Routing and Traffic Joint Scheduling Method and System Based on Link Load Peak

Abstract

The present invention discloses a routing and traffic joint scheduling method and system based on link load peaks, belonging to the research field of time-sensitive network traffic scheduling. The method includes: modeling the target network as a directed graph and defining a time-triggered flow set; combining the idea of load balancing, in a way that frame forwarding does not allow queuing, constructing a gated time slot allocation model for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting; establishing constraints for gated time slot allocation of time-triggered traffic based on joint routing scheduling and frame forwarding without waiting, taking minimizing the link load peak in the network as the scheduling objective, and constructing a joint scheduling problem; calling a joint routing and joint scheduling algorithm based on link load peaks to determine the scheduling scheme for time-triggered traffic. It avoids the transmission waiting caused by queue queuing and resource competition in traditional scheduling methods; at the same time, it reduces the high load of some links caused by the transmission of time-triggered flows.

Description

Technical Field

[0001] The present invention belongs to the field of time-sensitive network traffic scheduling research, and particularly relates to a routing and traffic joint scheduling method and system based on link load peaks. Background Art

[0002] Time-sensitive networks are a series of standards developed by the IEEE 802.1 working group, aiming to provide deterministic and low-latency communication services through Ethernet. Time-sensitive network technology has broad application prospects in fields such as industrial automation, in-vehicle networks, and professional audio and video applications. In these application scenarios, the real-time and reliability requirements of data streams are extremely high. Therefore, how to efficiently schedule data streams in the network to meet these strict delay requirements is an important research topic.

[0003] Existing time-sensitive network scheduling mechanisms mainly include scheduling methods based on priority and time-aware shapers. The scheduling method based on priority assigns different priorities to data streams to ensure that high-priority data streams are transmitted first, so as to reduce the transmission delay of critical data streams. However, the scheduling method based on priority relies on the queuing mechanism of network devices, which easily leads to the accumulation of waiting and delay of data streams and is difficult to achieve wait-free transmission. The scheduling method based on time-aware shapers ensures the transmission of critical data streams within a specific time window through a predefined gating list, thus ensuring the determinism and low latency of data streams. However, existing scheduling methods trigger flows using the shortest path transmission time or isolate link states and routing, without considering the dynamic impact of link states on routing, resulting in excessive load on some links in the network and affecting the transmission of other types of traffic.

[0004] Current research work does not consider the wait-free transmission of time-triggered traffic and link load together, and the transmission of time-triggered traffic will inevitably lead to changes in link load. Therefore, a routing and traffic joint scheduling method based on link load peaks is proposed. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a routing and traffic joint scheduling method and system based on link load peaks, which solves the problems in the prior art.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A routing and traffic joint scheduling method based on link load peaks includes the following steps:

[0008] Model the target network as a directed graph and define the time-triggered flow set;

[0009] Based on the time-triggered flow set, combined with the idea of load balancing, and in the manner that frame forwarding does not allow queuing, from the perspective of traffic planning, a gated time slot allocation model for time-triggered traffic based on joint routing scheduling and frame-forwarding without waiting is constructed;

[0010] Establish the gated time slot allocation constraints for time-triggered traffic based on joint routing scheduling and frame-forwarding without waiting, and based on the gated time slot allocation model, with minimizing the peak link load in the network as the scheduling objective, construct a joint scheduling problem;

[0011] Based on the joint scheduling problem, call the joint routing and joint scheduling algorithm based on the peak link load to determine the scheduling scheme for time-triggered traffic.

[0012] Furthermore, the directed graph is ; represents the set of network nodes, which consists of the set of terminal system nodes and the set of switch nodes ; represents the set of links between adjacent network nodes, and represent the bidirectional link between node and node ;

[0013] The time-triggered flow set is , and any time-triggered flow is characterized by < , , , , , >, representing the source node, destination node, size, period, generation time, and deadline respectively.

[0014] Furthermore, the gated time slot allocation model for time-triggered traffic based on joint routing scheduling and frame-forwarding without waiting is as follows:

[0015]

[0016]

[0017] Among them, represents the end-to-end delay of the time-triggered flow , represents the transmission path selected by the route for the time-triggered flow , represents the propagation delay of the time-triggered flow , represents the time-triggered flow Processing delay Indicates the clock synchronization delay Indicates the time-triggered flow Transmission delay Indicates the bandwidth Indicates the time-triggered flow Size

[0018] Furthermore, the gated slot allocation constraint for the time-triggered traffic in the joint routing scheduling and frame forwarding without waiting manner includes: frame constraint, frame transmission constraint, end-to-end constraint, and link constraint

[0019] The frame constraint is

[0020]

[0021] In the formula Indicates Any continuous link in Indicates On the link Transmission start time

[0022] The frame transmission constraint is

[0023]

[0024] Among them Indicates the previous link in the continuous link Indicates the next link in the continuous link Indicates On the link Transmission start time Indicates On the link Transmission start time

[0025] The end-to-end constraint is

[0026]

[0027] Among them Indicates the deadline of the time-triggered flow Deadline

[0028] The link constraint is

[0029]

[0030] Among them Indicates the hyperperiod Indicates the time-triggered flow Index value of the number of transmissions of the time-triggered flow under the hyperperiod Indicates the time-triggered flow Index value of the number of transmissions under the supercycle.

