JP2024516965A - Service Scheduling Method Based on Time-Triggered Ethernet - Google Patents

Abstract

The present invention discloses a service scheduling method based on time-triggered Ethernet, which calculates the number of end systems and switches according to service flow parameters, determines the topology structure, establishes a port connection matrix, plans the shortest path, allocates services in each matrix period, sorts services in each basic period, establishes an overall schedule in one matrix period, and generates a service schedule for each end node and switch. The present invention realizes the scheduling of TT services, so that each end node and switch can transmit TT tasks in order without contention, while saving network bandwidth and improving the utilization rate of network links. [Selected Figure]

Description

The present invention belongs to the technical field of computer network communications, and more specifically, to a service scheduling method based on time-triggered Ethernet.

With the development of the aerospace industry, the demand for avionics systems is increasing, including high bandwidth, high reliability, low latency and high fault tolerance. After a long period of research and development, TTE (Time Trigger Ethernet) has emerged. This technology adds time synchronization technology and time trigger technology to Ethernet, and can not only handle time-triggered services (TT) with very high real-time requirements, but also rate-limited services (RC) and ordinary Ethernet services (BE), which can simultaneously meet the needs of real-time and non-real-time applications.

The TT service has the strictest time certainty requirements among all services and has the highest priority. TT service transmission is based on a global clock and service transmission and reception must be performed at a specified period in strict accordance with a time schedule, thereby ensuring its high real-time and low delay jitter requirements and avoiding data frames competing with each other for the limited physical links in the network.

In view of this, the present invention aims to provide a service scheduling method based on time-triggered Ethernet that can realize transmission and reception control of TT services.

A service scheduling method based on time-triggered Ethernet, comprising:
Step 1: acquiring parameters of all time-triggered services to be transmitted and received;
Step 2 of calculating the number of required end nodes ES and switches SW based on the services to be transmitted and received, and obtaining a minimum system;
Step 3 of establishing a topology structure of the system based on the minimum system determined in step 2, in which both the end nodes ES and the switches SW are called connection nodes of the system, and an ordering is performed on the connection nodes;
Step 4: based on the topology structure of the system established in step 3, establishing a connection relationship matrix between connection nodes including a connection matrix between the connection ports of the switch SW and other switches SW or end nodes ES;
Step 5, based on the topology structure and the connection matrix of the connection nodes in steps 3 and 4, set the distance between two directly connected nodes to 1, set the connection distance of the same node to 0, and set the distance between nodes that are not directly connected to infinity, represented by X, and calculate the shortest transmission path between any two nodes, including the total distance of the shortest transmission path and the number of the intermediate connection node;
A step 6 of obtaining network parameters;
Step 7: Calculate a basic period BC and a matrix period MC according to the service flow parameters in step 1, where the matrix period MC includes multiple basic periods BC and is used to transmit each of all services once;
Step 8: assigning each service in each basic period BC in one matrix period MC according to the period of the service, and establishing a service assignment table in one matrix period MC;
Step 9: grouping all services allocated in each basic period BC according to different generation times of the services, sorting the services in each group based on priority, traversing each basic period BC in one matrix period MC, and completing sorting of the services according to priority;
Step 10: for a service in one matrix cycle MC, a transmission/reception route is determined using the shortest transmission route in step 5 based on the source number and destination number of the service, and a transmission/reception time is determined based on the priority of the service and the transmission/reception route, and finally a whole schedule is formed in one matrix cycle MC;
Step 11: identifying each connection node based on the overall schedule of step 10, obtaining a transmission/reception schedule of each connection node, and generating a schedule file;
and step 12 of scheduling transmission and reception for the service based on the schedule file of step 11.

Preferably, in step 8, after each service is assigned to each basic cycle BC, services with the same cycle are assigned to different basic cycles with a shift, to obtain a service assignment table.

Preferably, in step 5, the specific method for finally establishing one shortest transmission path table is as follows:
Arrange all the connection node numbers in row and column order,
The first intermediate node z between the connection node x and the connection node y is located at the xth row and yth column, the second intermediate node t is located at the zth row and yth column, and the tth row and yth column, and then the last intermediate node is located at the tth row and yth column. By analogy, the two connection nodes are traversed to obtain a shortest transmission path table.

Preferably, each switch SW includes four full-duplex ports.

