Routing path determination method and network device
By designing network devices and routing path determination methods in time-sensitive networks, the problem that the existing technology cannot plan multiple traffic types of routing paths is solved, and survivable routing path planning is realized in the online state, reducing the computational complexity and time Cost, meets the dynamic configuration requirements of Industry 4.0 and future industrial automation.
Patent Information
- Application Number
- CN202311594188.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-11-13
- Filing Date
- 2023-11-27
- Publication Date
- 2024-12-03
AI Technical Summary
The existing technology cannot effectively plan multiple traffic types of routing paths in an online state, resulting in increased computing volume and time complexity, and cannot meet the dynamic configuration requirements of Industry 4.0 and future industrial automation.
A network device and routing path determination method are designed. By receiving data flow requests, candidate mutually exclusive paths are determined, the final routing path is initialized, the temporary routing path is selected, and the final routing path is updated when the conditions are met to optimize the transmission of data flow. .
It realizes survivable routing path planning for multiple traffic types of data streams in the online state, reduces computing complexity and time cost, meets dynamic configuration requirements, and improves the performance utilization of time-sensitive networks.
Smart Images

Figure CN119071209A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a routing path determination method and a network device, and in particular to a routing path determination method and a network device for determining a survivable routing path in a time-sensitive network. Background Art
[0002] Time-sensitive networking (TSN) is a series of standards defined by the TSN working group in the IEEE 802.1 working group to achieve deterministic communication for safety-critical real-time applications in industrial networks. Time-sensitive networks can support data streams of various traffic types. However, determining the routing paths for data streams requires a huge amount of computation and time cost. Therefore, how to determine the routing paths for data streams and how to optimize the performance of time-sensitive networks have always been one of the important issues in this field.
[0003] The prior art has proposed a variety of methods for determining routing paths on time-sensitive networks. However, most of the prior art is about routing path planning algorithms in an off-line state. Therefore, the prior art cannot adapt to the interoperability required by Industry 4.0 and the dynamic configuration requirements of future industrial automation (such as dynamically increasing data flows). In addition, the prior art generally only considers routing path planning for a single type of traffic type data flow, such as only TSN data flow or only video bridging (AVB) data flow. Therefore, in this case, the prior art cannot make the most comprehensive and effective use of time-sensitive networks.
[0004] On the other hand, the TSN working group has also developed a fault-tolerant mechanism called Frame Replication and Elimination for Reliability (FRER), namely the IEEE 802.1CB standard. Among them, the frames of the data stream are copied at the sending end and transmitted to the receiving end through different paths at the same time, making the transmission of important data more real-time and reliable. Although the IEEE 802.1CB standard implements a redundant transmission mechanism on a time-sensitive network, under this mechanism, the routing paths required to transmit data streams increase exponentially, resulting in an increase in the amount of calculation and time complexity required for routing path planning. Therefore, there is an urgent need for a routing path planning algorithm that can take into account both computing efficiency and various traffic types in an online state to solve the problems faced in this field. Summary of the invention
[0005] Therefore, the main purpose of the present invention is to provide a method and device that can simultaneously consider various traffic types of data flows and perform survivable routing path planning in a limited time under an online state, so as to improve the shortcomings of the prior art.
[0006] The embodiment of the present invention discloses a network device for a time-sensitive network system, wherein the time-sensitive network system includes a plurality of background data streams. The network device includes a processing unit and a storage unit. The processing unit is used to execute a program code; the storage unit is coupled to the processing unit and is used to store the program code to instruct the processing unit to execute a routing path determination method. The routing path determination method sequentially comprises: (a) receiving a request for allocating a first plurality of data flows; (b) determining a first plurality of candidate edge-disjoint routing paths for each of the first plurality of data flows according to a network topology of the time-sensitive network system; (c) initializing a plurality of final routing path pairs for the first plurality of data flows and the plurality of background data flows according to a first plurality of lengths of the first plurality of candidate edge-disjoint routing paths; (d) selecting a plurality of temporary routing path pairs for the first plurality of data flows and the plurality of background data flows according to a first plurality of parameters; (e) updating the first plurality of parameters according to the plurality of temporary routing path pairs; and (f) comparing the plurality of final routing path pairs with the plurality of temporary routing path pairs, and updating the plurality of final routing path pairs with the plurality of temporary routing path pairs when the plurality of temporary routing path pairs are superior to the plurality of final routing path pairs. The time-sensitive network system transmits the first plurality of data flows and the plurality of background data flows according to the plurality of final routing paths.