[0031] Furthermore, the joint scheduling problem is as follows:

[0032]

[0033]

[0034]

[0035] Among them, represents the peak link load, represents the link load, represents the time-triggered flow on the link transmission delay. If the time-triggered flow does not pass through , then this value is 0.

[0036] Furthermore, the steps to determine the time-triggered traffic scheduling scheme include:

[0037] S401, input the network topology , the time-triggered flow set ;

[0038] S402, define the link time slot allocation table, link load table, transmission start time table, and routing entry table, and their initial values are all empty sets;

[0039] S403, traverse each time-triggered flow in the time-triggered flow set , and use depth-first traversal to calculate the path set of the time-triggered flow ;

[0040] S404, use the least common multiple method to calculate the supercycle of the time-triggered flow set ;

[0041] S405, sort the flows in the time-triggered flow set in non-decreasing order according to the supercycle size;

[0042] S406, traverse each time-triggered flow in the time-triggered flow set . If the traversal of the flow set is completed, enter S416; otherwise, enter S407;

[0043] S407, traverse the path set of the time-triggered flow , and find the path set with the smallest peak link load by comparing the peak link loads of each path, and sort them in non-decreasing order of path length;

[0044] S408. Traverse each path in the path set with the minimum peak link load. If the path traversal is completed, go to S412; otherwise, go to S409.

[0045] S409. Calculate the end-to-end delay of the time-triggered flow in the no-waiting mode with as the transmission path. If exceeds the deadline of the time-triggered flow , then go to S408 for the next loop; otherwise, go to S410.

[0046] S410. Call the earliest transmission start time calculation algorithm to calculate the transmission start time of the time-triggered flow in the no-waiting mode with as the transmission path; if the earliest transmission start time is found, i.e., , then go to S411; otherwise, go to S408 for the next loop.

[0047] S411. Update the transmission start time and the optimal path of the time-triggered flow . If one is found, terminate the path traversal and go to S412.

[0048] S412. If , go to S413; otherwise, go to S415.

[0049] S413. Update the link time slot allocation table hop by hop and cycle by cycle, re-sort the allocated time slots on the affected links in ascending order of the transmission start time, and update the link load table. Then go to S414.

[0050] S414. Update the transmission start time table and the routing entry table, and go to S406 for the next loop.

[0051] S415. Clear the link time slot allocation table, the transmission start time table, and the routing entry table, and go to S416.

[0052] S416. Return the link time slot allocation table, the transmission start time table, and the routing entry table.

[0053] The routing and traffic joint scheduling system based on the peak link load includes:

[0054] Flow set construction module: Model the target network as a directed graph and define the time-triggered flow set.

[0055] Time slot allocation model construction module: Based on the time-triggered flow set, combined with the idea of load balancing, in a way that frame forwarding does not allow queuing, starting from the perspective of traffic planning, construct a gated time slot allocation model for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting;

[0056] Scheduling problem construction module: Establish the gated time slot allocation constraints for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting, and based on the gated time slot allocation model, construct a joint scheduling problem with the goal of minimizing the peak link load in the network;

[0057] And, scheduling scheme determination module: Based on the joint scheduling problem, call the joint routing and joint scheduling algorithm based on the peak link load to determine the scheduling scheme for time-triggered traffic.

[0058] A computer storage medium stores a readable program that, when running, can execute the above-mentioned routing and traffic joint scheduling method based on the peak link load.

[0059] An electronic device includes: a processor, a memory, a communication interface, and a communication bus. The processor, the memory, and the communication interface complete communication with each other through the communication bus;

[0060] The memory is used to store at least one executable instruction, and the executable instruction causes the processor to execute the operations corresponding to the above-mentioned routing and traffic joint scheduling method based on the peak link load.

[0061] A computer program product includes computer instructions that instruct a computing device to execute the operations corresponding to the above-mentioned routing and traffic joint scheduling method based on the peak link load.

