Industrial internet traffic scheduling method and system

Abstract

The invention provides an industrial internet traffic scheduling method and system. The industrial internet traffic scheduling method comprises the following steps: classifying industrial internet traffic; then, INT stream detection is deployed on the classified industrial internet traffic to collect time delay and packet loss probability information of the industrial internet traffic; the core lies in traffic scheduling, and an optimal path bearing user traffic is calculated through a path constraint algorithm; and finally, performing path deployment, and constructing an end-to-end forwarding path based on the calculated optimal path. The system can be compatible with networks such as MPLS, SR and SRv6, and fine-grained real-time data plane monitoring is realized; and a multi-constraint parallel traffic algorithm is adopted to dynamically respond to network resource requirements, so that the network packet loss rate, the time delay, the traffic conflict and the energy consumption are effectively reduced, the resource utilization rate is improved, the network quality is guaranteed, and diversified requirements are met.

Description

Technical Field

[0001] The present invention belongs to the technical field of industrial Internet traffic scheduling, and specifically relates to an industrial Internet traffic scheduling method and system. Background Art

[0002] With the rapid development of digital technologies, the industrial environment is undergoing a comprehensive and profound digital transformation. The Industrial Internet plays a vital role in this transformation, becoming the core force driving this change. The rapid development of the Industrial Internet has profoundly transformed and upgraded the production methods and operating models of traditional industries, injecting new vitality into the production environment and achieving more efficient, intelligent, safe, reliable, and sustainable development.

[0003] As the Industrial Internet advances, challenges in real-time performance, flexibility, reliability, and security become increasingly prominent. Data traffic in intelligent industrial networks originates from a wide variety of sources, and service demands are also becoming increasingly diverse. This undoubtedly places unprecedented pressure on network resources. Furthermore, the processing quality of these massive industrial data streams varies widely, making it difficult to accurately assess their potential impact on network architectures. Network flexibility and reliability are the cornerstones of stable industrial operations, making the Industrial Internet's demands for high flexibility and reliability particularly pressing. However, existing network architectures are unable to meet these challenges, resulting in inadequate network service quality assurance. Summary of the Invention

[0004] In response to the deficiencies in the existing technology, the present invention provides an industrial Internet traffic scheduling method and system to solve the current deficiencies in the industrial Internet in terms of flexibility, latency, and packet loss rate.

[0005] The present invention achieves the above technical objectives through the following technical means.

[0006] An industrial Internet traffic scheduling method includes the following steps:

[0007] Traffic classification, classifying industrial Internet traffic;

[0008] INT on-stream detection: deploy INT on-stream detection on classified industrial Internet traffic to collect information on latency and packet loss rate of industrial Internet traffic;

[0009] Traffic scheduling, using a path constraint algorithm to calculate the optimal path to carry user traffic;

[0010] Path deployment, builds an end-to-end forwarding path based on the calculated optimal path.

[0011] Furthermore, the path constraint algorithm includes a delay priority algorithm and a packet loss rate priority algorithm:

[0012] Step 1: Use Dijkstra algorithm to find the shortest path P with the lowest delay / packet loss cost. min ;

[0013] Step 2: If P min If all the indicators are met, P min As the optimal path; if the calculated path delay / packet loss rate does not meet the index requirements, the path with the minimum current delay / packet loss rate is selected as the optimal path.

[0014] Furthermore, the path constraint algorithm includes a multi-constraint parallel routing algorithm:

[0015] First, use the Dijkstra algorithm to find the shortest routing path P with delay as the path cost t ;Then check whether this path meets the requirements based on the delay and packet loss requirements of the flow and the availability of limited resources;

[0016] If all conditions are met, the optimal path is obtained;

[0017] If the requirements are not met, the Dijkstra algorithm is used to calculate the shortest path P in the network with the packet loss rate as the path cost. d , and check whether the packet loss requirement is met. If it is met, the Lagrange dual function is used to calculate the path aggregation cost, and the Dijkstra algorithm is used to find the shortest path r based on this aggregation cost;

[0018] Finally, path r is replaced by P according to whether its delay and packet loss rate meet the traffic requirements. t or P d , repeat the iteration until convergence to obtain the optimal path.

