Round-cycle queue forwarding scheduling model and method for mixed time-sensitive traffic, and terminal

Abstract

The invention relates to the technical field of time-sensitive networks, and particularly provides a round-robin queue forwarding scheduling model and method for mixed time-sensitive traffic, and a terminal. The circular queue forwarding scheduling model comprises a delay sensitive queue, a first time sensitive queue and a second time sensitive queue. The method comprises the following steps: creating two time slot systems; defining fixed injection time and a queue scheduling strategy for the time-triggered traffic; carrying out pre-scheduling planning according to the knowability of the time-triggered stream, and obtaining injection time corresponding to the time-triggered stream and a corresponding injection queue; and for the emergently arrived event trigger flow, incremental scheduling is carried out according to real-time link resources, and a suitable time slot is obtained for forwarding. According to the invention, the event trigger traffic and the time trigger traffic are placed in different queues, and the model designs a scheduling mode of the corresponding event trigger traffic, so that the isolation of traffic types is effectively realized, the scheduling and forwarding of the time sensitive traffic are optimized, and the network communication efficiency is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of time-sensitive networks, and particularly to a cyclic queue forwarding scheduling model, method, and terminal for hybrid time-sensitive traffic. Background Art

[0002] With the development of the unmanned driving field and the automated industry, the traditional Ethernet with the best-effort transmission principle can no longer meet the requirements of some high-precision industries. For this reason, time-sensitive networks have emerged. Different from traditional Ethernet, time-sensitive networks can provide reliable delivery and low-latency and low-jitter transmission guarantees for some traffic with high real-time requirements, and this traffic is called time-sensitive traffic. To standardize the development of time-sensitive networks, the IEEE specifically established a TSN group in 2015, aiming to specify a series of standards for time-sensitive networks. Currently, mainstream scheduling models, such as the Time-Aware Shaper (TAS) and the Circular Queue Forwarding Model (CQF), are all for periodic and known time-sensitive traffic, that is, time-triggered traffic; for bursty time-sensitive traffic, that is, event-triggered traffic, no consideration and planning have been carried out. However, in actual application scenarios, there are not only periodic time-triggered traffic, such as the periodic control signals of driverless cars, but also some event-triggered traffic for facing emergencies, such as the emergency braking of cars. If event-triggered traffic is not considered during the scheduling process, it will not only affect the transmission of emergency event traffic, but may also occupy the bandwidth and queue resources reserved in advance for time-triggered traffic, affecting the transmission of the already planned time-triggered traffic. To address the above problems, based on some existing scheduling models, a new scheduling mechanism is designed to solve the scenario of hybrid scheduling of time-triggered and event-triggered traffic.

[0003] Time-sensitive networks mainly provide deterministic low latency and reliability guarantee for traffic with high real-time requirements. Among them, the purpose of traffic scheduling is to provide deterministic low latency for traffic. The essence of traffic scheduling is to determine the forwarding time of each piece of traffic at each switch node. The current mainstream scheduling method is synchronous traffic scheduling based on global time synchronization, such as the TAS scheduling model and the CQF scheduling model mentioned above. By uniformly managing the global time and space resources, the forwarding path and forwarding time of each piece of traffic are determined. And the way to achieve scheduling in time-sensitive networks is to determine the forwarding time of each piece of traffic by configuring the gating list. The TAS scheduling model schedules each piece of traffic one by one. All synchronous traffic scheduling is based on time slots for resource division and management. The so-called time slot can be understood as a time unit. Since TAS adopts refined scheduling, its time slot division is also very fine. The mechanism of TAS requires that by configuring the forwarding time of each piece of traffic on each switch, it is ensured that there is only one piece of time-sensitive traffic on each switch port in each time slot, and there will be no traffic competition and queuing. Through the reasonable allocation of global time slots and port resources, it can be said that each piece of traffic is transmitted according to the predetermined time and path, zero jitter of latency can be achieved, thereby realizing deterministic latency for traffic. At the same time, the specific scheduling time of traffic will be planned according to the corresponding latency index requirements and algorithms to meet the requirements of low latency. However, due to the refined scheduling of TAS, the configuration of its gating list is extremely complex. The complexity of the corresponding scheduling algorithm is also very high. It is not feasible for large-traffic scenarios.

[0004] To simplify the complexity of scheduling, the TSN group proposed another scheduling model - the dual-queue CQF. Different from the TAS scheduling method that is precise to each flow, CQF schedules based on queues. For this reason, the CQF scheduling model does not need to be as precise as the TAS scheduling model in dividing time slots, and the gating list of CQF is very simple. In the scheduling model of the dual-queue CQF, the gating lists of the two queues within each time slot are mutually exclusive, that is, one queue is used to receive data and the other queue is used to send data. In time slot T0, Q0 is in the sending state and Q1 is in the state of receiving data. In the next time slot, the states of the gating lists are mutually exclusive. Based on global clock synchronization, the operation of CQF is like a gear mechanism, sending the flow from the source node to the destination node step by step. Because of this mechanism, CQF requires that the transmission of the flow from one switch to another switch be completed within one time slot. Therefore, the delay is affected by the number of routing hops. When the path of the flow is determined, its delay is basically determined. However, compared with TAS, CQF has a delay jitter of two time slot sizes. And it is not difficult to see that CQF does not control the gating of the intermediate switching nodes in real time. Therefore, after the flow is injected into the network, the transmission of the flow is basically no longer controlled. Based on this, the scheduling strategy of CQF mainly focuses on the configuration of the flow injection time slot, selecting a reasonable time to inject into the network and a reasonable queue length in the switch on the basis of meeting the low-delay requirement, so as to ensure that there is no packet loss due to insufficient queue resources in each hop queue.

[0005] According to the mechanism of the dual-queue CQF mentioned above, it can be concluded that the size of the CQF time slot determines the size of the receiving and sending time window, and the size of the time slot division must ensure that all the data in the sending queue can be sent and all the data from the corresponding upstream nodes can be completely received. Due to the existence of transmission and processing delays, there is also a part of the bandwidth that cannot be utilized by time-triggered traffic in the CQF mechanism, which is also called dead time. For the data sent during the dead time, the downstream node cannot completely receive it in the receiving window. Therefore, there is a certain waste of bandwidth resources here.

[0006] It is not difficult to see from the mechanisms of the above two scheduling models that these are all for predictable traffic (and most of the research is on periodic traffic) to calculate the gating situation in advance for traffic forwarding. On the basis that the time-triggered sensitive traffic has been pre-allocated the corresponding time and queue resources, if some burst traffic occurs, not only may the event-triggered traffic cause packet loss because no corresponding resources are reserved for the burst traffic, but also the event-triggered traffic may preempt the resources of the original time-triggered traffic and affect the transmission of the planned traffic. Obviously, the traditional model is not very suitable for the scenario of mixing time-triggered and event-triggered time-sensitive traffic.

[0007] Prior art one, a Chinese patent with application number: 202410278029.X discloses an intelligent substation hybrid flow scheduling method and system based on a time-sensitive network. The method includes obtaining various types of traffic information of the intelligent substation and generating an intelligent substation traffic set based on the traffic information; constructing a hybrid flow scheduling mechanism based on a time-aware shaper according to the burst traffic transmission characteristics in the substation; formalizing the overall schedulability of burst flows and schedulable flows under scheduling constraints based on the hybrid flow scheduling mechanism; and calling a greedy tabu search algorithm based on the formalized overall schedulability to determine the gating list for TAS protocol traffic scheduling. Although it solves the problem of untimely GOOSE 1A transmission; uses the hybrid flow scheduling mechanism to evaluate the superiority of the scheduling scheme; and combines with the greedy tabu search algorithm to improve the overall schedulability of GOOSE 1A and SMV. However, the configuration of the gating list of the TAS protocol is extremely complex, and the complexity of the scheduling algorithm is also very high. There is a problem that the feasibility needs to be further improved in the face of large-traffic scenarios.

[0008] Prior art two, a Chinese patent with application number 202410660848.0 discloses a multi-scenario delay analysis method and system for a time-sensitive network in an intelligent substation. The method includes: generating an intelligent substation traffic delay calculation model according to the network structure of the intelligent substation and the internal structure of the time-sensitive network TSN switch; determining the transmission delay of GOOSE traffic and MMS traffic after traffic scheduling combined with a time-aware shaper TAS and a credit-based shaper CBS mechanism in a single protection window scenario; determining the transmission delay of GOOSE traffic and MMS traffic in a multi-protection window scenario; and determining the transmission delay of GOOSE traffic and MMS traffic based on the intelligent substation traffic delay calculation model in a multi-hop network scenario. Although it can accurately analyze the delay of the traffic transmission process in the substation and ensure the transmission performance of critical traffic. However, it does not consider event-triggered traffic, which will affect the transmission of emergency traffic and may also occupy the bandwidth and queue resources reserved in advance for time-triggered traffic, affecting the transmission of the already planned time-triggered traffic.