[0062] Advantages of the present invention:

[0063] Aiming at the scheduling problem of time-triggered traffic in a time-sensitive network, the present invention provides a routing and traffic joint scheduling method based on the peak link load in a time-sensitive network. With the goal of minimizing the peak link load in the network, it realizes true wait-free transmission, avoids transmission waiting caused by queue queuing and resource competition in traditional scheduling methods, thereby ensuring the bounded delay and jitter of time-triggered traffic; at the same time, it reduces the high load of some links caused by the transmission of time-triggered flows and reduces the impact on the transmission of other types of traffic. Description of the Drawings

[0064] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0065] Figure 1 is the flowchart of the routing and traffic joint scheduling method based on the link load peak of the present invention;

[0066] Figure 2 is the flowchart of the joint scheduling algorithm of the present invention;

[0067] Figure 3 is the flowchart of calculating the earliest transmission start time of the present invention;

[0068] Figure 4 is the network topology diagram of the joint scheduling method in Embodiment 2 of the present invention;

[0069] Figure 5 is the time slot allocation instance diagram in the case of fixed routing scheduling and no-waiting mode in Embodiment 2 of the present invention;

[0070] Figure 6 is the time slot allocation instance diagram in the case of joint routing scheduling and no-waiting mode in Embodiment 2 of the present invention;

[0071] Figure 7 is the network topology scenario diagram of the joint scheduling method in Embodiment 3 of the present invention;

[0072] Figure 8 is the link load peak comparison diagram of the joint scheduling method in Embodiment 3 of the present invention;

[0073] Figure 9 is the average scheduling rate comparison diagram of the joint scheduling method in Embodiment 3 of the present invention;

[0074] Figure 10 is the execution time comparison diagram of the joint scheduling method in Embodiment 3 of the present invention. Detailed implementation manners

[0075] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0076] Embodiment 1

[0077] As Figure 1As shown in the figure, in a time-sensitive network, a routing and traffic joint scheduling method based on the peak link load includes the following steps:

[0078] S1. Model the target network as a directed graph and define the time-triggered flow set ;

[0079] Model the target network as a directed graph ; represents the set of network nodes, which consists of the set of terminal system nodes and the set of switch nodes that comply with the IEEE 802.1Qbv standard Then there is ; represents the set of links between adjacent network nodes, and represents the node and the node The bidirectional link between;

[0080] For the time-triggered flow set in the scheduling task, any time-triggered flow is characterized by the tuple < , , , , , >, which represent the source node, destination node, size, period, generation time, and deadline respectively. The time-sensitive network operates at the second layer of the network, and time-triggered flows that exceed the maximum transmission unit size are split into a set of frames for continuous transmission.

[0081] S2. Based on the time-triggered flow set defined in S1, combined with the idea of load balancing, from the perspective of traffic planning, construct a gating time slot allocation model for time-triggered traffic in the way that frame forwarding does not allow queuing;

[0082] The end-to-end delay of a frame is defined as the time elapsed from the release of the frame at the source node to the reception of the frame at the destination node. This delay consists of four parts: propagation delay, processing delay, queuing delay, and transmission delay. The transmission delay of the time-triggered flow can be expressed as:

[0083] (1)

[0084] where represents the bandwidth, and the end-to-end delay of the time-triggered flow can be expressed as:

[0085] (2)

[0086] Wherein, represents the transmission path selected for the time-triggered flow ; represents the propagation delay of the time-triggered flow ; represents the processing delay of the time-triggered flow ; represents the clock synchronization delay represents the link on represents the queuing delay of the time-triggered flow on ;

[0087] The no-wait mode requires continuous frame transmission along the path, that is, queuing is not allowed during frame forwarding, and the queuing delay is zero. Then, the end-to-end delay of the time-triggered flow in the no-wait mode is:

[0088] (3)

[0089] S3. Establish the gating slot allocation constraint for the time-triggered traffic based on the joint routing scheduling and frame forwarding no-wait mode, and based on the gating slot allocation model constructed in S2, with minimizing the peak value of the link load in the network as the scheduling objective, and formalize the joint scheduling problem;

[0090] Based on the gating slot allocation constraint for the time-triggered traffic under the joint routing scheduling and frame forwarding no-wait mode, including: frame constraint, frame transmission constraint, end-to-end constraint, link constraint;

[0091] 1) Establish the frame constraint

[0092] The frame constraint requires that the transmission start time of the frame on the flowing-through link is non-negative, and it must be ensured that the transmission is completed within its period;

[0093] (4)

[0094] In the formula, represents any consecutive links in represents the transmission start time of on the link

[0095] 2) Establish the frame transmission constraint

[0096] The frame transmission constraint requires that the transmission start time of the same frame on the successor link must be greater than or equal to the transmission completion time on the predecessor link;

[0097] (5)

[0098] in, represents the previous link in a continuous chain of links, represents the next link in the continuous link, Represents a time-triggered flow In the link The transmission start time on Represents a time-triggered flow In the link The transmission start time on ;

[0099] 3) Establish end-to-end constraints

[0100] The end-to-end constraint requires that the time interval between the time when any frame arrives at the receiving end and the time when it is sent from the sending end must be less than or equal to the deadline. ;

[0101] (6)

[0102] in, Represents a time-triggered flow Deadline:

[0103] 4) Establish link constraints

[0104] Link constraints require that any two frames on the same link trigger the flow and There is no overlap in time. For a super period, this constraint can be converted into any two frames on the same link, where the transmission start time of one frame must be greater than or equal to the transmission end time of the other frame;