[0007] The embodiment of the present invention further discloses a routing path determination method for a time-sensitive network system. The time-sensitive network system includes a plurality of background data flows. The routing path determination method sequentially includes (a) receiving a request for allocating a first plurality of data flows; (b) determining a first plurality of candidate mutually exclusive path pairs for each data flow of the first plurality of data flows according to a network topology of the time-sensitive network system; (c) initializing a plurality of final routing path pairs for the first plurality of data flows and the plurality of background data flows according to a first plurality of lengths of the first plurality of candidate mutually exclusive path pairs; (d) selecting a plurality of temporary routing path pairs for the first plurality of data flows and the plurality of background data flows according to a first plurality of parameters; (e) updating the first plurality of parameters according to the plurality of temporary routing path pairs; and (f) comparing the plurality of final routing path pairs with the plurality of temporary routing path pairs, and updating the plurality of final routing path pairs with the plurality of temporary routing path pairs when the plurality of temporary routing path pairs are superior to the plurality of final routing path pairs. The time-sensitive network system transmits the first plurality of data flows and the plurality of background data flows according to the plurality of final routing paths. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] Figure 1 FIG. 4 is a schematic diagram of a time-sensitive network system according to an embodiment of the present invention. Figure 2 FIG. 4 is a schematic diagram of a network topology according to an embodiment of the present invention. Figure 3 A schematic diagram of a process of an embodiment of the present invention. Figure 4 FIG. 4 is a schematic diagram of a network device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0009] Certain words are used in the specification and the subsequent claims to refer to specific components. It should be understood by those skilled in the art that hardware manufacturers may use different terms to refer to the same component. This specification and the subsequent claims do not use differences in names as a way to distinguish components, but use differences in the functions of components as the criteria for distinction. The term "including" mentioned throughout the specification and the subsequent claims is an open-ended term and should be interpreted as "including but not limited to". In addition, the term "coupled" herein includes any direct and indirect electrical connection means. Therefore, if the text describes a first device coupled to a second device, it means that the first device can be directly electrically connected to the second device, or indirectly electrically connected to the second device through other devices or connection means.
[0010] Please refer to Figure 1 , Figure 1 FIG. 1 is a schematic diagram of a time-sensitive network system 10 according to an embodiment of the present invention. The time-sensitive network system 10 operates according to a fully centralized model of time-sensitive networking, and can be simply composed of a centralized network configuration (CNC) server 100, a centralized user configuration (CUC) server 110, end devices 120, 122, and time-sensitive network bridges (TSN bridges) 130, 132. The end devices 120, 122 and the time-sensitive network bridges 130, 132 are connected by physical links, and the end devices 120 and 122 can execute time-aware applications that require deterministic communication.
[0011] exist Figure 1 In the figure, the centralized network configuration server 100, the centralized user configuration server 110, the terminal devices 120, 122 and the time-sensitive network bridges 130, 132 are only used to illustrate the architecture of the time-sensitive network system 10. In fact, the terminal devices 120, 122 can be but not limited to devices such as robots, sensors, machines or programmable logic controllers (PLCs), and can perform end-to-end communication with other terminal devices. In other words, the terminal devices 120, 122 can be senders (talkers) or receivers (listeners) of data streams defined by multiple parameters. It should be noted that the terminal devices 120, 122 can be senders of one data stream and receivers of another data stream at the same time, and different transmission quality of service (QoS) requirements of different data streams in the time-sensitive network should be met in the end-to-end communication. In general, various QoS requirements of data streams can be stored in the terminal devices 120, 122 as the basis for scheduling and routing of data streams.
[0012] The centralized user configuration server 110 communicates with the centralized network configuration server 100 and the terminal devices 120 and 122, and can obtain the QoS requirements required for the data flow from the terminal devices 120 and 122, thereby requesting scheduling and routing configuration from the centralized network configuration server 100, and finally transmitting the configuration results of the data flow back to the terminal devices 120 and 122.
[0013] The centralized network configuration server 100 is a network device that calculates network routing, scheduling, and configures network resources. The centralized network configuration server 100 can communicate with the centralized user configuration server 110 and the time-sensitive network bridges 130 and 132 to obtain the QoS requirements of the data flow from the centralized user configuration server 110 and configure the time-sensitive network bridges 130 and 132. In other words, the centralized network configuration server 100 performs data flow scheduling and routing path calculation according to the QoS requirements obtained from the centralized user configuration server 110 to determine the network configuration, and deploys the network configuration results to the time-sensitive network bridges 130 and 132 to perform end-to-end communication of the data flow.