[0009] Prior Art III, a Chinese patent with application number 202311635255.0, discloses a deterministic delay routing scheduling method and system based on segment routing. The method includes: acquiring data packets and calculating the shortest delay path from the source node to the target node. Calculating the deadline offset vector of each relay node on the shortest delay path and writing it into the header information of the data packet. Transmitting the data packet along the shortest delay path. Before queuing at the relay node, calculating the discard probability of the data packet. If the probability is higher than the randomly generated number, the data packet is discarded at the current node. Using a three-queue cyclic queuing and forwarding model established by the cyclic queue forwarding model to calculate the queue into which the data packet should be queued at each relay node to minimize the queuing delay of the data packet at each relay node. Setting each relay node to transmit the data packet before its corresponding deadline offset vector. Although it can meet the transmission requirements of low latency and load balancing for deterministic delay network routing scheduling, there is a part of the bandwidth that cannot be utilized by time-triggered traffic, resulting in a certain waste of bandwidth resources.

[0010] Currently, Prior Art I, Prior Art II, and Prior Art III have the problem that no corresponding resources are reserved for bursty traffic, resulting in packet loss of event-triggered traffic and the event-triggered traffic preempting the resources of the original time-triggered traffic, thus affecting the transmission of the planned traffic. Therefore, the present invention provides a cyclic queue forwarding scheduling model, method, and terminal for hybrid time-sensitive traffic. Summary of the Invention

[0011] To achieve the above object, the present invention adopts the following technical solutions:

[0012] On the one hand, the present invention provides a cyclic queue forwarding scheduling model for hybrid time-sensitive traffic, including:

[0013] A delay-sensitive queue for event-triggered traffic, which processes traffic sensitive to delay; manages data packets of event-triggered traffic and performs priority scheduling during network congestion;

[0014] A first time-sensitive queue for managing the first type of time-triggered traffic, whose data packets arrive within a specified time interval and are processed within a predetermined time; by isolating the first time-sensitive queue from the delay-sensitive queue;

[0015] A second time-sensitive queue for cooperating with the first time-sensitive queue to form a ping-pong operation for receiving and parallel processing, ensuring the queue throughput while controlling the queuing delay.

[0016] In an optional embodiment, it further includes:

[0017] A first receiving gate, connected to the first time-sensitive queue, responsible for controlling whether the first time-sensitive queue belongs to the receiving queue in the current time slot. If in the receiving state, it allows data to be input into the queue;

[0018] A second receiving gate, connected to the second time-sensitive queue, is responsible for controlling whether the first time-sensitive queue is in a receiving state in the current time slot, and if in the receiving state, allows data to be input into the queue;

[0019] A first sending gate is used to determine whether the first time-sensitive queue is in a sending state, and if in the sending state, send the data packet in the queue to a final destination;

[0020] a second sending gate, used to determine whether the second time-sensitive queue is in a sending state, and if so, to send the data packets in the queue to a final destination;

[0021] The gate list module is responsible for configuring the switch state of each sending and receiving gate according to the traffic scheduling results calculated by the upper layer. The traffic scheduling results are reflected in the switch time and duration of each sending and receiving gate.

[0022] Gating logic GCL is used for traffic scheduling and priority management. It follows the established control strategy and dynamically adjusts the traffic forwarding time slot according to the real-time traffic status to ensure that event-triggered traffic is given priority while not affecting the established scheduling status of time-triggered traffic.

[0023] Another aspect of the present invention provides a method for creating two time slot systems, one for event-triggered traffic and the other for time-triggered traffic, and the two time slots need to be in multiple relationship; setting bandwidth reservation for time-triggered traffic time slots in the network to provide deterministic bandwidth resources for event-triggered traffic scheduling, thereby realizing independent scheduling of the two types of traffic; dividing data packets entering the network into time-triggered traffic and event-triggered traffic according to the nature of the traffic; setting the time slot size of the time-triggered traffic to not only meet the requirements of the cyclic forwarding mechanism but also reserve a part of the bandwidth to provide margin for the scheduling of the event-triggered flow;

[0024] Pre-schedule the time-triggered traffic with known arrival time and traffic size globally to determine the injection time and queue. For the event-triggered traffic that arrives suddenly, incremental scheduling is adopted according to the real-time remaining resources of the link; the time slots of the event-triggered traffic are divided into extremely detailed time slots, and each switch port can only receive one traffic forwarding in each time slot; schedule the time-triggered traffic one by one, and select the appropriate time slot for sending according to the bandwidth resources reserved in the current network; add the time-triggered traffic to the corresponding queue, follow the CQF scheduling mechanism, and adjust the penetration time according to the time slot resource table; before scheduling new traffic, check the status of the current switch port, and monitor the current reserved bandwidth when injecting traffic;

[0025] According to the predetermined time interval and injection time, the time-triggered traffic is added to the corresponding queue, following the CQF scheduling mechanism. According to the mechanism of event-triggered traffic, the biggest feature of time-triggered traffic is the known planned traffic, which can be scheduled in advance, and there is no sudden new traffic; choose the appropriate sending time to ensure that important traffic in the network will not be queued in a queue for too long or exceed the queue size and be discarded. By reasonably arranging the sending time, the traffic originally sent together can be staggered; dynamically adjust the sending time of time-triggered traffic according to the actual traffic situation.

[0026] For the received time-triggered traffic, select the time slot for forwarding according to the pre-scheduled results. For the newly arrived event-triggered traffic, schedule it according to the current global network status (queue resources and reserved bandwidth resources), and select the time slot for forwarding after obtaining the scheduling results. Record the forwarding results and bandwidth usage. Continuously monitor the traffic status, bandwidth usage and process, and select the appropriate sending time according to the actual traffic situation to ensure that important traffic in the network will not be queued in a queue for too long or exceed the queue size and be discarded. The injection time is the sending time.

[0027] In an optional implementation, the incremental scheduling strategy adopts an incremental method, scheduling each flow one by one, generating a flow, scheduling a flow, and scheduling only one flow at a time based on the current state; it is determined by adjusting the time from the generation of burst traffic to the injection into the network.

[0028] In an optional implementation manner, the incremental scheduling strategy specifically includes:

[0029] Monitor the status of each switch port in the network in real time, detect traffic arrival, determine its type as event-triggered traffic, and obtain characteristics such as the size and occurrence time of the traffic; before scheduling new traffic, evaluate the current network status, including the status of scheduled traffic, bandwidth usage, and queue length;

[0030] Find available time slots in the divided time slots, and select time slots based on the characteristics of the current traffic, network status, and other traffic already in the queue; when selecting time slots, consider the scheduling of time-triggered traffic and the set bandwidth reservation; after identifying the time slot, only schedule the event-triggered traffic that has arrived; during the traffic scheduling process, adjust the injection time of the traffic;

[0031] Take the selected event-triggered traffic from the relevant queue and forward it to the target location according to the scheduling result; for the forwarded traffic, record its forwarding results, including the sending time slot, bandwidth usage and processing delay information corresponding to the traffic;

[0032] Continuously monitor the status of traffic and bandwidth usage in the network, dynamically adjust the scheduling strategy for subsequent traffic based on the actual traffic situation monitored, and use scheduling algorithms to plan the injection time for sudden event-triggered flows and time-triggered flows because of their known nature.

[0033] In an optional implementation manner, selecting a suitable time slot for transmission includes:

[0034] Assign a unique number to each time slot, monitor the status of each switch port in real time, and promptly discover newly arrived event-triggered traffic; check the available fine time slots according to the bandwidth situation and detailed time slot division, and find the time slot for sending new traffic. The selected time slot does not conflict with the scheduled traffic;

[0035] If the selected time slot conflicts or exceeds the reserved bandwidth, make time adjustments in the available time slots and move the injection time to a suitable time slot; confirm that the selected time slot can effectively accommodate the new traffic and record it;

[0036] The selected event-triggered traffic is taken out from the corresponding queue and forwarded on the selected time slot; within the selected time slot, the traffic is sent to the target location as planned; for the forwarded traffic, detailed forwarding information is recorded, including the time slot number used, actual bandwidth usage, and delay data of traffic processing, to provide a basis for subsequent analysis.

[0037] Another aspect of the present invention provides a terminal, comprising a switch or a network card, characterized in that the scheduling result of the circular queue forwarding scheduling method for mixed time-sensitive traffic is obtained, a time slot corresponding to each traffic entering the network is generated, and the obtained result is deployed to the gating list information of the switch or the network card.

[0038] In an optional implementation, it includes:

[0039] The framing module is used to encapsulate the data transmitted by the upper layer to form the frame structure required by the network protocol; in the process of framing, necessary header and tail information are added; through framing, continuous data streams are integrated into frames; and the generated frames are ensured to comply with the corresponding network protocol specifications;

[0040] The frame parsing module is used to parse the received frames and extract the required information, including extracting the data payload, identifying the sender and receiver, and the protocol-related fields; the frame parsing process verifies the integrity of the data, including determining whether the data is damaged during transmission through checksums or other error detection mechanisms; and passes the parsed data or control information to the downstream modules for processing and scheduling decisions;

[0041] The scheduling module is used to manage and control the transmission order of data packets. According to the information from the virtual output queue module and the gating information of the gating logic module, it formulates the scheduling strategy for each data packet and generates the corresponding time slot scheduling results. By reading the gating list, it selects the data to be output based on the priority of different queues to realize the priority scheduling of traffic. It monitors the occupancy status of the current port and determines which queues' frames are scheduled. It returns the scheduling results to the virtual output queue module to process the corresponding data frames.