Preferably, the two switches SW should be directly connected and the two end systems ES should be connected via a switch SW.

Preferably, the parameters of the time-triggered service include the ID number, frame length, period, source system number, destination system number and service generation time of each service.

The present invention has the following beneficial effects:

The present invention discloses a service scheduling method based on time-triggered Ethernet, which calculates the number of end systems and switches according to service flow parameters, determines a topology structure, establishes a port connection matrix, plans the shortest path, allocates services in each matrix period, sorts services in each basic period, establishes an overall schedule in one matrix period, and generates a service schedule for each end node and switch. The present invention realizes the scheduling of TT services, allowing each end node and switch to transmit TT tasks in sequence without contention, while saving network bandwidth and improving the utilization rate of network links.

1 is a flow chart of an implementation of the present invention. FIG. 2 is a network topology diagram in a specific embodiment of the present invention.

The present invention will be described in detail below with reference to the drawings and examples.

1. Obtain service flow parameters. Obtain all time-triggered (TT) services to be sent and received. The ID number, frame length, period, source system number, destination system number, and service generation time of each service are shown in Table 1.

Table 1 Service flow parameters

2. Calculate the number of required end systems ES and switches SW, each switch SW contains four full-duplex ports, two switches SW may be directly connected, and two end systems ES need to be connected via a switch SW.

If the number of end systems involved in the service flow is 8, then the number of required ESs in the system is at least 8, and the number of required SWs is at least m = 8/2 - 1 = 5, i.e., we obtain a minimum system containing 8 end nodes and 5 switches.

3. Based on the minimum system determined in step 2, establish a topology structure, as shown in Figure 2, where the end system ES and the switch SW are both called connection nodes of the system, and order the connection nodes, with numbers 1-8 representing end nodes ES1-ES8 and numbers 9-13 corresponding to switches SW1-SW5.

4. Establish a port connection matrix. Based on the topology structure in step 3, a port matrix is established, as shown in Table 2.

Table 2 Port connection table

The matrix has three columns, the first and second columns represent the connection nodes, and the third column represents the SW port numbers connected to the two connection nodes. For example, in the first row, the two connection nodes connected to each other are 9 and 10, and the port number is 1, representing that node 9 is connected to node 10 via its own port 1.

5. Plan the shortest path. Establish a distance matrix based on the topology structure and node connection matrix in steps 3 and 4, and for any two nodes, the distance between two directly connected nodes is set to 1, the connection distance of the same node is 0, and the distance between nodes that are not directly connected is set to infinity, represented by X, as shown in Table 3.

Table 3 Connection distance table

The first column of the first row indicates that the distance between node 1 and itself is 0, the second column of the first row indicates that the distance between node 1 and node 2 is infinite, i.e., they are not directly connected, and the tenth column of the first row indicates that the distance between node 1 and node 10 is 1.

Based on the distance matrix, the shortest transmission path between any two nodes is calculated, including the total distance and intermediate node numbers, as shown in Tables 4 and 5, respectively.

Table 4 Transmission distance table

For example, the 7th column of the 1st row indicates that the total transmission distance from node 1 to node 7 is 3.

Table 5. Shortest transmission route table

For example, when inquiring about the transmission path from node 1 to node 7, first the 7th column of the 1st row is confirmed as 10, indicating that transmission will be from node 1 to node 10 in step 1, then the 7th column of the 10th row is confirmed as 11, indicating that transmission will be from node 10 to node 11 in step 2, then the 7th column of the 11th row is confirmed as 7, indicating that transmission will be from node 11 to node 7 in step 3. In other words, transmission from node 1 to node 7 is achieved in three steps.

6. Obtain network parameters, including link bandwidth (100 Mbps), synchronization accuracy (200 ns), delay between the end system and the switch (100 ns), delay of the end node itself (20 ns), delay of the switch itself (20 ns), etc.

7. Calculate the basic period BC and the matrix period MC. Based on the service flow parameters in step 1, calculate the basic period BC and the matrix period MC. The matrix period MC includes multiple basic periods BC, and transmits each of all services once. The basic period BC is set to the greatest common divisor of each service period MC, 1 ms, and the matrix period MC is set to the least common multiple of each service period, 12 ms.