[0014] The time-sensitive network bridges 130 and 132 may be Ethernet switches, which may receive network configurations from the centralized network configuration server 100 and identify traffic types of data flows. According to the gate control list (GCL) determined by the centralized network configuration server 100, the time-sensitive network bridges 130 and 132 forward data flows from one terminal device to another terminal device, thereby achieving deterministic communication in the time-sensitive network system 10.
[0015] According to the TSN specification, the traffic types of data flows can be TSN, AVB (audio video bridging) and BE (best effort) in descending order of priority. That is, the frames of TSN data flows can preempt the frames of AVB data flows and BE data flows and be transmitted preferentially in the TSN system 10. In the embodiment of the present invention, the TSN system 10 can support the routing path determination of TSN data flows and AVB data flows.
[0016] Time-Sensitive Networking Data Flow λ k Can be defined based on multiple parameters, including the source node (sender) I k , destination node (receiver) J k 、Size(size)S k 、Period (period) P k 、Offset F k and deadline k That is, with size S k The data needs to be at the time point {F k ,F k +P k ,F k +2Pk ,…} is periodically sent by the source node I k Send to destination node J k , and needs to be in D k Arrive at destination node J within time unit k .
[0017] Audio Video Bridging Data Streamλ k Can be defined based on multiple parameters, such as source node I k , destination node J k , Size S k , period P k , type (e.g. high priority level A or low priority level B) T k and deadline D k That is, with size S k and type T k The data must be in each P k The time unit is periodically sent by the source node I k Send to destination node J k , and need to be within the deadline D k Arrive at destination node J within time unit k .
[0018] Please refer to Figure 2 , Figure 2 Schematic diagram of a network topology 20 according to an embodiment of the present invention. The network topology 20 includes two terminal devices ED1 and ED2 connected by physical links and seven time-sensitive network bridges BR1 to BR7. The physical links have corresponding bandwidth capacities. In addition, the network topology 20 can be simulated as a directed-graph, the terminal devices ED1 and ED2 and the time-sensitive network bridges BR1 to BR7 can be simulated as vertices of the graph, and the physical links can be simulated as edges of the graph. It should be noted that Figure 2 This is only used to illustrate the concept of the present invention. In practical applications, the number of terminal devices, the number of time-sensitive network bridges, and the connection method are not limited thereto.
[0019] like Figure 2As shown, there may be multiple paths from the terminal device ED1 to the terminal device ED2, for example, such as the path {ED1, BR1, BR5, BR3, BR4, ED2} or the path {ED1, BR1, BR2, BR3, BR6, BR4, ED2}, etc., which can be used to transmit the data stream ST1 from the source node ED1 to the destination node ED2. In one embodiment, the path P1 {ED1, BR1, BR2, BR3, BR4, ED2} and the path P2 {ED1, BR1, BR5, BR6, BR4, ED2} are mutually exclusive path pairs (edge-disjoint routing paths) that can be used for survivable routing paths. On the paths P1 and P2, except for the links {ED1, BR1} and {BR4, ED2} between the terminal devices ED1 and ED2 and the time-sensitive network bridges BR1 and BR4 connecting the terminal devices ED1 and ED2, all other links are different (mutually exclusive). The frames of data stream ST1 can be copied and transmitted simultaneously through different paths P1 and P2 in the time-sensitive network bridge BR1, and the later frames are deleted in the time-sensitive network bridge BR4. Thus, a redundant transmission mechanism of data streams in the time-sensitive network is realized. Under this mechanism, the situation that a single network node or link failure causes transmission interruption can be avoided, thereby improving the reliability of data transmission.
[0020] exist Figure 2 In addition to path P1 and path P2, path P1 and P3 {ED1, BR7, ED2}, and path P2 and path P3, etc. are also mutually exclusive path pairs that can be used for survivable routing paths. It should be noted that there are multiple sets of mutually exclusive path pairs, such as path {ED1, BR1, BR5, BR3, BR4, ED2} and path {ED1, BR1, BR7, BR6, BR4, ED2}, etc. (not shown in the figure), which are all mutually exclusive path pairs that can be used for end-to-end survivable routing paths between terminal device ED1 and terminal device ED2. Therefore, in actual applications, facing a larger-scale network topology and a huge number of data streams, there will be more possible mutually exclusive path pairs for data streams to transmit data. In this case, the comprehensive planning of routing paths will be an extremely complex and time-consuming problem.