[0042] In an optional implementation, the scheduling module includes:

[0043] A virtual output queue submodule, which is used to create an independent queue for each input port and output port pair;

[0044] The in-queue submodule is used to temporarily store the queue of data packets to be processed or forwarded; data packets arrive from devices in the network and are stored in the in-queue, and after the corresponding scheduling and forwarding process is completed, they are sent to the next destination;

[0045] The scheduling submodule is used to arrange, organize and control the transmission order and time of data packets in network switches or routers; the scheduling strategy determines which data packets or traffic should be processed first under different conditions;

[0046] The scheduler submodule is used to manage and control the flow of data packets from the input queue to the output queue, and determines the data packets that each input queue sends to the output link at a specific time.

[0047] The present invention effectively implements the isolation of traffic types by placing event-triggered traffic and time-triggered traffic in different queues. ET Queue is used to process delay-sensitive event-triggered traffic, while TT Queue A and TTQueue B manage time-triggered traffic; isolation ensures that the processing priority and time characteristics of various types of traffic are effectively protected. In the case of network congestion, ET Queue can perform priority scheduling when processing delay-sensitive traffic. When network resources are tight, the system will give priority to delay-sensitive traffic without affecting the response time of time-triggered traffic due to high load. The first time-sensitive queue TT Queue A and the second time-sensitive queue TT Queue B each process traffic with different time constraints; through effective resource allocation and scheduling, the system ensures that these traffic can be processed and transmitted within their agreed time range; through classification and isolation of traffic, the overall system can run more stably during peak load periods, ensuring timely processing of time-sensitive traffic, thereby reducing delay and packet loss rate; through reasonable queue design, network resources can be effectively utilized to avoid resource waste. Coordination and management between different traffic types can maximize the throughput of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0049] Figure 1 A schematic diagram of the structure of a circular queue forwarding scheduling model for mixed time-sensitive traffic provided in Embodiment 1 of the present invention;

[0050] Figure 2 This is a flow chart of a circular queue forwarding scheduling method for mixed time-sensitive traffic provided in Embodiment 3 of the present invention;

[0051] Figure 3 This is a schematic diagram of a circular queue forwarding scheduling method for mixed time-sensitive traffic provided in Embodiment 3 of the present invention;

[0052] Figure 4 This is a process diagram of the incremental scheduling strategy provided in Embodiment 4 of the present invention;

[0053] Figure 5 A process diagram of selecting a suitable time slot for transmission provided in Embodiment 6 of the present invention;

[0054] Figure 6 This is a system design block diagram of a terminal provided in Embodiment 7 of the present invention;

[0055] Figure 7 A block diagram of an electronic device provided by an embodiment of the present invention;

[0056] Figure 8 A block diagram of a computer-readable storage medium provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present invention will be described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0058] In the following, the terms "first", "second", etc. are used only for convenience of description and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0059] In the present invention, unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense, for example, "connection" can be a fixed mechanical connection, or a detachable mechanical connection, or integrated; or, "connection" can be a direct connection, or an indirect connection through an intermediate medium. In addition, unless otherwise clearly specified and limited, the term "coupling" should be understood in a broad sense, for example, "coupling" can be a direct electrical connection, such as physical contact and electrical conduction between two components, and can also be understood as electrical connection between different components in a circuit structure through physical lines such as copper foil or wires on a printed circuit board (PCB) that can transmit electrical signals to transmit electrical signals; or, "coupling" can be an indirect electrical connection between two components through an intermediate medium; or, "coupling" can be an electrical connection between two components in an air-spaced / non-contact manner, for example, two components are electrically connected by capacitive coupling to transmit electrical signals.

[0060] In the embodiments of the present invention, directional terms such as "up", "down", "left" and "right" may be defined including but not limited to the orientation relative to the schematic placement of the components in the drawings. It should be understood that these directional terms may be relative concepts, which are used for relative description and clarification, and may change accordingly according to the change of the orientation of the components in the drawings.

[0061] The embodiments of the present invention are designed to effectively manage and schedule mixed types of time-sensitive traffic, and can maintain efficient service capabilities under different network conditions. This scheduling mechanism not only improves network performance and resource utilization, but also provides good support for multi-scenario and multi-service requirements, and has important technical and application value. The structured queue management method enhances the flexibility and scalability of the network, ensures the quality of real-time services, and improves user satisfaction. It is an important tool for network management.

[0062] The key technical features of the embodiments of the present invention are as follows: a three-queue architecture is adopted, namely, a delay-sensitive queue (ETQueue), a first time-sensitive queue (TT Queue A) and a second time-sensitive queue (TT Queue B). Including: Traffic isolation: different types of traffic are stored in independent queues to avoid competition and mutual interference between traffic. For example, ET Queue prioritizes event-triggered traffic, while TT Queue A and TT Queue B focus on time-triggered traffic; efficient resource utilization: by allocating specific queue resources to each traffic type, the model can maximize the efficiency of network resource utilization, avoid resource waste and unnecessary delays. The model implements a dynamic priority scheduling mechanism to cope with network congestion. Event-triggered traffic is prioritized when the network load is high. The model ensures that delay-sensitive traffic is processed first to maintain the stability of core services; time-sensitive traffic is guaranteed. The traffic in the first time-sensitive queue and the second time-sensitive queue can be guaranteed to be processed within the agreed time range, effectively reducing delays. The enhanced CQF scheduling model can flexibly respond to various types of traffic and provide better scheduling efficiency according to the needs of mixed scheduling scenarios. Compared with the traditional dual-queue CQF model, it adds a queue, enabling the system to better handle multiple types of traffic and avoid resource conflicts between different types of traffic.

[0063] Embodiment 1:

[0064] like Figure 1 As shown, an embodiment of the present invention provides a circular queue forwarding scheduling model for mixed time-sensitive traffic, including:

[0065] Delay-sensitive queue ET Queue, used for event-triggered traffic, handles delay-sensitive traffic; manages data packets of event-triggered traffic and performs priority scheduling when the network is congested;

[0066] A first time-sensitive queue is used to manage the first type of time-triggered traffic, whose packets arrive within a specified time interval and are processed within a predetermined time; by isolating the first time-sensitive queue from the delay-sensitive queue;

[0067] The second time-sensitive queue is used to cooperate with the first time-sensitive queue to form a pingpong operation for receiving parallel processing, and ensure queue throughput on the basis of controlling the queuing delay. In the above embodiment, the circular queue forwarding scheduling model for mixed time-sensitive traffic realizes effective management and scheduling of traffic of different natures by designing the delay-sensitive queue ET Queue, the first time-sensitive queue TT Queue A and the second time-sensitive queue TT Queue B.

[0068] In this embodiment, the model effectively implements the isolation of traffic types by placing event-triggered traffic and time-triggered traffic in different queues. ET Queue is used to process delay-sensitive event-triggered traffic, while TT Queue A and TT Queue B manage time-triggered traffic; isolation ensures that the processing priority and time characteristics of various types of traffic are effectively protected. In the case of network congestion, ET Queue can perform priority scheduling when processing delay-sensitive traffic. When network resources are tight, the system will give priority to delay-sensitive traffic without affecting the response time of time-triggered traffic due to high load. The first time-sensitive queue TT Queue A and the second time-sensitive queue TT Queue B each process traffic with different time constraints; through effective resource allocation and scheduling, the system ensures that these traffic can be processed and transmitted within their agreed time range; through classification and isolation of traffic, the overall system can run more stably during peak load periods, ensuring timely processing of time-sensitive traffic, thereby reducing delay and packet loss rate; through reasonable queue design, network resources can be effectively utilized to avoid resource waste. Coordination and management between different traffic types can maximize the throughput of the system.

[0069] The significance achieved by this embodiment is as follows: the scheduling and forwarding of time-sensitive traffic is optimized, and the communication efficiency of the network is enhanced, especially for the support of real-time, critical services, such as voice, video conferencing, and online games; by ensuring the timely transmission of delay-sensitive and time-triggered traffic, the user experience when using real-time applications and services is significantly improved, and user dissatisfaction caused by delays and packet loss is reduced; when multiple business needs coexist, the model can support the needs of complex application scenarios and provide reliable network support for various types of applications; it continues to support changes and expansions in network traffic, adapts to changes in future demands for different types of traffic, and enables the system to quickly adjust and adapt as technology develops and business needs change; through this traffic isolation mechanism, network operators can better plan and manage network resources, providing a better basis for future network upgrades and maintenance.