[0105] (7)

[0106] in, represents a super cycle, Represents a time-triggered flow The index value of the number of transmissions in the super period, Represents a time-triggered flow The index value of the number of transmissions in the super period;

[0107] Link Load It refers to the ratio of the sum of the transmission times of all time-triggered flows passing through a single link in one supercycle to the supercycle, which can be expressed as:

[0108] (8)

[0109] in, Represents a time-triggered flow The transmission delay on the link , if the time-triggered flow does not pass through the link , then this value is 0;

[0110] The peak link load refers to the maximum value of the loads of all links in the network under one hypercycle, and can be expressed as:

[0111] (9)

[0112] Based on the joint constraints of equations (4), (5), (6), and (7), with minimizing the peak link load in the network as the scheduling objective, the formalized joint scheduling problem is expressed as:

[0113] (10)

[0114] S4. Based on the joint scheduling problem formed in S3, call the joint routing and joint scheduling algorithm based on the peak link load to determine the scheduling scheme for the time-triggered traffic;

[0115] As Figure 2 shown, the specific steps to call the joint routing and joint scheduling algorithm based on the peak link load to determine the scheduling scheme for the time-triggered traffic are as follows:

[0116] S401. Input the network topology , the time-triggered flow set ;

[0117] S402. Define the link time slot allocation table, with the initial value being an empty set, used to record the allocated time slots of each link; define the link load table, with the initial value being an empty set, used to record the load of each link; define the transmission start time table, with the initial value being an empty set, used to record the transmission start time of the scheduled flows; define the routing entry table, with the initial value being an empty set, used to record the optimal paths of the scheduled flows;

[0118] S403. Traverse each time-triggered flow in the time-triggered flow set , and use depth-first traversal to calculate the simple path set of the time-triggered flow ;

[0119] S404. Use the least common multiple method to calculate the hypercycle of the time-triggered flow set ;

[0120] S405. Sort the flows in the time-triggered flow set in non-decreasing order according to the hypercycle size, and the hypercycle size can be expressed as ;

[0121] S406, Traverse each time-triggered flow in , if the traversal of the flow set is completed, enter S416, otherwise enter S407;

[0122] S407, Traverse the set of simple paths of the time-triggered flow , find the set of paths with the minimum link load peak by comparing the link load peaks of each path , and sort them in non-decreasing order of path length;

[0123] S408, Traverse each path in the set of paths with the minimum link load peak in , if the path traversal is completed, enter S412, otherwise enter S409;

[0124] S409, Calculate the end-to-end delay of the time-triggered flow when using as the transmission path in the non-waiting mode , if exceeds the deadline of the time-triggered flow , then enter S408 for the next loop, otherwise enter S410;

[0125] S410, Call the earliest transmission start time calculation algorithm to calculate the transmission start time of the time-triggered flow when using as the transmission path in the non-waiting mode ; if the earliest transmission start time is found, that is , then enter S411; otherwise enter S408 for the next loop;

[0126] S411, Update the transmission start time and the optimal path of the time-triggered flow , terminate the path traversal when one is found, and enter S412;

[0127] S412, If , enter S413, otherwise enter S415;

[0128] S413, Update the link time slot allocation table hop by hop and cycle by cycle, and re-sort the allocated time slots on the affected links in ascending order of transmission start time to improve the detection efficiency of available time slots on the subsequent links, and at the same time update the link load table, then enter S414;

[0129] S414. Update the transmission start time table and the routing entry table, and enter S406 for the next loop;

[0130] S415. Clear the link time slot allocation table, the transmission start time table, and the routing entry table, and enter S416;

[0131] S416. Return the link time slot allocation table, the transmission start time table, and the routing entry table.

[0132] As Figure 3 shown, in S410, the earliest transmission start time calculation algorithm specifically includes the following steps:

[0133] Step 1. Input the time-triggered flow , the transmission path ;

[0134] Step 2. Define the time interval as , where represents the latest transmission start time of the time-triggered flow in the no-wait mode; represents the transmission end time of the previous hop of the time-triggered flow in the current calculation during the time interval traversal; define the time slot interval as , where represents the start time of the time slot interval, represents the end time of the time slot interval; represents the index value of the number of transmissions of the time-triggered flow in the supercycle;

[0135] Step 3. Calculate the end-to-end delay of the time-triggered flow in the no-wait mode with as the transmission path;

[0136] Step 4. Traverse the moment values in the time interval . To ensure that the flow is transmitted within its period, the frame constraint must be satisfied, that is, , otherwise let , and enter Step 10;

[0137] Step 5. Let be the transmission start time of the next hop of the time-triggered flow in the current calculation to ensure that the transmission start time of the same frame on the subsequent link must be equal to the transmission completion time on the previous link, and the frame transmission constraint must be satisfied;

[0138] Step 6. Traverse each link in . If the link traversal is completed, enter Step 10, otherwise enter Step 7;

[0139] Step 7, update the time slot interval to , and the transmission start time of the next hop after update is ;

[0140] Step 8, traverse the time-triggered flow In each cycle of the supercycle, if the cycle traversal is completed, go to Step 6 for the next loop, otherwise go to Step 9;

[0141] Step 9, update the time slot interval to , and judge whether the time slot interval is occupied link by link and cycle by cycle. If it is occupied, let go to Step 4 for the next loop, otherwise go to Step 8 for the next loop;

[0142] Step 10, return the earliest transmission start time .