[0021] In the embodiment of the present invention, the centralized network configuration server 100 determines the survivable routing path and schedule of each data flow according to the plurality of parameters of each data flow obtained by the centralized user configuration server 110, and configures the time-sensitive network bridge according to the determined routing path and schedule. In addition to determining the routing path for the first plurality of data flows that need to be dynamically added, the centralized network configuration server 100 also considers the routing paths of the plurality of background data flows, so as to maximize the number of schedulable data flows. The method for determining the survivable routing path and schedule can be summarized as a process 30, such as Figure 3 Process 30 includes the following steps:
[0022] Step 300: Start.
[0023] Step 302: Receive a request for allocating a first plurality of data flows.
[0024] Step 304: Determine a first plurality of candidate mutually exclusive path pairs for each data flow of a first plurality of data flows according to the network topology of the time-sensitive network system.
[0025] Step 306: Initialize a plurality of final routing path pairs for the first plurality of data flows and a plurality of background data flows according to a first plurality of lengths of the first plurality of candidate mutually exclusive path pairs.
[0026] Step 308: Select a plurality of temporary routing path pairs for the first plurality of data flows and the plurality of background data flows according to the first plurality of parameters.
[0027] Step 310: Update a first plurality of parameters according to a plurality of temporary routing path pairs.
[0028] Step 312: Compare the plurality of final routing path pairs and the plurality of temporary routing path pairs, and when the plurality of temporary routing path pairs are better than the plurality of final routing path pairs, update the plurality of final routing path pairs with the plurality of temporary routing path pairs.
[0029] Step 314: Determine whether the timer has expired. If so, execute step 316; if not, continue to execute step 308.
[0030] Step 316: The time-sensitive network system transmits the first plurality of data flows and the plurality of background data flows according to the plurality of final routing path pairs.
[0031] Step 318: End.
[0032] According to the process 30, the centralized network configuration server 100 receives a request for allocating a first plurality of data flows that need to be dynamically added to the time-sensitive network system 10 (step 302), and determines a first plurality of candidate mutually exclusive path pairs for each data flow of the first plurality of data flows according to the request content and the network topology of the time-sensitive network system 10 (step 304). The centralized network configuration server 100 initializes a plurality of final routing path pairs for the first plurality of data flows and a plurality of background data flows according to a first plurality of lengths of the first plurality of candidate mutually exclusive path pairs (step 306). Next, the centralized network configuration server 100 selects a plurality of temporary routing path pairs for the first plurality of data flows and a plurality of background data flows according to a first plurality of parameters (step 308), and updates the first plurality of parameters according to the plurality of temporary routing path pairs (step 310). The centralized network configuration server 100 compares the plurality of final routing path pairs and the plurality of temporary routing path pairs, and selects a better temporary routing path pair therefrom to update the plurality of final routing path pairs (step 312). Before a timer expires, the centralized network configuration server 100 repeats steps 308 to 312 to continuously search for a better plurality of final routing path pairs (step 314). Finally, after the timer expires, the time-sensitive network system 10 configures and transmits the first plurality of data streams and the plurality of background data streams according to the best plurality of final routing path pairs found at that time (step 316). Accordingly, the time-sensitive network system 10 can dynamically add a new data stream, and can simultaneously consider the newly added data stream and the background data stream to efficiently determine the routing path, schedule, and configure for all data streams.
[0033] Specifically, in step 302, the centralized network configuration server 100 receives a request for allocating the first plurality of data streams. In the embodiment of the present invention, the fully centralized mode of the time-sensitive network is taken as an example, and the centralized user configuration server 110 obtains the demand for configuring the data stream from the terminal devices 120 and 122 and transmits the request for allocating the first plurality of data streams to the centralized network configuration server 100 accordingly, but is not limited thereto. For example, according to different models of different time-sensitive networks, such as a centralized network / distributed user model, the centralized network configuration server 100 may receive the request for allocating the first plurality of data streams from the time-sensitive network bridges 130 and 132.
[0034] In step 304, the centralized network configuration server 100 determines a first plurality of candidate mutually exclusive path pairs for each data flow of the first plurality of data flows according to the network topology of the time-sensitive network system 10. As mentioned above, in a large-scale time-sensitive network system, each data flow may have multiple mutually exclusive path pairs that may be used to transmit data. Therefore, compared with determining a routing path according to all the first plurality of mutually exclusive path pairs that may be used for each data flow of the first plurality of data flows, the embodiment of the present invention determines the first plurality of candidate mutually exclusive path pairs from the first plurality of mutually exclusive path pairs to reduce the complexity of determining the routing path.