[0070] In summary, the circular queue forwarding scheduling model for mixed time-sensitive traffic in this embodiment not only improves the performance and resource utilization of the network, but also achieves efficient network service capabilities through effective isolation and priority scheduling of traffic, providing good support for the needs of multiple scenarios and multiple services, and has important technical and application significance. This embodiment adopts a three-queue mode, two queues process time-triggered sensitive traffic, the queue length affects the time slot size and gating delay, and the other queue processes event-triggered sensitive traffic, and the queue length is the maximum frame length of a frame.

[0071] This embodiment is based on the mainstream CQF scheduling model and is improved to adapt to the situation of the hybrid scheduling model; the enhanced CQF model of this embodiment is different from the classic dual-queue CQF model and uses three queues to store data. Two of the queues are used to cyclically forward time-triggered traffic, and another queue is configured separately for event-triggered traffic. When traffic passes through the network card, the network card identifies the traffic type and forwards the event-triggered traffic into an independent queue for event-triggered traffic. The queue for processing time-triggered flow adopts a dual-queue cyclic forwarding mechanism. In the same time slot, one queue is used as a receiving queue, which only receives frames input by the network card, and the other queue is used as a sending queue to inject the accumulated frames in the queue into the network. The functions of the two queues are exchanged in the next time slot. The network card determines the roles of the two queues by reading the information of the gating list, that is, it is determined by the switch status of the receiving gate and the sending gate, and the received time-triggered flow is transferred to the receiving queue of the current time slot. The two types of traffic are isolated from the queue resources to avoid the influence of event-triggered traffic occupying time-triggered traffic queue resources. .

[0072] Embodiment 2:

[0073] like Figure 1 As shown, based on Example 1, the circular queue forwarding scheduling model for mixed time-sensitive traffic provided by the embodiment of the present invention further includes:

[0074] The first receiving gate is connected to TT Queue A and is used to control the incoming traffic from the previous stage of reception. It cooperates with the sending gate to determine whether TT Queue A is in the sending state or the receiving state. If it is in the receiving state, it is allowed to receive data. The second receiving gate is connected to TT Queue B and is used to control the incoming traffic from the previous stage of reception. It cooperates with the sending gate to determine whether TT Queue is in the sending state or the receiving state. If it is in the receiving state, it is allowed to receive data. The first sending gate is used to send the data packets in TT Queue A to the final destination, and cooperates with the queue receiving gate to determine whether TT Queue A is in the sending state or the receiving state. If it is in the sending state, the data from the time-triggered traffic queue is transmitted to the destination in a timely manner. The second sending gate is used to send the data packets in TTQueue B to the final destination, and cooperates with the queue receiving gate to determine whether TT Queue B is in the sending state or the receiving state. If it is in the sending state, the data from the time-triggered traffic queue is transmitted to the destination in a timely manner. The gate control list module is responsible for configuring the switch state of each sending and receiving gate according to the traffic scheduling results calculated by the upper layer, where the traffic scheduling results are reflected in the switch time and duration of each sending and receiving gate.

[0075] The gate control logic GCL is used for traffic scheduling and priority management. It follows the established control strategy and dynamically adjusts the traffic forwarding time slot according to the real-time traffic status to ensure that the event-triggered traffic is given priority while not affecting the established scheduling state of the time-triggered traffic. In the above embodiment, this embodiment uses two receiving gates and two sending gates to determine the state of two time-sensitive traffic queues. The two queues follow the mechanism of circular forwarding. In the same time slot, one queue serves as a receiving queue to receive the time-sensitive traffic transmitted by the network card, and the other queue serves as a sending queue to input the data stored in the previous time slot into the network. The two queue states are switched in the next time slot. Through the mechanism of circular forwarding, the queuing delay of the traffic is controlled while ensuring the throughput of the time-triggered traffic. The gate control logic GCL implements the circular forwarding mechanism by sending the gate control information to the sending and receiving gates of each queue. The circular forwarding mechanism enables the transmission time of the time-sensitive traffic between two nodes in the link to have an upper bound, that is, a time slot unit. When the path is determined, the upper and lower bounds of the transmission delay of the entire link can be determined. The event-triggered flow has an independent queue. For the event-triggered flow that is about to enter the network, the upper layer will perform real-time scheduling and send the scheduling results to the gating logic for execution, ensuring that the event-triggered flow will not conflict in network transmission and that the reserved bandwidth resources can ensure that the delay jitter upper bound of the time-triggered flow is not affected. Through the structural design of this embodiment, the system ensures that under high load conditions, the processing capacity of time-sensitive traffic remains stable without being affected by potential event-triggered traffic, improving the overall network reliability and performance, and realizing the rapid forwarding of urgent event-triggered flows.

[0076] The first receiving gate is connected to the first time-sensitive queue and is responsible for controlling whether the first time-sensitive queue belongs to the receiving queue in the current time slot. If it is in the receiving state, data is allowed to enter the queue; the second receiving gate is connected to the second time-sensitive queue and is responsible for controlling whether the first time-sensitive queue is in the receiving state in the current time slot. If it is in the receiving state, data is allowed to enter the queue; the first sending gate is used to determine whether the first time-sensitive queue is in the sending state. If it is in the sending state, the data packets in the queue are sent to the final destination; the second sending gate is used to determine whether the second time-sensitive queue is in the sending state. If it is in the sending state, the data packets in the queue are sent to the final destination; the gate control list module is responsible for configuring the switch state of each sending and receiving gate according to the traffic scheduling results calculated by the upper layer, wherein the traffic scheduling results are reflected in the switch time and duration of each sending and receiving gate.

[0077] The significance achieved by this embodiment is as follows: it meets the diverse application needs, especially the services with high real-time requirements, such as financial transactions, video conferencing, and online games; it ensures that the needs of different traffic types are effectively solved and improves the user experience; by optimizing the scheduling and resource allocation of traffic, the network can configure bandwidth and resources with higher efficiency, reduce idle resources, and improve the overall throughput of the system; effectively isolate and prioritize event-triggered traffic and time-triggered traffic, significantly reducing the delay and packet loss rate during data transmission, thereby improving the service quality of the network; the model has strong adaptability, and as the business needs change, the traffic processing and scheduling strategies can be flexibly adjusted, which provides convenience for the evolution of future technologies and the expansion of the network; through the reasonable configuration of gating logic and gateways, the system can more effectively monitor network traffic conditions, perform troubleshooting and performance analysis, and provide reliable data support for network managers.

[0078] In summary, the circular queue forwarding scheduling model for mixed time-sensitive traffic in this embodiment achieves efficient traffic management and precise scheduling in a complex and changeable network environment through its construction, which fully reflects the good technical effect and application significance. It provides strong support for various traffic characteristics and application scenarios that may appear in the future, and lays the foundation for achieving high-performance, high-reliability and high-user satisfaction network services.

[0079] Embodiment 3:

[0080] like Figure 2 As shown, based on Example 1, the circular queue forwarding scheduling method for mixed time-sensitive traffic provided by the embodiment of the present invention includes the following steps:

[0081] Step S100: Create two time slot systems, one for event-triggered traffic and the other for time-triggered traffic, and the two time slots must be in multiples; set the bandwidth reservation for the time-triggered traffic time slot in the network to provide deterministic bandwidth resources for event-triggered traffic scheduling, so as to achieve independent scheduling of the two types of traffic; according to the nature of the traffic, divide the data packets entering the network into time-triggered traffic and event-triggered traffic; according to the nature of the traffic, set the time slot size of the time-triggered traffic to not only meet the cyclic forwarding mechanism, but also reserve a part of the bandwidth to provide margin for the scheduling of the event-triggered flow;

[0082] Step S200: Perform global scheduling in advance for time-triggered traffic with known arrival time and traffic size, and determine the injection time and queue. For event-triggered traffic that arrives suddenly, incremental scheduling is adopted according to the real-time remaining resources of the link; the time slots of event-triggered traffic are divided into extremely detailed time slots, and each switch port can only receive one traffic forwarding in each time slot; calculate the scheduling of traffic one by one, and select the appropriate time slot for sending according to the current network status; add the time-triggered traffic to the corresponding queue, follow the CQF scheduling mechanism, and adjust the penetration time according to the time slot resource table; before scheduling new traffic, check the status of the current switch port, and monitor the current reserved bandwidth when injecting traffic;

[0083] Step S300: according to the predetermined time interval and injection time, add the time-triggered traffic to the corresponding queue, follow the CQF scheduling mechanism, according to the mechanism of event-triggered traffic, the biggest feature of time-triggered traffic is the known planned traffic, which can be scheduled in advance, and there is no sudden new traffic; select the appropriate sending time to ensure that important traffic in the network will not be queued in a queue for too long or be discarded due to exceeding the queue size, and arrange the sending time reasonably to stagger the traffic that was originally sent together; dynamically adjust the sending time of time-triggered traffic according to the actual traffic situation;

[0084] Step S400: For the received time-triggered traffic, select a time slot for forwarding according to the pre-scheduled results; for the newly arrived event-triggered traffic, schedule it according to the current global network status (queue resources and reserved bandwidth resources), and select a time slot for forwarding after obtaining the scheduling results; and record the forwarding results and bandwidth usage; continuously monitor the traffic status, bandwidth usage and process, and select a suitable sending time according to the actual traffic situation to ensure that important traffic in the network will not be queued in a queue for too long or exceed the queue size and be discarded; the injection time is the sending time.