[0019] Furthermore, the multi-constraint parallel routing algorithm includes the following steps:

[0020] Step 1: Use Dijkstra algorithm to calculate the shortest path P with the minimum delay cost. t ;

[0021] Step 2,

[0022] Path P t If all conditions are met, the path P t Output as the optimal path result;

[0023] Path P t If the delay meets the requirement but the packet loss rate does not meet the requirement, execute step 3;

[0024] Path P tIf the delay and packet loss rate meet the requirements, but the resources are insufficient, return to step 1 to recalculate a shortest path P with the minimum delay cost. t ;

[0025] Step 3: Use Dijkstra algorithm to calculate the shortest path P with the lowest packet loss cost. d ;

[0026] Step 4, Path P d If the packet loss rate meets the requirement, go to step 5;

[0027] Step 5: Use the Lagrange dual function to calculate the path, aggregate the constraints into the objective function, and define the Lagrange multiplier as λ. λ (e) is defined as the sum of the delay and weighted packet loss of link e, and uses the link aggregation cost G λ Calculate the shortest path r using Dijkstra's algorithm:

[0028]

[0029] λ←[T f (P t )-T f (P d )] / [D f (P t )-D f (P d )]

[0030] r←Dijkstra(s,t,G λ )

[0031] Where T(e) is the delay parameter of link e, D(e) is the packet loss rate parameter of link e, T(e) and D(e) are obtained by INT flow detection, T f (P t ) is the path P t The delay parameter, T f (P d ) is the path P d The delay parameter, D f (P t ) is the path P t The packet loss rate parameter, D f (P d ) is the path P d The packet loss rate parameter, T f (P t 、T f (P d ), D f (P t ), D f (Pd ) is calculated by Dijkstra algorithm, r is the cost from source node s to target node t based on the aggregation cost G λ Calculate the shortest path;

[0032] When G is satisfied λ (r) = G(P t )=G λ (P d ), then execute step 6;

[0033] When G is not satisfied λ (r) = G(P t )=G λ (P d ), then determine whether the path r meets the requirements of both delay and packet loss rate. If the path r meets the requirements of delay and packet loss rate, the path r is set as the new P d On the contrary, if the path r does not meet the requirements of both delay and packet loss rate, the path r is set as the new P t ; Set path r to the new P d or P t Then, repeat step 5 until the solution that satisfies G is calculated. λ (r) = G(P t )=G λ (P d ) path r;

[0034] Step 6: Check whether the path r meets the resource requirements, where:

[0035] When resources are sufficient, path r is output as the optimal path result;

[0036] When resources are insufficient, return to step 3 to recalculate a shortest path P with the lowest packet loss cost. d .

[0037] Furthermore, traffic is classified based on network parameter information, where the network parameter information includes source and destination IP addresses, port numbers, protocol types, and DSCP.

[0038] Furthermore, the traffic classification includes:

[0039] For services with fixed source and destination IP address information, they are classified by IP address;

[0040] For business traffic without fixed source and destination IP addresses, it is classified by service port number or protocol number;

[0041] For service traffic carried by VPN, traffic classification is performed based on VPN information.

[0042] Furthermore, in the INT flow detection, an INT detection message is inserted into the classified data message, and the INT detection message is encapsulated in the data message in the form of an MPLS or SR stack bottom label.

[0043] Furthermore, the path deployment is delivered to the forwarding plane device in the form of MPLS labels, SR or SRv6 labels to build an end-to-end forwarding path.

[0044] An industrial Internet traffic scheduling system implements traffic scheduling based on the above-mentioned industrial Internet traffic scheduling method. It includes:

[0045] Traffic classification module, responsible for implementing traffic classification and sending the classification results to the INT management module;

[0046] The INT management module is responsible for inserting INT detection messages into classified data messages;

[0047] The traffic scheduling module calculates the optimal path based on the detected delay and packet loss information and the indicator requirements;

[0048] The path deployment module builds traffic forwarding channels and establishes the optimal forwarding path based on the optimal path results of the traffic scheduling module;

[0049] The global view module displays path information as well as traffic delay and packet loss rate information.

[0050] The beneficial effects of the present invention are:

[0051] (1) The present invention provides an industrial Internet traffic scheduling method and system that is compatible with MPLS, SR, and SRv6 networks, wherein flow classification is performed through ACL, and INT detection messages are inserted into the classified label messages for flow detection, thereby realizing fine-grained real-time data plane monitoring.