[0035] In detail, the embodiment of the present invention obtains the first plurality of mutually exclusive path pairs for each data stream of the first plurality of data streams according to a k-pair mutually exclusive path algorithm (Efficient algorithms for listing k disjoint st-paths in graphs by Grossi et al.). The k-pair mutually exclusive path algorithm has the characteristics of being unique and exhaustive, and the mutually exclusive path pairs obtained according to the k-pair mutually exclusive path algorithm are all paths suitable for survivable routing. However, the k-pair mutually exclusive path algorithm cannot guarantee the acquisition of k mutually exclusive path pairs with the shortest path length. In a time-sensitive network, using a path with a shorter path length for transmission can reduce scheduling conflicts, thereby improving the overall schedulability of the network. Therefore, the embodiment of the present invention uses a deviation partitioning technique to enable the k-pair mutually exclusive path algorithm to output k mutually exclusive path pairs with the shortest path length as candidate mutually exclusive path pairs in ascending order according to the length of the mutually exclusive path pairs. Accordingly, the first plurality of candidate mutually exclusive path pairs for each data stream of the first plurality of data streams can be determined according to the plurality of lengths of the first plurality of mutually exclusive path pairs for each data stream of the first plurality of data streams. A mutually exclusive path pair is composed of two mutually exclusive paths whose links are not repeated, and the length of the mutually exclusive path pair is the sum of the lengths of the two mutually exclusive paths whose links are not repeated. In the embodiment of the present invention, the length of the path is calculated using a number of hop counts of the path, but is not limited thereto.
[0036] According to the process 30, in step 306, the centralized network configuration server 100 initializes a plurality of final routing path pairs for the first plurality of data streams and the plurality of background data streams according to the first plurality of lengths of the first plurality of candidate mutually exclusive path pairs. As mentioned above, the network topology of the time-sensitive network system 10 can be simulated as a directed graph, and the process of determining the routing path for each data stream according to the network topology is actually a solution process of calculating the best available path on the directed graph, and the plurality of final routing path pairs for the first plurality of data streams and the plurality of background data streams in the embodiment of the present invention are the solution to the problem. The embodiment of the present invention selects the mutually exclusive path pair with the shortest path length in the first plurality of candidate mutually exclusive paths for each data stream of the first plurality of data streams as the routing path of the first plurality of data streams in the plurality of final routing path pairs, and in addition, selects the current mutually exclusive path pair of each data stream for each data stream of the plurality of background data streams as the routing path of the plurality of background data streams in the plurality of final routing path pairs, thereby initializing the plurality of final routing path pairs. The current mutually exclusive path pair of each data flow of the plurality of background data flows is the current mutually exclusive path pair of each data flow for transmitting the plurality of background data flows selected according to process 30 from the second plurality of candidate mutually exclusive path pairs for each data flow of the plurality of background data flows before the centralized network configuration server 100 receives a request to allocate the first plurality of data flows.
[0037] In this embodiment, in addition to initializing a plurality of final routing path pairs, the centralized network configuration server 100 also needs to initialize a first plurality of parameters and a second plurality of parameters. The first plurality of parameters include a plurality of pheromone values for a plurality of background data streams and a first plurality of data streams according to an Ant Colony Optimization (ACO) algorithm. In detail, the pheromone value of each mutually exclusive path pair of the second plurality of candidate mutually exclusive path pairs of each data stream of the plurality of background data streams is a known value, and the pheromone value of each mutually exclusive path pair of the first plurality of candidate mutually exclusive path pairs of each data stream of the first plurality of data streams can be initialized to 1, but is not limited thereto.
[0038] The second plurality of parameters include a plurality of visibility values for the plurality of background data streams and the first plurality of data streams according to the ant colony optimization algorithm. Specifically, if the data stream belongs to an audio video bridge data stream, the visibility value of each mutually exclusive path pair of the corresponding candidate mutually exclusive path pair may be the inverse of the worst-case delays of the mutually exclusive path pair; if the data stream belongs to a time-sensitive network data stream or other traffic types, the visibility value of each mutually exclusive path pair of the corresponding candidate mutually exclusive path pair may be the inverse of the exponential path length of the mutually exclusive path pair. In addition, the visibility value of the current mutually exclusive path pair of each data stream of the plurality of background data streams may be multiplied by a preference parameter. The preference parameter may be predetermined according to an empirical rule to increase the visibility value of the currently used mutually exclusive path pair of the plurality of background data streams, so that the chance of the currently used mutually exclusive path pair being selected again according to the ant colony optimization algorithm is increased. Under this mechanism, the situation where the routing paths of the plurality of background data streams are replaced can be reduced. In one embodiment, the preference parameter may be determined according to the cost of rerouting the plurality of background data flows, so as to adjust the probability of changing the routing paths of the plurality of background data flows.