[0143] Based on a similar inventive concept, an embodiment of the present invention further provides a computer storage medium storing a readable program which, when running, can execute the above-mentioned routing and traffic joint scheduling method based on the link load peak.

[0144] Based on a similar inventive concept, an embodiment of the present invention provides an electronic device, including: a processor, a memory, a communication interface, and a communication bus. The processor, the memory, and the communication interface complete communication with each other through the communication bus;

[0145] The memory is used to store at least one executable instruction, and the executable instruction causes the processor to execute the operations corresponding to the above-mentioned routing and traffic joint scheduling method based on the link load peak.

[0146] Based on a similar inventive concept, an embodiment of the present invention further provides a computer program product including computer instructions, and the computer instructions direct a computing device to execute the operations corresponding to the above-mentioned routing and traffic joint scheduling method based on the link load peak.

[0147] Embodiment 2

[0148] To verify and illustrate the technical effects adopted in the routing and traffic joint scheduling method based on the link load peak of the present invention, this embodiment separately tests the fixed routing scheduling and the time slot allocation instance in the no-waiting mode and the joint routing scheduling and the time slot allocation instance in the no-waiting mode in the present invention.

[0149] The network topology is as Figure 4 shown. Assume that the network bandwidth is 1 Gbps, the propagation delay, the processing delay, and the clock synchronization delay are all zero. The scheduling task consists of 4 time-triggered flows, and the time-triggered flow The attribute is <source node , destination node , 250B, 100 μs, 0, 100 μs >, time-triggered flow The attribute is <source node , destination node , 250B, 150 μs, 0, 150 μs >, time-triggered flow The attribute is <source node , destination node , 250B, 100 μs, 0, 100 μs >, time-triggered flow The attribute is <source node , destination node , 250B, 150 μs, 0, 150 μs >, scheduled one by one according to the flow subscript order. Let denote the supercycle, then there is = 300 μs, the sizes of all 4 time-triggered flows are 250B. According to formula (1), the transmission delay of all 4 flows is 2 μs.

[0150] Figure 5 Shows an example of time slot allocation in fixed routing scheduling and non-waiting mode. Since the shortest path transmission is adopted, the transmission paths of flows with the same source node and destination node are the same. Let denote the path , denote the path , denote the path , denote the path , denote the path , denote the path , assuming that from the source node to the destination node select as the transmission path, from the source node to the destination node select as the transmission path. After the time slot allocation of the time-triggered flows and , since the time slot interval with a start time of 4 μs and an end time of 6 μs on has been allocated to the time-triggered flow , in non-waiting mode, the transmission start time of the flow time-triggered flow on is postponed by 2 μs, which in turn causes the time-triggered flow The earliest transmission start time is postponed by 2 μs. The time-triggered flow Similarly. In summary, after the scheduling of 4 time-triggered flows is completed, at this time in the network the link load is the highest, reaching 4.