8. Establish a TT service allocation table in one matrix cycle MC. Depending on the status of the TT service, allocate the TT service to each basic cycle BC in one matrix cycle MC, and further perform a basic cycle BC shift process for the TT service in the same cycle to obtain an optimized allocation means and avoid the contention of the TT service, as shown in Table 6.

Table 6 TT service allocation table for one matrix period MC

9. Process the allocation table within each BC. Process the TT services allocated within each basic cycle, group them based on the different generation times of the TT services, and perform priority processing for the services in each group based on the port's transmission and reception volume. First, process the first BC. Before processing, the results are shown in Table 7, and after processing, the results are shown in Table 8.

Table 7 Before treatment

Table 8 After treatment

10. Establish an overall schedule for one matrix cycle MC. The same process is performed for each basic cycle BC in step 9 until service processing for all basic cycles BC in one matrix cycle MC is completed. Next, time scheduling is performed for the TT services processed in one matrix cycle MC, and the transmission and reception paths are determined using the shortest transmission path table based on the source number and destination number of the service, and the transmission and reception times are determined based on the priority of the service and the transmission and reception paths, and finally an overall schedule for one matrix cycle MC is formed.

The first column is the ID number, the second column is the basic cycle number, the third to sixth columns represent transmission from the source node, and are the source node number, transmission start time, transmission end time, and window length, respectively, followed by intermediate node reception and intermediate node transmission. There are four columns in one group, and each group represents the intermediate node number, reception/transmission start time, reception/transmission completion time, and window length, respectively, and the last four columns represent destination node reception, and are the destination node number, reception start node, reception completion node, and window length, respectively.

11. Generate a schedule for each node. Identify each node based on the overall schedule in step 10, obtain the transmission and reception schedule for each node, and generate a schedule file.

The sending table of source system 1, the receiving table of switch 2, the sending table of switch 3, and the receiving table of destination system 4 are shown in Tables 9-12, respectively.

Table 9: Source System 1 Transmission Table

Table 10. Switch 2 receive table

Table 11. Switch 3 Transmission Table

Table 12 Reception table of destination system 4

As mentioned above, the preferred embodiment of the present invention has been described, but it is not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made without departing from the spirit and principles of the present invention should be included in the scope of protection of the present invention.

(Additional Note)
(Appendix 1)
Step 1: acquiring parameters of all time-triggered services to be transmitted and received;
Step 2 of calculating the number of required end nodes ES and switches SW based on the services to be transmitted and received, and obtaining a minimum system;
Step 3 of establishing a topology structure of the system based on the minimum system determined in step 2, in which both the end nodes ES and the switches SW are called connection nodes of the system, and an ordering is performed on the connection nodes;
Step 4: based on the topology structure of the system established in step 3, establishing a connection relationship matrix between connection nodes including a connection matrix between the connection ports of the switch SW and other switches SW or end nodes ES;
Step 5, based on the topology structure and the connection matrix of the connection nodes in steps 3 and 4, set the distance between two directly connected nodes to 1, set the connection distance of the same node to 0, and set the distance between nodes that are not directly connected to infinity, represented by X, and calculate the shortest transmission path between any two nodes, including the total distance of the shortest transmission path and the number of the intermediate connection node;
A step 6 of obtaining network parameters;
Step 7: Calculate a basic period BC and a matrix period MC according to the service flow parameters in step 1, where the matrix period MC includes multiple basic periods BC and is used to transmit each of all services once;
Step 8: assigning each service in each basic period BC in one matrix period MC according to the period of the service, and establishing a service assignment table in one matrix period MC;
Step 9: grouping all services allocated in each basic period BC according to different generation times of the services, sorting the services in each group based on priority, traversing each basic period BC in one matrix period MC, and completing sorting of the services according to priority;
Step 10: for a service in one matrix cycle MC, a transmission/reception route is determined using the shortest transmission route in step 5 based on the source number and destination number of the service, and a transmission/reception time is determined based on the priority of the service and the transmission/reception route, and finally a whole schedule is formed in one matrix cycle MC;
Step 11: identifying each connection node based on the overall schedule of step 10, obtaining a transmission/reception schedule of each connection node, and generating a schedule file;
Step 12: scheduling transmission and reception for the service based on the schedule file of step 11.