[0039] According to process 30, in step 308, the centralized network configuration server 100 selects a plurality of temporary routing path pairs for the first plurality of data streams and the plurality of background data streams according to the first plurality of parameters. In an embodiment of the present invention, the selection of the plurality of temporary routing path pairs is to select a mutually exclusive path pair as a plurality of temporary routing paths for each data stream from the first plurality of candidate mutually exclusive path pairs of each data stream of the first plurality of data streams and the second plurality of candidate mutually exclusive path pairs of each data stream of the plurality of background data streams. The mutually exclusive path pair of each data stream is selected according to a transition probability, and the transition probability is determined according to the first plurality of parameters and the second plurality of parameters for the plurality of background data streams and the first plurality of data streams. The transition probability can be calculated according to the following formula 1:
[0040]
[0041] Among them, χ[λ k ] represents the data flow λ k The multiple candidate mutually exclusive path pairs, ψ[λ k ][z] represents the data flow λ k The visibility value of the zth candidate mutually exclusive path pair, φ[λ k ][z] represents the data flow λ k The pheromone value of the zth candidate mutually exclusive path pair, Probability represents the data flow λ kThe centralized network configuration server 100 selects the candidate mutually exclusive path pair with the largest conversion probability as the plurality of temporary routing path pairs according to each data stream of the first plurality of data streams and the plurality of background data streams.
[0042] According to process 30, in step 310, the centralized network configuration server 100 updates the first plurality of parameters according to the plurality of temporary routing path pairs. The plurality of pheromone values are updated according to an evaporation coefficient of the ant colony optimization algorithm. The evaporation coefficient is a predetermined value, which can be between 0 and 1, and is used to reduce the pheromone values of all candidate mutually exclusive path pairs for each data stream of the plurality of background data streams and the first plurality of data streams in each iteration of the ant colony optimization algorithm. Data stream λ k The updated pheromone value of the zth candidate mutually exclusive path pair can be ρ·φ[λ k ][z], where ρ is the evaporation coefficient. In addition, the pheromone values corresponding to the plurality of temporary routing paths selected in step 308 need to be increased according to a predetermined value such as Where cost is an incremental parameter.
[0043] Next, in step 312, the centralized network configuration server 100 compares the plurality of final routing path pairs and the plurality of temporary routing path pairs, and when the plurality of temporary routing path pairs are better than the plurality of final routing path pairs, the plurality of final routing path pairs are updated with the plurality of temporary routing path pairs. The comparison of the plurality of final routing path pairs and the plurality of temporary routing path pairs is performed based on an objective function, and the plurality of temporary routing path pairs are better than the plurality of final routing path pairs when a value of the objective function of the plurality of temporary routing path pairs is less than a value of the objective function of the plurality of final routing path pairs.
[0044] In detail, the objective function of the embodiment of the present invention can be expressed as follows:
[0045] Objective()=W 1 O 1 +W 2 O 2 +W 3 O 3 +W 4 O 4 (Formula 2)
[0046] Among them, Objective() represents the objective function, 1 , O 2 , O 3 and O 4represents the number of unschedulable time-sensitive network data flows, the number of unschedulable audio video bridging data flows, the number of rescheduled background data flows, and the sum of delays of all audio video bridging data flows, and is not limited thereto. It should be noted that according to the worst-case delay of the data flow and the deadline D of the data flow, k , the schedulability of the data flow is determined. If the worst-case delay is not greater than the deadline D k , the data flow is a schedulable data flow. If the worst-case delay is greater than the deadline D k , the data flow is an unschedulable data flow. It should be noted that the worst-case delay depends on the timing model of the time-sensitive network. In addition, W 1 , W 2 , W 3 and W 4 Respectively represent O 1 , O 2 , O 3 and O 4 The corresponding weight values are restricted as follows:
[0047]
[0048] in, represents a set of audio video bridging data streams in the first plurality of data streams, Represents a collection of audio video bridge data streams in multiple background data streams. Represents a collection of multiple background data streams.
[0049] The centralized network configuration server 100 calculates the objective function values of the plurality of temporary routing path pairs and the plurality of final routing path pairs according to the objective functions defined in Formula 2 and Formula 3, respectively. When the objective function values of the plurality of temporary routing path pairs are less than the objective function values of the plurality of final routing path pairs, it means that the plurality of temporary routing path pairs are better than the plurality of final routing path pairs. Therefore, the plurality of final routing path pairs are updated with the plurality of temporary routing path pairs.
