the current link will cause a cascade port conflict. If there is a conflict, the next link under the same output group is selected for configuration. When all forwarding paths are configured and there are no conflicts, the configuration phase ends.
8. The management method according to claim 7, characterized in that: The fixed port polling path algorithm also includes: Determine whether the input port is a 100G port. If it is a 100G port, split a complete policy cycle into 10 small cycles, and use a 10G table position for polling in each small cycle.
9. The management method according to any one of claims 3 to 6, characterized in that: In step S4, the data path is automatically calculated by a flexible port polling path algorithm, and the flexible port polling path algorithm includes: Determine whether it is still within the polling cycle. If not, stop polling and clean up resources. If it is within the polling cycle, obtain all output ports and fixed forwarding input and output ports; Determine whether the output port of the polling forwarding path is included in the obtained output port. If included, sort all the polling forwarding entries, group and sort them according to the input port, and further sort them according to the number of input ports; Combining forwarding links from the first cycle, determining the number of cascade ports required according to the number of output ports, allocating input and output ports according to the number of cascade ports, and allocating corresponding cascade ports and output ports to each input and output port.
10. The management method according to claim 9, characterized in that: The flexible port polling path algorithm also includes: Start polling and determine whether the input port is a 100G port. If it is a 100G port, split one cycle into 10 small cycles, and use the 10G port for polling in each small cycle.