[0151] Figure 6 Shows an example of time slot allocation under joint routing scheduling and the no-waiting method. By using the peak link load as the condition for load balancing, without considering the link load, the time-triggered flow has 3 paths to choose from, which are respectively 、 and , at this time the link load of each link is 0, so these 3 paths are all the paths with the minimum peak link load of the time-triggered flow , and are sorted in non-decreasing order of path length. Taking as the transmission path, the earliest transmission start time is calculated first and it is found that it can be scheduled. At this time the earliest transmission start times are all 0 μs, so the optimal path of the flow is , and the link time slot allocation table is updated with the earliest transmission start time calculated from the optimal path. Without considering the link load, the time-triggered flow has 3 paths to choose from, which are respectively 、 and , at this time the cumulative link load is 1, which is the maximum value of the load of 、 、 、 、 、 as well as these 7 links (all the links that make up 、 and ), so there are 2 paths with the minimum link load for the time-triggered flow , which are and . Since is shorter than , taking as the transmission path, the earliest transmission start time is calculated first and it is found that it can be scheduled. At this time the earliest transmission start time is 0 μs, so the optimal path of the time-triggered flow is , and the link time slot allocation table is updated with the earliest transmission start time calculated from the optimal path. Without considering the link load, the time-triggered flow has 3 paths to choose from, which are respectively 、 and , at this time 、 、 The cumulative link loads of are 1 respectively, and it is 、 、 、 、 、 and the maximum value of the link loads of these 7 links. Therefore, these 3 paths are all time-triggered flows with the minimum link load peak of the path, sorted in non-decreasing order of path length. Taking as the transmission path, the earliest transmission start time is calculated first, and it is found that it can be scheduled. At this time The earliest transmission start time of is 1 μs. Therefore, the optimal path of the time-triggered flow is , and the link time slot allocation table is updated with the earliest transmission start time calculated from the optimal path. Without considering the link load, the time-triggered flow has 3 paths to choose from, which are 、 and , at this time 、 The cumulative link loads of are 2 respectively, and it is 、 、 、 、 、 and the maximum value of the link loads of these 7 links. Therefore, there are 2 paths with the minimum link load for the time-triggered flow , which are divided into and . Since is shorter than , taking as the transmission path, the earliest transmission start time is calculated first, and it is found that it can be scheduled. At this time The earliest transmission start time of is 1 μs. Therefore, the optimal path of the time-triggered flow is , and the link time slot allocation table is updated with the earliest transmission start time calculated from the optimal path. To sum up, after 4 flows are scheduled, at this time in the network 、 、 、 have the highest link loads, all reaching 2, which is 50% lower than the fixed routing. By comparing Figure 5 and Figure 6The allocated time slots indicate that the time-triggered flow time slot allocation model based on joint routing scheduling and the wait-free method reduces the link link load, that is, reduces the peak value of the link load in the network.

[0152] Embodiment 3

[0153] In this embodiment, the technical effect of the joint scheduling method of the present invention is verified and explained by comparing and testing the joint scheduling method of the present invention with two other methods.

[0154] In this embodiment, the two scheduling methods used for comparison with the method of the present invention are respectively:

[0155] (1) Shortest path scheduling method: Use the shortest path to replace the routing part of the method of the present invention.

[0156] (2) Equal-cost multi-path scheduling method: Use the equal-cost multi-path method to replace the routing part of the method of the present invention. The equal-cost multi-path method has 1 or more shortest paths according to different network topologies, and randomly selects 1 path when selecting a path.

[0157] The network topology scenario diagram in this embodiment is as Figure 7 shown, a mesh network topology composed of 10 time-sensitive network switches, and each switch is connected to 1 terminal device.

[0158] Parameter settings: The link bandwidth is uniformly set to 1 Gbps, and the processing delay of the time-sensitive network switch is set to zero. Set the time-triggered flow template shown in Table 1 in the flow set data generator and generate multiple flow sets with different numbers [50, 500]. Each flow in each flow set accounts for 33.33%, and each flow set contains 20 groups of scheduling tasks. Take the average data as the final result. All simulations are run on a Ubuntu 22.04 machine with an 8-core 3.6 GHz Intel Core i7-7700 CPU and 8 GB of memory.

[0159] Table 1 Time-triggered flow template

[0160]

[0161] The reduction rate of the link load peak is as Figure 8 shown. It can be seen that: compared with the other two scheduling methods, the routing and traffic joint scheduling method based on the link load peak of the present invention can effectively reduce the link load peak in the network, and is respectively 45.41% and 10.88% lower than the link load peaks of the shortest path scheduling method and the equal-cost multi-path scheduling method.

[0162] The scheduling rate is as Figure 9As shown in the figure, it can be seen that the routing and traffic joint scheduling method based on the peak link load is superior to the other two scheduling methods in terms of schedulability, with the average scheduling rate increased by 41.25% and 47.49% respectively compared with the shortest path scheduling method and the equal cost multi-path scheduling method.

[0163] The execution time is as Figure 10 As shown in the figure, it can be seen that the routing and traffic joint scheduling method based on the peak link load has almost the same execution time as the other two scheduling methods.

[0164] Embodiment 4

[0165] Based on the routing and traffic joint scheduling method based on the peak link load mentioned in Embodiment 1, in this embodiment, a routing and traffic joint scheduling system based on the peak link load is proposed, which specifically includes:

[0166] Flow set construction module: Model the target network as a directed graph and define the time-triggered flow set;

[0167] Time slot allocation model construction module: Based on the time-triggered flow set, combined with the idea of load balancing, starting from the perspective of traffic planning, construct a gated time slot allocation model for time-triggered traffic under the joint routing scheduling and frame forwarding without waiting mode;

[0168] Scheduling problem construction module: Establish the gated time slot allocation constraints for time-triggered traffic under the joint routing scheduling and frame forwarding without waiting mode, and based on the time slot allocation model, formalize the joint scheduling problem with the goal of minimizing the peak link load in the network;

[0169] And, a scheduling scheme determination module: Based on the joint scheduling problem, call the joint routing and joint scheduling algorithm based on the peak link load to determine the scheduling scheme for time-triggered traffic.