[0050] According to process 30, in step 314, the centralized network configuration server 100 determines whether to continue iteration according to whether a predetermined timer expires. Before the timer expires, the centralized network configuration server 100 continues to execute steps 308 to 312 to obtain a better plurality of final routing path pairs. After the timer expires, the time-sensitive network system 10 can perform network configuration according to the plurality of final routing path pairs in step 316. In detail, the centralized network configuration server 100 responds to the request for allocating the first plurality of data streams sent by the centralized user configuration server 110 according to the routing paths and schedules of the plurality of final routing path pairs, and the centralized user configuration server 110 then transmits the relevant network configuration to the terminal devices 120 and 122. In addition, the centralized network configuration server 100 is deployed to the time-sensitive network bridges 130 and 132 according to the routing paths and schedules of the plurality of final routing path pairs. It should be noted that the centralized network configuration server 100 may only configure the time-sensitive network bridges passed by the dynamically added first plurality of data flows and the background data flows with modified routing paths.
[0051] Accordingly, the time-sensitive network system 10 can dynamically add new data flows, and can simultaneously consider the newly added data flows and background data flows to determine routing paths, schedules, and configurations for all data flows within a limited time.
[0052] For further information, please refer to Figure 4 , which is a schematic diagram of a network device 40 according to an embodiment of the present invention. The network device 4 can be used to implement the centralized network configuration server 100 according to an embodiment of the present invention. Figure 4 As shown, the network device 4 may include a processing unit 400 and a storage unit 410. The processing unit 400 may be a general purpose processor, a microprocessor, an application specific integrated circuit (ASIC), etc. or a combination thereof. The storage unit 410 is coupled to the processing unit 400 and may be any data storage device for storing a program code 412, and the program code 412 is read and executed by the processing unit 400. For example, the storage unit 410 may be a read-only memory (ROM), a flash memory, a random access memory (RAM), a hard disk, an optical data storage device, and a non-volatile storage unit, etc., but is not limited thereto.
[0053] The network device 40 is used to represent the necessary elements required to implement the embodiment of the present invention, and those skilled in the art may make different modifications and adjustments accordingly, without limitation thereto. For example, when the centralized network configuration server 100 is implemented with the network device 40, the process 30 may be compiled into a program code 412 and stored in the storage unit 410, and the routing path determination method is executed by the processing unit 400. In addition, the storage unit 410 is also used to store the data and parameters required when the routing path determination method is running, without limitation thereto.
[0054] In summary, the present invention provides a routing path determination method and device, which can simultaneously consider various traffic type data flows and plan survivable routing paths in a limited time under an online state, thereby solving the problems faced in the art.
[0055] The above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made according to the scope of the claims of the present invention should fall within the scope of the present invention. Description of Reference Numerals
[0056] 10: Time-Sensitive Networking Systems 100:Centralized network configuration server 110:Centralized User Configuration Server 120, 122: Terminal device 130, 132: Time-Sensitive Network Bridge 20: Network Topology ED1, ED2: terminal device BR1~BR7: Time-Sensitive Network Bridge ST1: Data Flow P1~P3: Path 30: Process 300~314: Steps 40: Network Devices 400: Processing unit 410: Storage unit 412: Program code
Claims
1. A routing path determination method, It is characterized in that For a time-sensitive network system, the time-sensitive network system includes a plurality of background data flows, and the routing path determination method sequentially includes: (a) receiving a request to allocate a first plurality of data streams; (b) determining a first plurality of candidate mutually exclusive path pairs for each data flow of the first plurality of data flows according to a network topology of the time-sensitive network system; (c) initializing a plurality of final routing path pairs for the first plurality of data flows and the plurality of background data flows according to a plurality of first lengths of the plurality of candidate mutually exclusive path pairs; (d) selecting a plurality of temporary routing path pairs for the first plurality of data flows and the plurality of background data flows according to a first plurality of parameters; (e) updating the first plurality of parameters according to the plurality of temporary routing path pairs; and (f) comparing the plurality of final routing path pairs with the plurality of temporary routing path pairs, and updating the plurality of final routing path pairs with the plurality of temporary routing path pairs when the plurality of temporary routing path pairs are better than the plurality of final routing path pairs; The time-sensitive network system transmits the first plurality of data flows and the plurality of background data flows according to the plurality of final routing paths.
2. The routing path determination method according to claim 1, It is characterized in that The method also includes: before a timer expires, returning to step (d) after executing step (f).