[0085] In the above embodiment, step S100 system initialization and traffic classification ensure that the time slots of event-triggered traffic and time-triggered traffic can be effectively isolated and managed; adjust the allocation of internal time slot resources so that different types of traffic can coexist in the network without interfering with each other; ensure that the network can meet the low latency requirements of time-triggered traffic under high load by reserving bandwidth; classify data packets according to traffic characteristics, which is helpful for the implementation of subsequent scheduling strategies, so that different types of traffic can run under their respective scheduling strategies. Significance: Improves the efficiency of the network and the rationality of resource allocation, and ensures the priority processing of time-sensitive data packets; through good resource management and scheduling strategies, the stability and reliability of the system are enhanced. Step S200 distinguishes between time-triggered flow and event-triggered flow: time-triggered flow is known (size, source address, destination address, etc.) and periodic; event-triggered flow is bursty and non-periodic, reducing its uncertainty and improving the success rate of sending; incremental scheduling is defined for event-triggered flow, and a single flow is scheduled according to the real-time status to avoid simultaneous conflicts; only one flow is forwarded in each time slot to ensure the orderliness and stability of processing; the flow is scheduled one by one, and the scheduling decision is based on the current network status to ensure flexibility and adaptability. Significance: Enhances the processing capability of burst traffic and reduces network delay, especially event-triggered flow. Step S300 Traffic injection and status monitoring, adds time-triggered traffic to the queue according to the scheduled time slot, ensures that the traffic will not conflict in the network and meets the delay requirements, and improves the real-time and reliability of data transmission; uses the CQF scheduling mechanism to determine the time to inject into the network according to the time slot resources to ensure the timely transmission of traffic; Significance: Improves the data processing capability of the entire network, ensures the effective injection and optimized transmission of traffic, and achieves the goal of reducing packet loss rate and improving network quality. Step S400 Dynamic scheduling and real-time monitoring, select the appropriate time slot according to the actual network status, effectively forward the event-triggered traffic, and ensure that the traffic is accurately sent from the corresponding queue; continuously monitor the bandwidth usage, optimize and adjust according to the real-time situation, so that the system has adaptive capabilities; dynamically adjust parameters according to the real-time changes of the current traffic, and optimize the scheduling effect of time-triggered traffic and event-triggered traffic. Significance: Improves the network's adaptive capabilities and intelligent management level, can cope with dynamically changing network conditions and traffic conditions, and further enhances the user's experience of network services (for specific principles, please refer to the attached Figure 3 ). Due to the complexity of the actual network environment, there are great differences in the traffic conditions triggered by sudden events. Therefore, this mechanism adopts a dynamic and adaptive reserved bandwidth method to determine the actual application needs. That is, the number of event-triggered traffic that fails to be scheduled will be counted over a period of time. If it exceeds a certain threshold, the reserved bandwidth will be increased. Otherwise, the reserved bandwidth can be reduced accordingly. Figure 3The queue in the sending state of node 0 is used as an example for explanation. In a network without sudden event-triggered traffic, since the time-triggered flow is scheduled in advance, the scheduling result can ensure that there will be no conflict of queue resources or insufficient time resources in the link. Under normal circumstances, the time-triggered flow will be sent to the receiving queue of the next node in one time slot. However, when there is unplanned high-priority traffic in the link, that is, event-triggered traffic, the original time-triggered traffic may not be sent out in the sending time slot or cannot reach the next hop node on time, affecting the transmission of subsequent traffic. This design reserves a part of bandwidth resources for the event-triggered flow in the time slot of each time-triggered flow. When the event-triggered flow arrives, it can be forwarded immediately in the time dimension, that is, it can interrupt other time-sensitive traffic in the same sending time slot space. Because there is reserved bandwidth, as long as the size of the event-triggered flow does not exceed the reserved bandwidth, the original time-triggered flow can still complete all forwarding in one time slot. Therefore, the event-triggered flow that has been incrementally scheduled follows the restrictions of the reserved bandwidth and will not affect the upper bound of the transmission delay of the original time-triggered flow.

[0086] In summary, through the implementation of the above steps, the circular queue forwarding scheduling method for mixed time-sensitive traffic in this embodiment realizes an efficient, intelligent and flexible traffic management system. It not only optimizes the scheduling of time-triggered and event-triggered traffic, and provides good support for diverse application environments; it also emphasizes the importance of efficient use of network resources and low-latency communication, thereby improving overall performance and user satisfaction in actual network environments. The time-triggered time-sensitive traffic in this embodiment still follows the scheduling mechanism of CQF and is scheduled in a queue manner. A portion of the bandwidth is reserved in each time slot for the transmission of event-triggered traffic.

[0087] This embodiment reserves a portion of bandwidth for time-triggered sensitive traffic in the time slot for processing time-triggered sensitive traffic. The reserved bandwidth can ensure that the transmission of the two types of traffic does not affect each other. The adaptability of the reserved bandwidth can meet the needs of event-triggered traffic in different environments; the bandwidth size can be appropriately increased in the scenario where the event-triggered traffic surges, and the reserved bandwidth size can be reduced in the case of less event-triggered traffic.

[0088] Embodiment 4:

[0089] like Figure 4 As shown, based on Example 3, the incremental scheduling strategy in step S200 provided in this embodiment of the present invention includes:

[0090] Step S201: monitor the status of each switch port in the network in real time, determine the type of traffic as event-triggered traffic, and obtain characteristics such as the size and occurrence time of the traffic; before scheduling new traffic, evaluate the current network status, including the status of scheduled traffic, bandwidth usage, and queue length, etc.;

[0091] Step S202: Find an available time slot in the divided time slots, and select a time slot based on the characteristics of the current traffic, the network status, and other traffic already in the queue; when selecting a time slot, consider the scheduling of time-triggered traffic and the set bandwidth reservation; after identifying the time slot, only schedule the event-triggered traffic that has arrived currently; during the traffic scheduling process, adjust the injection time of the traffic;

[0092] Step S203: Take the selected event-triggered traffic from the relevant queue and forward it to the target location according to the scheduling result; for the forwarded traffic, record the forwarding result, including the sending time slot, bandwidth usage and processing delay corresponding to the traffic;

[0093] Step S204: Continuously monitor the traffic status and bandwidth usage in the network, dynamically adjust the scheduling strategy for subsequent traffic based on the actual traffic situation monitored, and use a scheduling algorithm to plan the injection time for sudden event-triggered flows and time-triggered flows because of their known nature.

[0094] In the above embodiment, step S201 real-time monitoring and network status evaluation can quickly identify that the arriving traffic is event-triggered traffic, and can be processed in a targeted manner; by analyzing the current network status, the scheduled traffic, bandwidth usage and queue length are understood, and the basic data support required for decision-making is provided. Significance: Through real-time monitoring and status evaluation, it is possible to quickly adapt to the changing network environment, ensure the timeliness and accuracy of traffic scheduling, ensure that network resources are reasonably used, and reduce the risk of conflicts and delays caused by burst traffic. Step S202 available time slot search and scheduling, find suitable available time slots in the divided time slots, and only schedule traffic on available resources to reduce the probability of conflicts. Significance: Enhance the flexibility of network scheduling, optimize the use of network bandwidth by scientifically selecting time slots and adjusting the traffic injection time in time, and improve the processing efficiency of event-triggered traffic; at the same time, reduce the transmission delay caused by resource competition, and improve the user experience. Step S203 traffic forwarding and result recording, extract the selected event-triggered traffic from the corresponding queue, and forward it to ensure that the traffic can reach the destination smoothly as planned; record the results of traffic forwarding, including sending time slots and bandwidth utilization. Significance: The effective execution of forwarding operations ensures lasting network performance. At the same time, by recording relevant data, it enhances the traceability and analyzability of network traffic performance; it helps engineers evaluate the effectiveness of traffic scheduling. Step S204 continuously monitors and dynamically adjusts, monitors the traffic status and bandwidth usage in the network in real time, and can quickly respond to network changes; according to the actual traffic monitoring situation, dynamically adjusts the traffic scheduling strategy, including improving the incremental scheduling of event-triggered traffic or the injection strategy of time-triggered traffic. Significance: The adaptive capability of the system is realized, so that the network can flexibly adjust the scheduling strategy according to the real-time situation, so as to better operate in burst traffic and changing environments; dynamic adjustment not only improves the efficiency of network use, but also provides users with important performance guarantees and reduces the risk of delay and packet loss.