[0170] The method of the present invention can be implemented in hardware, firmware, or be implemented as software or computer code that can be stored in a recording medium (such as a CDROM, RAM, floppy disk, hard disk, or magneto-optical disk), or be implemented as computer code that is originally stored in a remote recording medium or a non-transitory machine-readable medium and downloaded through a network and will be stored in a local recording medium, so that the method described herein can be stored on such a software process on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component (such as RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the method shown herein, the execution of the code converts the general-purpose computer into a dedicated computer for executing the method shown herein.

[0171] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and what is described in the above embodiments and the specification is only to illustrate the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed.

Claims

1. A routing and traffic joint scheduling method based on the peak value of link load, characterized in that It includes the following steps: Model the target network as a directed graph and define the time-triggered flow set; Based on the time-triggered flow set, combined with the idea of load balancing, and in a way that frame forwarding does not allow queuing, from the perspective of traffic planning, construct a gated time slot allocation model for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting; Establish the gated time slot allocation constraints for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting, and based on the gated time slot allocation model, construct a joint scheduling problem with minimizing the peak link load in the network as the scheduling objective; Based on the joint scheduling problem, call the joint routing and joint scheduling algorithm based on the peak link load to determine the scheduling scheme for time-triggered traffic; The gated time slot allocation model for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting is: Among them, represents the end-to-end delay of the time-triggered flow , represents the transmission path selected for the time-triggered flow , represents the propagation delay of the time-triggered flow , represents the processing delay of the time-triggered flow , represents the clock synchronization delay; represents the transmission delay of the time-triggered flow , represents the bandwidth, represents the size of the time-triggered flow . The gated time slot allocation constraints for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting include: frame constraints, frame transmission constraints, end-to-end constraints, and link constraints; The frame constraints are: In the formula, represents any arbitrary continuous link in represents the start time of transmission on the link ; is the period of the time-triggered flow . The frame transmission constraints are: Among them, represents the previous link in the continuous link, represents the subsequent link in the continuous link, represents the start time of transmission on link ; represents the start time of transmission on link ; The end-to-end constraints are: Among them, represents the deadline of the time-triggered flow ; The link constraints are: Among them, represents a hypercycle, represents a time-triggered flow which is the index value of the number of transmissions under the hypercycle, represents a time-triggered flow which is the index value of the number of transmissions under the hypercycle, represents the set of links between adjacent network nodes; is the period of the time-triggered flow ; The steps to determine the time-triggered traffic scheduling scheme include: S401, Input network topology , time-triggered flow set ; S402, Define the link time slot allocation table, link load table, transmission start time table, and routing entry table, and their initial values are all empty sets; S403, traverse each time-triggered flow in the time-triggered flow set and use depth-first traversal to calculate the path set of the time-triggered flow ; ​ S404, calculate the hyperperiod of the time-triggered flow set using the least common multiple method of ; S405, sort the flows in the time-triggered flow set in non-decreasing order according to the supercycle size; S406, Traverse each time-triggered flow in the time-triggered flow set If the traversal of the flow set is completed, enter S416; otherwise, enter S407 for each time-triggered flow in the set S407, Traverse the path set of the time-triggered flow , and by comparing the link load peaks of each path, find the path set with the smallest link load peak, and sort it in non-decreasing order of path length; S408. Traverse each path in the path set with the minimum peak link load. If the path traversal is completed, go to S412; otherwise, go to S409. S409, Calculate the time-triggered flow In the no-wait mode The end-to-end delay when is used as the transmission path , if exceeds the deadline of the time-triggered flow , then enter S408 for the next loop, otherwise enter S410; S410, call the earliest transmission start time calculation algorithm to calculate the time-triggered stream In the non-waiting mode, take as the transmission start time when it is used as the transmission path ; if the earliest transmission start time is found, that is , then enter S411; otherwise, enter S408 for the next loop; S411, Update Time Triggered Flow of the transmission start time and the optimal path , terminate the path traversal once one is found, and enter S412; S412, if , go to S413, otherwise go to S415; S413, Update the link time slot allocation table hop by hop and cycle by cycle, re-sort the allocated time slots on the affected links in ascending order of transmission start time, and update the link load table, then enter S414; S414, Update the transmission start time table and routing entry table, and enter S406 for the next cycle; S415, Clear the link time slot allocation table, transmission start time table, and routing entry table, and enter S416; S416, Return the link time slot allocation table, transmission start time table, and routing entry table.

2. The routing and traffic joint scheduling method based on the peak value of link load according to claim 1, characterized in that The directed graph is ; which represents a set of network nodes, consisting of the set of terminal system nodes and the set of switch nodes ; which represents a set of links between adjacent network nodes, and represents the two-way link between node and node . The time-triggered flow set is , and any time-triggered flow is represented by < , , , , , , representing the source node, destination node, size, period, generation time, and deadline respectively.

3. The routing and traffic joint scheduling method based on the peak value of link load according to claim 1, wherein The joint scheduling problem is: Among them, represents the peak link load, represents the link load, represents the time-triggered flow on the link transmission delay, if the time-triggered flow does not pass through , then the value is 0.