3. The routing path determination method according to claim 1, It is characterized in that Step (b) comprises: Obtaining a first plurality of mutually exclusive path pairs for each data stream of the first plurality of data streams according to a k-pair mutually exclusive path algorithm; and The first plurality of candidate mutually exclusive path pairs for each data stream of the first plurality of data streams are determined according to the plurality of lengths of the first plurality of mutually exclusive path pairs for each data stream of the first plurality of data streams.
4. The routing path determination method according to claim 3, It is characterized in that The length of each of the plurality of lengths of the first plurality of mutually exclusive path pairs of each data stream of the first plurality of data streams is the sum of a path length of each path of each of the mutually exclusive path pairs.
5. The routing path determination method according to claim 4, It is characterized in that The path length of each path is a hop count of each path.
6. The routing path determination method according to claim 1, It is characterized in that Also included: Before receiving the request, a current mutually exclusive path pair for transmitting each data flow of the plurality of background data flows is selected from a second plurality of candidate mutually exclusive path pairs for each data flow of the plurality of background data flows.
7. The routing path determination method of claim 6, wherein step (c) is to initialize the plurality of final routing path pairs according to the first plurality of lengths of the first plurality of candidate mutually exclusive path pairs for each data flow of the first plurality of data flows and the current mutually exclusive path pair for each data flow of the plurality of background data flows.
8. The routing path determination method according to claim 6, It is characterized in that Step (d) is to select the plurality of temporary routing path pairs from the first plurality of candidate mutually exclusive path pairs of each data flow of the first plurality of data flows and the second plurality of candidate mutually exclusive path pairs of each data flow of the plurality of background data flows.
9. The routing path determination method according to claim 1, It is characterized in that The first plurality of parameters include a plurality of pheromone values for the plurality of background data streams and the first plurality of data streams according to an ant colony optimization algorithm.
10. The routing path determination method according to claim 9, It is characterized in that Step (d) is to select the plurality of temporary routing path pairs according to a switching probability, wherein the switching probability is determined according to the first plurality of parameters and the second plurality of parameters for the plurality of background data flows and the first plurality of data flows.
11. The routing path determination method according to claim 10, It is characterized in that The second plurality of parameters include a plurality of visibility values for the plurality of background data streams and the first plurality of data streams according to the ant colony optimization algorithm.
12. The routing path determination method according to claim 11, It is characterized in that The plurality of visibility values are calculated as follows: A visibility value of a mutually exclusive path pair for a time-sensitive network data flow is the inverse of the exponential length of the mutually exclusive path pair; A visibility value for a mutually exclusive path pair for an audio video bridge data stream is the inverse of the worst case delay of the mutually exclusive path pair; and A visibility value of each current mutually exclusive path pair of the plurality of background data flows is increased according to a preference parameter.
13. The routing path determination method according to claim 9, It is characterized in that Step (e) is to update the plurality of pheromone values according to an evaporation coefficient of the ant colony optimization algorithm, and to increase the pheromone values corresponding to the plurality of temporary routing path pairs in the plurality of pheromone values.
14. The routing path determination method according to claim 1, It is characterized in that Step (f) is to compare according to an objective function, and the plurality of temporary routing path pairs are better than the plurality of final routing path pairs if a value of the objective function of the plurality of temporary routing path pairs is smaller than a value of the objective function of the plurality of final routing path pairs.
15. The routing path determination method according to claim 14, It is characterized in that The objective function includes at least one of the following parameters: A number of unschedulable time-sensitive network data flows; a number of unschedulable audio video bridge streams; An amount by which the plurality of background data flows are rescheduled; as well as The sum of the delays of all Audio Video Bridge streams.
16. The routing path determination method according to claim 1, It is characterized in that The path determination method is a path determination method that supports time-sensitive network data streams and audio video bridging data streams.
17. A network device, It is characterized in that Used in a time-sensitive network system, the time-sensitive network system includes a plurality of background data streams, and the network device includes: a processing unit for executing a program code; as well as A storage unit, coupled to the processing unit, is used to store the program code to instruct the processing unit to execute the routing path determination method as described in any one of claims 1-16.
18. The network device according to claim 17, It is characterized in that The network device is a centralized network configuration server of the time sensitive network.
19. The network device according to claim 18, It is characterized in that The request of the routing path determination method is received from a centralized user configuration server of the time-sensitive network system.
20. The network device according to claim 18, It is characterized in that The step of transmitting the first plurality of data streams and the plurality of background data streams according to the plurality of final routing paths by the time-sensitive network system further comprises: The centralized network configuration server configures a plurality of time-sensitive network bridges connected to the centralized network configuration server according to the plurality of final routing path pairs, wherein the plurality of final routing path pairs pass through the plurality of time-sensitive network bridges.