[0095] In summary, through the implementation of the above steps, the incremental scheduling strategy of this embodiment can effectively provide an efficient and flexible scheduling solution for event-triggered traffic, ensuring that the network still exhibits excellent performance under high load and dynamic changes. It not only improves the utilization efficiency of network resources, but also helps to improve user experience, optimizes the stability and timeliness of overall traffic transmission, and provides reliable technical guarantee for meeting the high demands of users and applications.

[0096] The enhanced CQF scheduling mechanism of this embodiment adopts different scheduling methods for time-triggered traffic and event-triggered traffic. Time-triggered traffic follows the queue-based scheduling method of CQF to adjust the injection time to ensure no congestion and packet loss in the network and low latency. The occurrence time of event-triggered traffic is uncertain and bursty. Therefore, the scheduling method adopts an incremental, one-by-one traffic scheduling method. Unlike joint scheduling, which is based on scheduling all traffic at the same time, incremental scheduling generates a traffic and schedules a traffic, scheduling only one traffic at a time based on the current state; the mechanism of incremental scheduling is also determined by adjusting the time from the generation of burst traffic to the injection into the network. It is assumed that two burst flows fb1 and fb2 have been scheduled in the network at the beginning, and the reserved bandwidth is set to two packets of data. At this time, the third flow fb3 comes. If it is sent out directly without scheduling, fb3 will conflict with fb1 on the port of switch SW1. In addition, if a time slot in slot_t0 is selected, the size of the set reserved bandwidth will be exceeded. Therefore, in order to meet the latency requirement of event-triggered traffic, fb3 selects slot_b4 in slot_t1 for forwarding, which can not only meet the size of the reserved bandwidth but also ensure that no conflict occurs. Refer to Table 1;

[0097] Table 1

[0098]

[0099] Embodiment 5:

[0100] like Figure 5 As shown, based on Embodiment 4, the step S200 provided in the embodiment of the present invention selects a suitable time slot for transmission, including:

[0101] Step S205: assign a unique number to each time slot, monitor the status of each switch port in real time, and promptly discover newly arrived event-triggered traffic; check the available fine time slots according to the bandwidth situation and the detailed time slot division, and find the time slot for sending new traffic. The selected time slot does not conflict with the scheduled traffic;

[0102] Step S206: If the selected time slot conflicts or exceeds the reserved bandwidth, time adjustment is performed in the available time slots to move the injection time to a suitable time slot; confirm that the selected time slot can effectively accommodate the new traffic and record it;

[0103] Step S207: Take the selected event-triggered traffic from the corresponding queue and forward it on the selected time slot; within the selected time slot, send the traffic to the target location as planned; for the forwarded traffic, record detailed forwarding information, including the time slot number used, actual bandwidth usage, and traffic processing delay, etc., to provide a basis for subsequent analysis.

[0104] In the above embodiment, step S205 allocates time slot numbers and checks available time slots, allocates unique numbers to each time slot, facilitates management and identification, and forms a systematic time slot management mechanism; monitors the status of each switch port in real time, promptly discovers newly arrived event-triggered traffic, and ensures that it can quickly respond to changes in network traffic; based on the current bandwidth situation and refined time slot division, dynamically checks and evaluates available time slots to ensure that the selected time slot can accommodate new traffic; before selecting a time slot, checks whether it conflicts with scheduled traffic to ensure the effectiveness of traffic scheduling. Significance: Ensures refined management of the network, improves the flexibility and response speed of traffic scheduling; ensures reasonable allocation of time slots, reduces resource competition, and efficiently guides important traffic to the appropriate location; at the same time, reduces the risk of data packet loss and delay, and improves the overall performance and reliability of the network. Step S206 Time slot adjustment and confirmation: if it is found that the selected time slot has a conflict or exceeds the reserved bandwidth, time adjustment is performed in the available time slot to ensure that the new traffic can be successfully injected; while confirming that the selected time slot can effectively accommodate the new traffic, relevant information is recorded for subsequent scheduling and adjustment. Significance: It emphasizes the flexibility and adaptability of network scheduling. No matter how the traffic changes, it can effectively respond by dynamically adjusting the time slot; it ensures that there will be no traffic loss or delay due to time slot conflicts, and improves the timeliness and reliability of transmission; the recorded confirmation information provides valuable data support for subsequent traffic analysis and optimized scheduling. Step S207 Traffic forwarding and result recording, take out the selected event-triggered traffic from the corresponding queue, and forward it on the selected time slot to ensure that the traffic is sent to the target location smoothly as planned; for the forwarded traffic, the forwarding information of the traffic is recorded in detail, including the time slot number used, the actual bandwidth usage and the delay of traffic processing, etc., to provide data support for further analysis and optimization. Significance: It ensures that the scheduled traffic can be forwarded to the target location quickly and accurately to achieve the purpose of data transmission; at the same time, the detailed forwarding records also provide important data basis for the evaluation of network performance, troubleshooting and improvement of scheduling strategies in the later stage; the combination of effective traffic forwarding and information recording further improves the stability and overall efficiency of the network.

[0105] This embodiment divides the time slot of event-triggered traffic into fine time slots, and schedules the event-triggered traffic flow by flow. The requirement for solving the scheduling time slot is that each switch port can only receive the forwarding of one flow in each time slot, and it does not exceed the reserved bandwidth of the time-sensitive traffic division. Since only one flow is scheduled at a time, the time complexity of the full search solution is a constant with an upper bound. The scheduling time slot obtained based on this criterion can ensure that the event-triggered flow is forwarded without waiting on the link node, and since the reserved bandwidth limitation is taken into account during the solution, the forwarding of the event-triggered traffic will not affect the original delay upper bound of the time-triggered traffic. . Under this mechanism, the queue of the event-triggered traffic only needs to be able to store one maximum event-triggered flow.

[0106] In summary, this embodiment ensures that event-triggered traffic can be transmitted efficiently and reliably through time slot management and dynamic scheduling, thereby maximizing the utilization of network resources, reducing delays and data conflicts, and providing users with a better service experience. The practicality and flexibility in handling burst traffic provide a guarantee for the high-performance operation of the network.

[0107] Embodiment 6:

[0108] like Figure 6 As shown, on the basis of Example 4, the terminal provided by the embodiment of the present invention includes a switch or a network card, etc.; the scheduling result is to generate a time slot corresponding to each flow entering the network, and the obtained result is deployed to the gating list information of the switch or the network card; including:

[0109] The framing module is used to encapsulate the data transmitted by the upper layer to form the frame structure required by the network protocol; in the process of framing, necessary header and tail information is added, such as source address, destination address, frame type and checksum, etc.; through framing, continuous data streams are integrated into frames of a certain length; ensuring that the generated frames comply with the corresponding network protocol specifications;

[0110] The frame parsing module is used to parse the received frames and extract the required information, including extracting the data payload, identifying the sender and receiver, and other protocol-related fields; the frame parsing process verifies the integrity of the data, including determining whether the data is damaged during transmission through checksums or other error detection mechanisms; and passes the parsed data or control information to downstream modules (such as the scheduling module) for further processing and scheduling decisions;

[0111] The scheduling module is used to manage and control the transmission order of data packets. According to the information from the Virtual Output Queue (VOQ) module and the gating information of the gating logic GCL module, it formulates the scheduling strategy for each data packet and generates the corresponding time slot scheduling results; by reading the gating list, it selects the data to be output based on the priority of different queues to realize the priority scheduling of traffic; monitors the occupancy status of the current port to determine which queues' frames can be scheduled; returns the scheduling results to the Virtual Output Queue module to process the corresponding data frames.

[0112] In the above embodiments, the framing module can encapsulate the data transmitted by the upper layer into a frame structure that complies with the network protocol, which is the basis for data transmission from source to destination; ensure that the generated frames follow specific network protocol specifications (such as Ethernet, IP protocol, etc.), which helps to ensure compatibility between interconnected devices; integrate continuous data streams into frames of a certain length, which is convenient for subsequent processing and optimizes the use of network bandwidth. Significance: The design of the framing module enables data to be effectively transmitted in the network, which is the first step to achieve data communication; by adding information such as source address and destination address, it can ensure that data reaches the destination accurately in a complex network, reducing the risk of data loss. At the same time, the data integrity identifier (such as checksum) in the encapsulation process helps to further improve the reliability of the network. The frame parsing module extracts the effective load and other protocol-related information from the received data frame; through error detection mechanisms such as checksum, it ensures that the data is not damaged during transmission; the parsed data and control information are quickly passed to the scheduling module to achieve further processing of the traffic. Significance: The frame parsing module plays an important role in ensuring the correct transmission of data. By effectively cracking the received frames, it lays the foundation for the priority scheduling and forwarding of data flows, ensuring that the system can process real and valid data. At the same time, integrity checking can reduce various problems caused by data corruption and provide necessary feedback information, which is helpful for network maintenance and fault diagnosis. The scheduling module formulates effective scheduling strategies and time slot arrangements based on the information of the virtual output queue (VOQ) and the gating information provided by the gating logic (GCL), ensuring that data packets are reasonably scheduled according to priority and network status; selects the data to be output according to the priority of different queues to ensure that key data can be processed first, thereby balancing the utilization of network resources and dynamically monitoring the occupancy of ports to ensure the real-time and effectiveness of scheduling. Significance: The effective operation of the scheduling module is the key to achieving efficient traffic control. By reasonably configuring time slots and scheduling strategies, it can improve the overall performance of the network, reduce delays and congestion, and ensure that important traffic is given priority under high load conditions. This not only improves the user experience, but also enhances the reliability and stability of the network. In addition, the scheduling results feed back information to the virtual output queue (VOQ) module to form a closed-loop control, which helps to further optimize the scheduling strategy.