[0052] (2) The present invention adopts a multi-constraint parallel traffic algorithm to dynamically respond to network resource requirements, effectively reduce network packet loss rate, delay, traffic conflict and energy consumption, improve resource utilization, ensure network quality, and meet diverse needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a diagram of the SDN network architecture used in the embodiments of the present application;

[0054] Figure 2 This is an architecture diagram of the industrial Internet traffic scheduling system in an embodiment of this application;

[0055] Figure 3 This is a flowchart of a multi-constraint parallel routing algorithm in an embodiment of the present application. DETAILED DESCRIPTION

[0056] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be understood as limiting the present invention.

[0057] 1. Methods

[0058] like Figure 1 As shown, this embodiment adopts an SDN network architecture, including 1 SDN controller and multiple SDN switches. The specific number of SDN switches can be determined according to the service type of the industrial Internet. In this embodiment, each industrial Internet service platform (or system) is dual-homed to 2 SDN switches. In the example, there are 3 platform access SDN switches, 2 core SDN switches, 3 aggregation SDN switches, 1 system access SDN switch, and 1 terminal access SDN switch; wherein the industrial Internet terminal device is connected to the terminal access SDN switch, and then connected to the 3 aggregation SDN switches via the terminal access SDN switch, and the 3 aggregation SDN switches are respectively connected to the 2 core SDN switches, and the 2 core SDN switches are respectively connected to the 3 platform access SDN switches, and the traffic scheduling server is connected to the system access SDN switch via the SDN controller, and the system access SDN switch is then connected to any aggregation SDN switch.

[0059] For the above network, the embodiment of the present application provides a traffic scheduling method:

[0060] 1. Traffic classification

[0061] Traffic is classified based on network parameters such as source and destination IP addresses, port numbers, protocol types, and DSCP (Differentiated Services Code Point). Traffic classification distinguishes flows for control, identification, and operation and maintenance monitoring of the Industrial Internet, providing a basis for INT (Inband Network Telemetry) data packets.

[0062] When classifying traffic, the received data packets (sent by various industrial Internet devices) are first parsed to extract network parameter information such as IP (Internet Protocol) addresses and port numbers; then, according to the classification strategy set in the ACL (Access Control List) issued by the SDN controller, the data packets are classified according to different business needs.

[0063] Specific methods of traffic classification include but are not limited to:

[0064] 1) For services with fixed source and destination IP address information, classify them by IP address;

[0065] 2) For business traffic without fixed source and destination IP addresses, classify it by service port number or protocol number;

[0066] 3) For business traffic carried by VPN (Virtual Private Network), flow classification is performed based on VPN information.

[0067] Traffic classification based on IP address, port number, and protocol number requires network administrators to create different ACLs at the service ingress node and differentiate traffic using different ACL IDs (Identity Documents). Matching rules must be defined within the ACL ID. For example, if the target IP address of an industrial internet service is 10.1.1.1, ACL 2001 can be configured to match data with a destination address of 10.1.1.1 / 32. Subsequent INT flow detection only requires calling ACL 2001 to select the service.

[0068] VPN-based traffic classification requires network administrators to configure VPN services, identify them by VPN name or ID, and then direct industrial Internet traffic into the configured VPN service channel. INT flow detection simply requires calling the VPN name or ID to perform in-band telemetry on the current service.

[0069] 2.INT flow detection

[0070] Deploy INT in-stream detection to collect information such as latency and packet loss rate of industrial internet traffic. By inserting in-band telemetry information (INT probe messages) such as latency and packet loss into classified data packets and encapsulating these INT probe messages within the data packets as MPLS or SR stack bottom labels, industrial internet traffic carries telemetry data and supports multiple industrial extranets, including MPLS, SR, and SRv6. This collected telemetry information is then passed to the traffic scheduling module for subsequent optimal path calculation.

[0071] Before performing latency performance statistics, the SDN controller and SDN routers / switches must synchronize their PTP clocks and calculate latency using timestamps. Finally, the administrator must configure performance statistics, including point-by-point and end-to-end. Point-by-point performance statistics primarily monitor performance between directly connected devices, while end-to-end performance statistics primarily monitor the performance of the entire service path.