4. A routing and traffic joint scheduling system based on the peak value of link load, characterized in that It includes: Flow set construction module: Model the target network as a directed graph and define the time-triggered flow set; Time slot allocation model construction module: Based on the time-triggered flow set, combined with the idea of load balancing, and in a way that frame forwarding does not allow queuing, from the perspective of traffic planning, construct a gated time slot allocation model for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting; Scheduling problem construction module: Establish the gated time slot allocation constraints for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting, and based on the gated time slot allocation model, construct a joint scheduling problem with minimizing the peak link load in the network as the scheduling objective; And, scheduling scheme determination module: Based on the joint scheduling problem, call the joint routing and joint scheduling algorithm based on the peak link load to determine the scheduling scheme for time-triggered traffic; The gated time slot allocation model for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting is: Among them, represents the end-to-end delay of the time-triggered flow , represents the transmission path selected by the time-triggered flow as the route, represents the propagation delay of the time-triggered flow , represents the processing delay of the time-triggered flow , represents the clock synchronization delay; represents the transmission delay of the time-triggered flow , represents the bandwidth, represents the size of the time-triggered flow . The gated time slot allocation constraints for time-triggered traffic based on joint routing scheduling and frame forwarding without waiting include: frame constraints, frame transmission constraints, end-to-end constraints, and link constraints; The frame constraints are: In the formula, represents any arbitrary continuous link in represents the start time of transmission on link ; is the period of the time-triggered flow . The frame transmission constraints are: Among them, represents the previous link in the continuous link, represents the next link in the continuous link, represents the start time of transmission on the link ; represents the start time of transmission on the link ; The end-to-end constraint is as follows: Among them, represents the deadline of the time-triggered flow; The link constraint is as follows: Among them, represents a hypercycle, represents a time-triggered flow which is the index value of the number of transmissions under the hypercycle, represents a time-triggered flow which is the index value of the number of transmissions under the hypercycle, represents the set of links between adjacent network nodes; is the period of the time-triggered flow ; The steps for determining the time-triggered traffic scheduling scheme include: S401, Input network topology , time-triggered flow set ; S402, define a link time slot allocation table, a link load table, a transmission start time table, and a routing entry table, and their initial values are all empty sets; S403, traverse each time-triggered flow in the time-triggered flow set and calculate the path set of the time-triggered flow using depth-first traversal for each time-triggered flow in the time-triggered flow set; S404, calculate the super-period of the time-triggered flow set using the least common multiple method of ; S405, non-decreasingly sort the flows in the time-triggered flow set according to the hyperperiod size; S406, Traverse each time-triggered flow in the time-triggered flow set in the time-triggered flow set , if the traversal of the flow set is completed, enter S416; otherwise, enter S407; S407, Traverse the path set of the time-triggered flow, and by comparing the link load peaks of each path, find the path set with the smallest link load peak and sort it in non-decreasing order of path length; ​ S408, traverse each path in the path set with the minimum peak link load , if the path traversal is completed, go to S412, otherwise go to S409; S409, Calculate the time-triggered flow In the non-waiting mode, with as the transmission path, the end-to-end delay , if exceeds the time-triggered flow 's deadline , then enter S408 for the next loop, otherwise enter S410; S410, call the earliest transmission start time calculation algorithm to calculate the time-triggered stream in the non-waiting mode at as the transmission start time when it is the transmission path ; if the earliest transmission start time is found, that is , then enter S411; otherwise, enter S408 for the next loop; S411, Update time-triggered flow of the transmission start time and the optimal path , terminate the path traversal once one is found and enter S412; S412, if , go to S413, otherwise go to S415; S413, update the link time slot allocation table hop by hop and cycle by cycle, re-sort the allocated time slots on the affected links in ascending order of the transmission start time, and update the link load table, then enter S414; S414, update the transmission start time table and the routing entry table, and enter S406 for the next cycle; S415, clear the link time slot allocation table, the transmission start time table, and the routing entry table, and enter S416; S416, return the link time slot allocation table, the transmission start time table, and the routing entry table.

5. A computer storage medium stores a readable program, characterized in that, When the program runs, it can execute the routing and traffic joint scheduling method based on the link load peak described in any one of claims 1-3.

6. An electronic device, characterized in that, It includes: A processor, a memory, a communication interface, and a communication bus. The processor, the memory, and the communication interface complete communication with each other through the communication bus; The memory is used to store at least one executable instruction, and the executable instruction causes the processor to execute the operations corresponding to the routing and traffic joint scheduling method based on the link load peak described in any one of claims 1-3.

7. A computer program product comprising computer instructions, characterized in that, The computer instruction instructs the computing device to execute the operations corresponding to the routing and traffic joint scheduling method based on the link load peak described in any one of claims 1-3.

Images