[0113] In summary, this embodiment realizes the complete process of data from framing, parsing to scheduling and forwarding, which is specifically manifested in: each module provides guarantee for the efficient transmission and processing of data, ensuring that data can flow quickly and accurately in the network; through traffic priority management, the scheduling module can ensure priority service for important data, improve user experience and satisfaction; through integrity verification and protocol encapsulation, the reliability of the network is improved and the risks in data transmission are reduced; the system can maintain flexibility and scalability when processing complex traffic, and can adapt to the ever-changing network environment.

[0114] Embodiment 7:

[0115] On the basis of Example 6, the scheduling module provided in this embodiment of the present invention includes:

[0116] A virtual output queue submodule, which is used to create an independent queue for each input port and output port pair;

[0117] The ingress queue submodule is a queue in a network switch or router that is used to temporarily store data packets to be processed or forwarded. Data packets arrive from other devices in the network and are stored in the ingress queue. After the corresponding scheduling and forwarding process is completed, they are sent to the next destination.

[0118] The scheduling submodule (Schedule) is used to arrange, organize and control the transmission order and time of data packets in network switches or routers; the scheduling strategy determines which data packets or traffic should be processed first under different conditions;

[0119] The scheduler submodule is used to manage and control the flow of data packets from the input queue to the output queue, and determines the data packets sent to the output link from each input queue at a specific time.

[0120] In the above embodiments, the virtual output queue sub-module (VOQ) creates independent queues between each input port and output port, effectively isolating different traffic flows and avoiding competition between traffic from different sources; by managing multiple virtual output queues, the data flow can be better controlled and scheduled, thereby improving the throughput and efficiency of the entire network system. Significance: The design of the virtual output queue sub-module can effectively reduce conflicts and competition, and improve the fairness and efficiency of data transmission. This mechanism enables the system to allocate bandwidth more reasonably, improve network performance, and at the same time support fair processing of multiple traffic types, which helps to ensure the stability and reliability of data transmission under high load conditions. The ingress queue sub-module provides a buffer space for the data packets to be processed or forwarded, ensuring a smooth transition of data in the processing flow; dynamically manages the received data traffic, and performs effective traffic regulation according to the network bandwidth and status to prevent overflow and packet loss. Significance: The existence of the ingress queue sub-module enables network devices to flexibly respond to different traffic spikes, ensuring that data packets are not immediately discarded when they arrive, providing time and space for scheduling and forwarding, and helping to improve the stability of the network and the effectiveness of streaming data; when the traffic fluctuates, this module can make a smooth transition to ensure the continuity and reliability of the system. The schedule sub-module arranges the transmission order and time of data packets according to the real-time network status, realizing dynamic management of the data flow; determines the data packets that need to be processed preferentially under different conditions according to the preset scheduling strategy, improving the service level of specific traffic. Significance: The design of the schedule sub-module aims to achieve efficient traffic management, effectively allocate priorities under changing network conditions, ensure the priority transmission of important traffic, and thus enhance the user experience; through reasonable scheduling, network congestion and delay can be reduced, and the utilization rate of network resources can be optimized, which plays a crucial role in the improvement of the network. The scheduler sub-module manages the flow of data packets from the ingress queue to the egress queue, ensuring that the appropriate data packets can be sent at a specific moment; according to the current queue status and scheduling strategy, it decides in real time which data packets can be sent to the output link, minimizing delay and packet loss. Significance: As the core of the scheduling module, the scheduler sub-module can ensure the smooth flow of data packets in the network. Through precise scheduling decisions, the performance of the entire network device can be optimized, unnecessary waiting time can be reduced, and the throughput capacity can be enhanced. Dynamically adjusts the flow of data packets and performs detailed traffic control according to the current network usage conditions, ensuring high availability and quality of service.

[0121] In this embodiment, the traffic scheduling time slots obtained by the scheduling model and algorithm are sent to the gating logic GCL module through the AXI bus. The gating logic GCL module is a module that stores gating information, which includes the gating information of each priority queue, that is, whether to open or close, and the duration of the corresponding gating information; the Scheduler scheduling module reads the data in the gating list, and selects which queue's data to output according to the occupancy of the current port, and returns the scheduling result to the virtual output queue (VOQ) module; the main function of the virtual output queue (VOQ) module is to store the basic information of the extracted data frames, and output the basic information situation in the internal cache to the scheduler sub-module (Scheduler) for it to obtain the scheduling result. At the same time, the virtual output queue (VOQ) module will also send the basic information of the corresponding data frames to be output to the subsequent ingress queue sub-module (Ingress Queue) according to the scheduling result of the scheduler; the main function of the ingress queue sub-module (Ingress Queue) is to cache the incoming data frames, and select the data frames of the corresponding queue for forwarding according to the basic information of the scheduled data frames provided by the virtual output queue (VOQ) module. At the same time, after transmitting a frame of data, this module provides a signal indicating that the transmission is completed to the scheduling sub-module (Schedule) so that it can schedule the next data. Thus, through the scheduling part module designed by the end system, the scheduling result injects the corresponding data frames into the network from the sending port according to their corresponding forwarding time slots.

[0122] In summary, in this embodiment, by combining the various sub-modules of the overall scheduling module, it can be seen that a highly efficient and flexible network data processing mechanism is formed through cooperation. By effectively managing data streams and reducing contention, the throughput and efficiency of network protocols can be improved; dynamic scheduling and priority control help accelerate the transmission of important data and reduce latency; it can adapt to a variety of different traffic demands and network loads, ensuring the stability and reliability of the system in a complex network environment. In the face of changing environments and different traffic loads, it can achieve efficient and reliable service delivery.

[0123] Figure 7 The block diagram of an exemplary electronic device suitable for implementing the embodiments of the present invention is shown.

[0124] The electronic device may include a central processing unit / microprocessor / master control chip, etc.; a storage medium, coupled to the central processing unit / microprocessor / master control chip, etc., and storing computer-executable instructions therein for performing the steps of the various methods of the embodiments of the present invention when executed by the processor.

[0125] The central processing unit / microprocessor / master control chip, etc. may include, but are not limited to, for example, one or more processors or microprocessors, etc.

[0126] The storage medium may include, but is not limited to, for example, random access memory (RAM), read-only memory (ROM), flash memory, EPROM memory, EEPROM memory, registers, computer storage media (such as hard disks, floppy disks, solid state drives, removable disks, CD-ROMs, DVD-ROMs, Blu-ray discs, etc.).

[0127] In addition, the electronic device may further include (but is not limited to) a data bus, an input / output bus / external bus / device bus, etc., a display, and input / output devices (such as a keyboard, a mouse, a speaker, etc.).

[0128] The central processing unit / microprocessor / master control chip, etc. may communicate with external devices (etc.) via a wired or wireless network (not shown) through the I / O bus.

[0129] The storage medium may also store at least one computer-executable instruction for performing each function and / or method step in the embodiments described in the present technology when run by the central processing unit / microprocessor / master control chip, etc.

[0130] In one embodiment, the at least one computer-executable instruction may also be compiled into or form a software product, and when one or more computer-executable instructions are run by a processor, each function and / or method step in the embodiments described in the present technology is performed.

[0131] Figure 8 A schematic diagram of a computer-readable storage medium according to an embodiment of the present invention is shown.

[0132] As Figure 8 shown, instructions are stored on a non-transitory computer-readable storage medium, and the instructions are, for example, computer-readable instructions. When the computer-readable instructions are run by a processor, the various methods described above can be executed. The non-transitory computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may, for example, include random access memory (RAM) and / or cache memory, etc. Non-transitory non-volatile memory may, for example, include read-only memory (ROM), hard disks, flash memory, etc. For example, the non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device runs the computer-readable instructions stored on the non-transitory computer-readable storage medium, the various methods described above can be performed.

[0133] In several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of devices or units can be in electrical, mechanical, or other forms.

[0134] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0135] In addition, each functional unit in various embodiments of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0136] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (English full name: Read-Only Memory, English abbreviation: ROM), random access memories (English full name: Random Access Memory, English abbreviation: RAM), magnetic disks, or optical discs and other various media that can store program codes.

[0137] The above, the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of various embodiments of the present invention.