[0072] INT probe packets are injected into the network according to the network administrator's configuration and queued for forwarding along with service traffic. At each router / switch along the forwarding path, the probe packet extracts the internal device status and pushes it into the INT label stack. The probe packet also counts the ingress and egress timestamps and the number of labeled packets. The resulting statistics are then sent to the traffic scheduling system for analysis. The difference is calculated to determine the latency and packet loss rate, completing link performance analysis.

[0073] 3. Traffic Scheduling

[0074] Based on key QoS (Quality of Service) performance indicators such as latency, bandwidth, and packet loss in the network, the path constraint algorithm is used to calculate the optimal path to carry user traffic, ensuring that business traffic can always remain on the path that meets the QoS performance indicator requirements.

[0075] The path constraint algorithm includes a delay priority algorithm, a packet loss rate priority algorithm, and a multi-constraint parallel routing algorithm.

[0076] 1) Delay priority algorithm / packet loss rate priority algorithm

[0077] This approach aims to provide low-latency / low-packet-loss network services for real-time applications in industrial Internet scenarios. Specifically, it uses the Dijkstra algorithm to find the optimal path that minimizes latency / packet-loss. The steps involved are:

[0078] Step 1: Use Dijkstra algorithm to find the shortest path P with the lowest delay / packet loss cost. min ;

[0079] Step 2: If P min If all the indicators are met, P min As the optimal path; if the calculated path delay / packet loss rate does not meet the index requirements, it means that there is no optimal solution for the service path. In this case, the path with the minimum current delay / packet loss rate is selected as the optimal path.

[0080] 2) Multi-constraint parallel routing algorithm

[0081] Reference Figure 3 As shown in FIG, the multi-constraint parallel routing algorithm proposed in this embodiment calculates the optimal path by aggregating the cost. First, the shortest routing path P is found using the Dijkstra algorithm with the delay as the path cost. t , and then check whether this path meets the requirements based on the delay and packet loss requirements of the flow and the effectiveness of the limited resources. If all conditions are met, the optimal path is obtained; if not, the shortest path P in the network is calculated using the Dijkstra algorithm with the packet loss rate as the path cost. d, and check whether the packet loss requirement is met. If it is met, the Lagrange dual function is used to calculate the path aggregation cost, and the Dijkstra algorithm is used to find the shortest path r based on this aggregation cost. Replace it with P according to whether the delay and packet loss rate of the shortest path r meet the traffic requirements. t or P d , step by step iteration, and finally converge to an optimal path that satisfies the requirements of delay, packet loss and limited resources.

[0082] The specific steps include:

[0083] Step 1: Use Dijkstra algorithm to calculate the shortest path P with the minimum delay cost. t ;

[0084] Step 2: Compare the indicators to the following situations:

[0085] ① Path P t If all conditions are met, the path P t Output as the optimal path result;

[0086] ②Path P t If the latency does not meet the indicator requirements, it means that there is no optimal solution; this means that the network is not suitable for carrying this type of industrial Internet business and requires hardware or technology upgrades;

[0087] ③Path P t If the latency meets the requirement but the packet loss rate does not, proceed to step 3.

[0088] ④Path P t If the delay and packet loss rate meet the index requirements, but the resources (total bandwidth, remaining bandwidth, jitter, etc.) are insufficient, return to step 1 and recalculate a shortest path P with the minimum delay cost. t ;

[0089] Step 3: Use Dijkstra algorithm to calculate the shortest path P with the lowest packet loss cost. d ;

[0090] Step 4: Compare the indicators to the following situations:

[0091] ① Path P d If the packet loss rate meets the index requirements, proceed to step 5;

[0092] ②Path P d If the packet loss rate does not meet the indicator requirements, it means that there is no optimal solution; this means that the network is not suitable for carrying this type of industrial Internet business and requires hardware or technology upgrades;

[0093] Step 5: Use the Lagrange dual function to calculate the path, aggregate the constraints into the objective function, and define the Lagrange multiplier as λ. λ (e) is defined as the sum of the delay and weighted packet loss of link e, and uses the link aggregation cost G λ Calculate the shortest path r using Dijkstra's algorithm:

[0094]

[0095] λ←[T f (P t )-T f (P d )] / [D f (P t )-D f (P d )]

[0096] r←Dijkstra(s,t,G λ )

[0097] Where T(e) is the delay parameter of link e, D(e) is the packet loss rate parameter of link e, T(e) and D(e) are obtained by INT flow detection, T f (P t ) is the path P t The delay parameter, T f (P d ) is the path P d The delay parameter, D f (P t ) is the path P t The packet loss rate parameter, D f (P d ) is the path P d The packet loss rate parameter, T f (P t 、T f (P d ), D f (P t ), D f (P d ) is calculated by Dijkstra algorithm, r is the cost from source node s to target node t based on the aggregation cost G λ Calculate the shortest path.