(Appendix 2)
The service scheduling method based on time-triggered Ethernet described in Appendix 1, characterized in that in step 8, after each service is assigned to each basic period BC, services with the same period are assigned to different basic periods with a shift, thereby obtaining a service assignment table.

(Appendix 3)
In the step 5, the specific method for finally establishing one shortest transmission path table is as follows:
Arrange all the connection node numbers in row and column order,
The method for scheduling a service based on time-triggered Ethernet as described in Supplementary Note 1, characterized in that the first intermediate node z between the connecting node x and the connecting node y is located at the xth row, yth column, the second intermediate node t is located at the zth row, yth column, and the tth row, yth column, and so on until the last intermediate node, traversing the two connecting nodes to obtain a shortest transmission path table.

(Appendix 4)
The method for scheduling services based on time-triggered Ethernet according to claim 1, characterized in that each switch SW includes four full-duplex ports.

(Appendix 5)
The method for service scheduling based on time-triggered Ethernet according to claim 1, characterized in that two switches SW need to be directly connected and two end systems ES need to be connected via a switch SW.

(Appendix 6)
The method for scheduling services based on time-triggered Ethernet according to claim 1, wherein the parameters of the time-triggered services include the ID number, frame length, period, source system number, destination system number and service generation time of each service.

Claims (6)

Step 1: acquiring parameters of all time-triggered services to be transmitted and received;
Step 2 of calculating the number of required end nodes ES and switches SW based on the services to be transmitted and received, and obtaining a minimum system;
Step 3 of establishing a topology structure of the system based on the minimum system determined in step 2, in which both the end nodes ES and the switches SW are called connection nodes of the system, and an ordering is performed on the connection nodes;
Step 4: based on the topology structure of the system established in step 3, establishing a connection relationship matrix between connection nodes including a connection matrix between the connection ports of the switch SW and other switches SW or end nodes ES;
Step 5, based on the topology structure and the connection matrix of the connection nodes in steps 3 and 4, set the distance between two directly connected nodes to 1, set the connection distance of the same node to 0, and set the distance between nodes that are not directly connected to infinity, represented by X, and calculate the shortest transmission path between any two nodes, including the total distance of the shortest transmission path and the number of the intermediate connection node;
A step 6 of obtaining network parameters;
Step 7: Calculate a basic period BC and a matrix period MC according to the service flow parameters in step 1, where the matrix period MC includes multiple basic periods BC and is used to transmit each of all services once;
Step 8: assigning each service in each basic period BC in one matrix period MC according to the period of the service, and establishing a service assignment table in one matrix period MC;
Step 9: grouping all services allocated in each basic period BC according to different generation times of the services, sorting the services in each group based on priority, traversing each basic period BC in one matrix period MC, and completing sorting of the services according to priority;
Step 10: for a service in one matrix cycle MC, a transmission/reception route is determined using the shortest transmission route in step 5 based on the source number and destination number of the service, and a transmission/reception time is determined based on the priority of the service and the transmission/reception route, and finally a whole schedule is formed in one matrix cycle MC;
Step 11: identifying each connection node based on the overall schedule of step 10, obtaining a transmission/reception schedule of each connection node, and generating a schedule file;
Step 12: scheduling transmission and reception for the service based on the schedule file of step 11. The method for scheduling services based on time-triggered Ethernet according to claim 1, characterized in that in step 8, after each service is assigned to each basic period BC, services of the same period are assigned to different basic periods with a shift, thereby obtaining a service assignment table. In the step 5, the specific method for finally establishing one shortest transmission path table is as follows:
Arrange all the connection node numbers in row and column order,
The method for scheduling a service based on time-triggered Ethernet according to claim 1, characterized in that the first intermediate node z between the connecting node x and the connecting node y is located at the xth row, yth column, the second intermediate node t is located at the zth row, yth column, and the tth row, yth column, and so on until the last intermediate node, traversing the two connecting nodes to obtain a shortest transmission path table. The method for service scheduling based on time-triggered Ethernet according to claim 1, characterized in that each switch SW includes four full-duplex ports. The method for service scheduling based on time-triggered Ethernet according to claim 1, characterized in that two switches SW must be directly connected and two end systems ES must be connected via a switch SW. The method for scheduling services based on time-triggered Ethernet according to claim 1, characterized in that the parameters of the time-triggered services include an ID number, frame length, period, source system number, destination system number, and service generation time of each service.

Images