Claims

1. A circular queue forwarding scheduling model for mixed time-sensitive traffic, characterized in that: Include: Delay-sensitive queues, used for event-triggered traffic, handle delay-sensitive traffic; manage packets of event-triggered traffic and perform priority scheduling when the network is congested; A first time-sensitive queue is used to manage the first type of time-triggered traffic, whose packets arrive within a specified time interval and are processed within a predetermined time; by isolating the first time-sensitive queue from the delay-sensitive queue; The second time-sensitive queue is used to cooperate with the first time-sensitive queue to form a pingpong operation for receiving parallel processing, thereby ensuring queue throughput on the basis of controlling queuing delay.

2. The circular queue forwarding scheduling model for mixed time-sensitive traffic according to claim 1, characterized in that: Also includes: A first receiving gate, connected to the first time-sensitive queue, is responsible for controlling whether the first time-sensitive queue belongs to a receiving queue in the current time slot, and allows data to be input into the queue if it is in a receiving state; A second receiving gate, connected to the second time-sensitive queue, is responsible for controlling whether the first time-sensitive queue is in a receiving state in the current time slot, and if in the receiving state, allows data to be input into the queue; A first sending gate is used to determine whether the first time-sensitive queue is in a sending state, and if in the sending state, send the data packet in the queue to a final destination; a second sending gate, used to determine whether the second time-sensitive queue is in a sending state, and if so, to send the data packets in the queue to a final destination; The gate list module is responsible for configuring the switch state of each sending and receiving gate according to the traffic scheduling results calculated by the upper layer. The traffic scheduling results are reflected in the switch time and duration of each sending and receiving gate.

3. The circular queue forwarding scheduling model for mixed time-sensitive traffic according to claim 1, characterized in that: Also includes: Gating logic is used for traffic scheduling and priority management. For event-triggered traffic, the sending time is dynamically adjusted according to the real-time traffic status of the link, and priority is given to it without causing traffic queuing conflicts. For time-triggered traffic, a round-robin forwarding method is adopted, and the sending gate and receiving gate of the first time-sensitive queue and the second time-sensitive queue are mutually exclusive.

4. A circular queue forwarding scheduling method for mixed time-sensitive traffic, characterized in that: The following steps are involved: Create two time slot systems, one for event-triggered traffic and the other for time-triggered traffic, and the two time slots must be in multiples. Set bandwidth reservation for time-triggered traffic time slots in the network to provide deterministic bandwidth resources for event-triggered traffic scheduling, thereby achieving independent scheduling of the two types of traffic. According to the nature of the traffic, the data packets entering the network are divided into time-triggered traffic and event-triggered traffic. In addition to meeting the requirements of the cyclic forwarding mechanism, the time slot size for time-triggered traffic also needs to reserve some bandwidth to provide margin for the scheduling of event-triggered flows. Global scheduling is performed in advance for time-triggered traffic with known arrival time and traffic size to determine the injection time and queue; incremental scheduling is used for event-triggered traffic that arrives suddenly based on the real-time remaining resources of the link; the time slots of event-triggered traffic are divided into extremely detailed time slots, and each switch port can only receive one traffic forwarding in each time slot; Schedule time-triggered traffic one by one, and select appropriate time slots for transmission based on the bandwidth resources reserved in the current network; add time-triggered traffic to the corresponding queue, follow the CQF scheduling mechanism, and adjust the penetration time according to the time slot resource table; Before scheduling new traffic, check the current switch port status and monitor the current reserved bandwidth when injecting traffic; According to the scheduled time interval and injection time, the time-triggered traffic is added to the corresponding queue, following the CQF scheduling mechanism. According to the mechanism of event-triggered traffic, the biggest feature of time-triggered traffic is that the planned traffic is known and scheduled in advance, and there is no sudden new traffic; Select an appropriate sending time to ensure that important traffic in the network will not be queued in a queue for too long or be discarded if it exceeds the queue size. By properly arranging the sending time, the traffic that was originally sent together can be sent at different times. Dynamically adjust the sending time of time-triggered traffic according to actual traffic conditions. For the received time-triggered traffic, select the time slot for forwarding according to the pre-scheduled results. For the newly arrived event-triggered traffic, schedule it according to the current state of the global network, which includes queue resources and reserved bandwidth resources. After obtaining the scheduling results, select the time slot for forwarding; and record the forwarding results and bandwidth usage; continuously monitor the traffic status, bandwidth usage and process, and select the appropriate sending time according to the actual traffic situation to ensure that important traffic in the network will not queue in a queue for too long or exceed the queue size and be discarded; the injection time is the sending time.

5. The method for circular queue forwarding scheduling of mixed time-sensitive traffic according to claim 4, characterized in that: The incremental scheduling strategy uses incremental scheduling, which schedules traffic one by one. It generates one traffic and schedules one traffic at a time, scheduling only one traffic based on the current state. This is determined by adjusting the time from when burst traffic is generated to when it is injected into the network.

6. The method for circular queue forwarding scheduling of mixed time-sensitive traffic according to claim 5, characterized in that: The incremental scheduling strategy specifically includes: Monitor the status of each switch port in the network in real time, detect traffic arrival, determine its type as event-triggered traffic, and obtain the size and occurrence time characteristics of the traffic; Before scheduling new traffic, evaluate the current network status, including the status of scheduled traffic, bandwidth usage, and queue length; Find available time slots in the divided time slots, and select time slots based on the characteristics of the current traffic, network status, and other traffic already in the queue; when selecting time slots, consider the scheduling of time-triggered traffic and the set bandwidth reservation; after identifying the time slot, only schedule the event-triggered traffic that has arrived; during the traffic scheduling process, adjust the injection time of the traffic; Take the selected event-triggered traffic from the relevant queue and forward it to the target location according to the scheduling result; for the forwarded traffic, record its forwarding results, including the sending time slot, bandwidth usage and processing delay information corresponding to the traffic; Continuously monitor the status of traffic and bandwidth usage in the network, dynamically adjust the scheduling strategy for subsequent traffic based on the actual traffic situation monitored, and use scheduling algorithms to plan the injection time for sudden event-triggered flows and time-triggered flows because of their known nature.

7. The method for circular queue forwarding scheduling of mixed time-sensitive traffic according to claim 4, characterized in that: Select the appropriate time slot for transmission, including: Assign a unique number to each time slot, monitor the status of each switch port in real time, and promptly discover newly arrived event-triggered traffic; check the available fine time slots according to the bandwidth situation and detailed time slot division, and find the time slot for sending new traffic. The selected time slot does not conflict with the scheduled traffic; If the selected time slot conflicts or exceeds the reserved bandwidth, make time adjustments in the available time slots and move the injection time to a suitable time slot; confirm that the selected time slot can effectively accommodate the new traffic and record it; The selected event-triggered traffic is taken out from the corresponding queue and forwarded on the selected time slot; within the selected time slot, the traffic is sent to the target location as planned; for the forwarded traffic, detailed forwarding information is recorded, including the time slot number used, actual bandwidth usage, and delay data of traffic processing, to provide a basis for subsequent analysis.

8. A terminal, characterized in that: It includes a switch or a network card, and is characterized in that the scheduling result of the circular queue forwarding scheduling method for mixed time-sensitive traffic of any one of claims 4-7 is obtained, a time slot corresponding to each traffic entering the network is generated, and the obtained result is deployed to the gating list information of the switch or the network card.

9. The terminal according to claim 8, characterized in that Include: The framing module is used to encapsulate the data transmitted by the upper layer to form the frame structure required by the network protocol; in the process of framing, the necessary header and tail information are added; through framing, the continuous data stream is integrated into a frame; Ensure that the generated frames comply with the corresponding network protocol specifications; The frame parsing module is used to parse the received frames and extract the required information, including extracting the data payload, identifying the sender and receiver, and the protocol-related fields; the frame parsing process verifies the integrity of the data, including determining whether the data is damaged during transmission through checksum or error detection mechanisms; and passes the parsed data or control information to the downstream modules for processing and scheduling decisions; The scheduling module is used to manage and control the transmission order of data packets, formulate a scheduling strategy for each data packet based on the information from the virtual output queue module and the gating information from the gating logic module, and generate the corresponding time slot scheduling results; By reading the gating list, the data to be output is selected based on the priorities of different queues to implement priority scheduling of traffic; the occupancy status of the current port is monitored to determine which queues' frames are scheduled; The scheduling result is returned to the virtual output queue module to process the corresponding data frame.

10. The terminal according to claim 9, characterized in that Scheduling module, including: A virtual output queue submodule, which is used to create an independent queue for each input port and output port pair; The in-queue submodule is used to temporarily store the queue of data packets to be processed or forwarded; data packets arrive from devices in the network and are stored in the in-queue, and after the corresponding scheduling and forwarding process is completed, they are sent to the next destination; The scheduling submodule is used to arrange, organize and control the transmission order and time of data packets in a network switch or router; The scheduling policy determines which data packets or flows should be prioritized under different conditions; The scheduler submodule is used to manage and control the flow of data packets from the input queue to the output queue, and determines the data packets sent to the output link from each input queue at a specific time.

Images