[0098] ①When G is satisfied λ (r) = G(P t )=G λ (P d ), then execute step 6;

[0099] ② When G is not satisfied λ(r) = G(P t )=G λ (P d ), then determine whether the path r meets the requirements of both delay and packet loss rate. If the path r meets the requirements of delay and packet loss rate, the path r is set as the new P d On the contrary, if the path r does not meet the requirements of both delay and packet loss rate, the path r is set as the new P t ; Set path r to the new P d or P t Then, repeat step 5 until the solution that satisfies G is calculated. λ (r) = G(P t )=G λ (P d ) path r.

[0100] Step 6: Check whether path r meets resource requirements (total bandwidth, remaining bandwidth, jitter, etc.), where:

[0101] ① When resources are sufficient, path r is output as the optimal path result;

[0102] ② When resources are insufficient, return to step 3 and recalculate a shortest path P with the lowest packet loss cost. d .

[0103] 4. Path deployment

[0104] The calculated optimal path is delivered to the forwarding plane device using MPLS labels, SR labels, or SRv6 labels to build an end-to-end forwarding path.

[0105] 2. System

[0106] like Figure 2 As shown, based on the same inventive concept as the above-mentioned industrial Internet traffic scheduling method, the present application also provides an industrial Internet traffic scheduling system, including:

[0107] Traffic classification module: responsible for implementing traffic classification and sending the classification results to the INT management module;

[0108] INT management module: responsible for inserting INT detection messages into classified data messages;

[0109] Traffic scheduling module: Calculates the optimal path based on the detected delay and packet loss information and the indicator requirements;

[0110] Path deployment module: builds traffic forwarding channels and establishes the optimal forwarding path based on the optimal path results of the traffic scheduling module;

[0111] Global view module: displays all traffic classifications, traffic policies, and path information after traffic scheduling, as well as statistical information and alarm information such as traffic delay and packet loss rate.

[0112] The present invention is not limited to the above-mentioned embodiments. Any obvious improvement, replacement or modification that can be made by those skilled in the art without departing from the essence of the present invention shall fall within the scope of protection of the present invention.

Claims

1. An industrial Internet traffic scheduling method, characterized in that: The following steps are involved: Traffic classification, classifying industrial Internet traffic; INT on-stream detection: deploy INT on-stream detection on classified industrial Internet traffic to collect information on the latency and packet loss rate of industrial Internet traffic; Traffic scheduling, using a path constraint algorithm to calculate the optimal path to carry user traffic; Path deployment, builds an end-to-end forwarding path based on the calculated optimal path.

2. The industrial Internet traffic scheduling method according to claim 1 is characterized in that: The path constraint algorithm includes a delay priority algorithm and a packet loss rate priority algorithm: Step 1: Call the Dijkstra algorithm to find the shortest path P with the lowest delay / packet loss rate cost min ; Step 2: If P min If all the indicators are met, P min As the optimal path; If the calculated path delay / packet loss rate does not meet the index requirements, the path with the minimum current delay / packet loss rate is selected as the optimal path.

3. The industrial Internet traffic scheduling method according to claim 1 is characterized in that: The path constraint algorithm includes a multi-constraint parallel routing algorithm: First, use the Dijkstra algorithm to find the shortest routing path P with delay as the path cost. t ;Then check whether this path meets the requirements based on the delay and packet loss requirements of the flow and the availability of limited resources; If all conditions are met, the optimal path is obtained; If the requirement is not met, the Dijkstra algorithm is used to calculate the shortest path P in the network with the packet loss rate as the path cost. d , and check whether the packet loss requirement is met. If it is met, the Lagrange dual function is used to calculate the path aggregation cost, and the Dijkstra algorithm is used to find the shortest path r based on this aggregation cost; Finally, path r is replaced by P according to whether its delay and packet loss rate meet the traffic requirements. t or P d , repeat the iteration until convergence to obtain the optimal path.

4. The industrial Internet traffic scheduling method according to claim 3 is characterized in that: The multi-constraint parallel routing algorithm comprises the following steps: Step 1: Call the Dijkstra algorithm to calculate the shortest path P with the minimum delay cost t ; Step 2, Path P t If all conditions are met, path P t Output as the optimal path result; Path P t If the delay meets the requirement but the packet loss rate does not meet the requirement, execute step 3; Path P t If the delay and packet loss rate meet the requirements, but the resources are insufficient, return to step 1 to recalculate a shortest path P with the minimum delay cost. t ; Step 3: Call the Dijkstra algorithm to calculate the shortest path P with the lowest packet loss cost d ; Step 4, Path P d If the packet loss rate meets the requirement, execute step 5; Step 5: Use the Lagrange dual function to calculate the path, aggregate the constraints into the objective function, and define the Lagrange multiplier as λ. λ (e) is defined as the sum of the delay and weighted packet loss of link e, and uses the link aggregation cost G λ Calculate the shortest path r using Dijkstra's algorithm: λ←[T f (P t )-T f (P d )] / [D f (P t )-D f (P d )] r←Dijkstra(s,t,G λ ) Where T(e) is the delay parameter of link e, D(e) is the packet loss rate parameter of link e, T(e) and D(e) are obtained by INT flow detection, T f (P t ) is the path P t The delay parameter, T f (P d ) is the path P d The delay parameter, D f (P t ) is the path P t The packet loss rate parameter, D f (P d ) is the path P d The packet loss rate parameter, T f (P t 、T f (P d )、D f (P t )、D f (P d ) is calculated by the Dijkstra algorithm, r is the sum of the cost G from the source node s to the target node t. λ Calculate the shortest path; When G is satisfied λ (r) = G(P t )=G λ (P d ), then execute step 6; When G is not satisfied λ (r) = G(P t )=G λ (P d ), then determine whether path r meets both the delay and packet loss rate requirements. If path r meets both the delay and packet loss rate requirements, path r is set as the new P d On the contrary, if path r does not meet the requirements of both delay and packet loss rate, path r is set as the new P t ; Set path r as the new P d or P t Then, repeat step 5 until the solution that satisfies G is calculated. λ (r) = G(P t )=G λ (P d ) path r; Step 6: Check whether the path r meets the resource requirements, where: When the resources are sufficient, the path r is output as the optimal path result; When resources are insufficient, return to step 3 to recalculate a shortest path P with the lowest packet loss cost. d .

5. The industrial Internet traffic scheduling method according to claim 1 is characterized in that: Traffic is classified based on network parameter information, which includes source and destination IP addresses, port numbers, protocol types, and DSCP.

6. The industrial Internet traffic scheduling method according to claim 5 is characterized in that: The traffic classification includes: For services with fixed source and destination IP address information, they are classified by IP address; For business traffic without fixed source and destination IP addresses, it is classified by service port number or protocol number; For service traffic carried by VPN, traffic classification is performed based on VPN information.

7. The industrial Internet traffic scheduling method according to claim 1 is characterized in that: In the INT flow detection, an INT detection message is inserted into the classified data message, and the INT detection message is encapsulated in the data message in the form of an MPLS or SR stack bottom label.

8. The industrial Internet traffic scheduling method according to claim 1 is characterized in that: The path deployment is delivered to the forwarding plane device in the form of MPLS labels, SR or SRv6 labels to build an end-to-end forwarding path.

9. An industrial Internet traffic scheduling system, characterized in that: Traffic scheduling is implemented according to the industrial Internet traffic scheduling method as described in any one of claims 1 to 8.

10. The industrial Internet traffic scheduling system according to claim 9, characterized in that: include: Traffic classification module, responsible for implementing traffic classification and sending the classification results to the INT management module; INT management module, responsible for inserting INT detection messages into classified data messages; The traffic scheduling module calculates the optimal path according to the detected delay and packet loss information and the indicator requirements; The path deployment module builds the traffic forwarding channel and establishes the optimal forwarding path based on the optimal path result of the traffic scheduling module; The global view module displays path information as well as traffic delay and packet loss rate information.

Images