A bus network dynamic reconfiguration system supporting multi-priority traffic

Abstract

The application relates to a bus network dynamic reconstruction system supporting multi-priority flow, and belongs to the technical field of wired communication network resource dynamic scheduling and planning. The system comprises the following: a network perception and management module, which is used for acquiring topology information and network flow information of the bus network in real time; an end node demand processing module, which is used for identifying the demand type of a source end node, calling a control center subsystem to solve a corresponding configuration scheme based on the demand type, the topology information and the network flow information; and the control center subsystem, which is used for solving the configuration scheme and dispatching the configuration scheme to corresponding switches and target end nodes when the solving is successful. The system of the application introduces a feasibility prediction unit to preferentially select a new TT flow admission configuration scheme which does not affect the end-to-end delay constraint of RC flows in the network, introduces an elastic reconstruction module to elastically adjust the bandwidth utilization of RC flows to weaken the influence of the new TT flow on the quality of service of the RC flows, and solves the RC flow packet loss and communication interruption problems caused by the admission of the new TT flow.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of wired communication network resource dynamic scheduling and planning, and particularly relates to a bus network dynamic reconstruction system supporting multi-priority traffic. BACKGROUND

[0002] In order to save costs and benefit from the rapid development of the increasingly widespread and perfect Ethernet technology, the trend of merging bus networks of various integrated electronic systems such as aerospace, aviation and vehicles with Ethernet is very strong. This trend is reflected in the activities of standardization organizations, such as the IETF Deterministic Network (DetNet) plan and the IEEE Time Sensitive Network (TSN). Software-defined networks (SDN) are more introduced into real-time control systems, further enhancing the explicit control of networks on traffic and promoting the research on dynamic planning and configuration of time-triggered network traffic that needs static verification and deployment in the past.

[0003] However, even under the premise of passing the scheduling verification, the reconfiguration scheme of TT flow (time-triggered traffic) (such as the addition of new traffic) can cause changes in switch port load and network bandwidth resources, thereby affecting the end-to-end delay of rate-constrained low-priority traffic (RC flow), and even causing them to miss the worst deadline, causing resource waste or more serious consequences, thereby triggering the secondary configuration of RC flow, which is not conducive to maintaining the stability and reliability of network traffic communication. Therefore, how to generate a more optimized new TT flow admission configuration scheme to avoid the transmission time of low-priority traffic exceeding the end-to-end delay deadline and causing secondary configuration is a problem that needs to be solved urgently. SUMMARY

[0004] In view of the above analysis, the application aims to provide a bus network dynamic reconstruction system supporting multi-priority traffic, which introduces a feasibility prediction unit to preferentially select a new TT flow admission configuration scheme that does not affect the end-to-end delay constraints of all low-priority traffic (RC flow) in the network, and introduces an elastic reconstruction module to elastically adjust the bandwidth utilization of low RC flow through the solution and dispatch of RC flow compression scheme and RC flow decompression scheme, thereby weakening the impact of new TT flow on the quality of service of RC flow and avoiding packet loss and communication interruption.

[0005] The bus network dynamic reconstruction system supporting multi-priority traffic of the application comprises:

[0006] A network perception and management module is configured to acquire topology information and network traffic information of the bus network in real time.

[0007] The end node demand processing module is configured to identify a demand type of a source end node, wherein the demand type comprises a plurality of new TT flow admission requests of the source end node and a traffic decommission notification of a TT flow; and based on the demand type, the topology information and network flow traffic information, the control center subsystem is called to solve a configuration scheme corresponding to each demand type, and if the solution is successful, the corresponding configuration scheme is fed back to the source end node, and if the solution fails, when the demand type is the plurality of new TT flow admission requests, the source end node is fed back that the admission request is not passed, and when the demand type is the traffic decommission notification of the TT flow, only the end node is fed back the failure result.

[0008] The control center subsystem is configured to solve the configuration scheme and dispatch the configuration scheme to corresponding switches and target end nodes when the solution is successful.

[0009] Further, the control center subsystem executes the following process to solve the configuration scheme:

[0010] When the demand type is the plurality of new TT flow admission requests, a first configuration scheme of the plurality of new TT flows satisfying that the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay limit is solved, and if the first configuration scheme fails to be solved, a second configuration scheme of the plurality of new TT flows and RC flows satisfying that the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay limit is solved.

[0011] When the demand type is the traffic decommission notification of the TT flow, if there is an RC flow in a compressed state in the network, a third configuration scheme of the RC flow decompression scheme satisfying that the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay limit is solved.

[0012] Further, the control center subsystem comprises:

[0013] The routing solving module is configured to solve a first routing scheme of the plurality of new TT flows satisfying that the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay limit based on the plurality of new TT flow admission requests.

[0014] The scheduling solving module is configured to solve a first scheduling scheme based on the first routing scheme, and the first scheduling scheme successfully solved is taken as the first configuration scheme.

[0015] an elastic reconstruction module configured to solve a second scheduling scheme of the multiple new TT flows with the least number of RC flows whose end-to-end worst-case latency exceeds the corresponding end-to-end latency deadline when solving the first routing scheme or the first scheduling scheme fails, solve a second configuration scheme based on the second scheduling scheme if the solving fails, solve a partial RC flow compression scheme that satisfies all RC flows whose end-to-end worst-case latency does not exceed the corresponding end-to-end latency deadline based on the second scheduling scheme if the solving succeeds, and take the partial RC flow compression scheme and the second scheduling scheme as the second configuration scheme if the solving succeeds; and solve a decompression scheme of the RC flows as a third configuration scheme when the demand type is a traffic withdrawal notification of the TT flow;

[0016] a comprehensive interface module configured to dispatch the first, second or third configuration scheme to a corresponding switch, target end node and end node demand processing module respectively, and feed back that the admission request is not passed to the end node demand processing module when the second configuration scheme fails, and feed back the failure result to the end node demand processing module when the third configuration scheme fails;

[0017] Further, the elastic reconstruction module solving the partial RC flow compression scheme that satisfies all RC flows whose end-to-end worst-case latency does not exceed the corresponding end-to-end latency deadline based on the second scheduling scheme further includes: taking the second scheduling scheme as the second configuration scheme if the solving of the partial RC flow compression scheme fails, and feeding back the second configuration scheme and a traditional RC traffic reconstruction method for RC flows whose worst-case latency exceeds the corresponding end-to-end latency deadline to the source end node through the comprehensive interface module.

[0018] Further, the routing solving module includes:

[0019] a routing pool solving unit configured to calculate a set of optional routing schemes based on the multiple new TT flow admission requests and the topology information;

[0020] a routing distribution unit configured to solve a preferred routing scheme based on the set of optional routing schemes within a maximum number of iterations based on the network traffic information, with the goal of optimizing the delay time rate of each RC flow affected by the new TT flow;

[0021] a feasibility prediction unit configured to determine whether the preferred routing scheme obtained in each round of iteration is feasible, and if the end-to-end worst-case latency of each RC flow does not exceed the corresponding end-to-end latency deadline, the preferred routing scheme is determined to be feasible and is taken as the first routing scheme, otherwise it is not feasible and the next round of iteration is returned to the routing distribution unit.

[0022] Further, the method for calculating the end-to-end worst-case latency of each RC flow by the feasibility prediction unit includes:

[0023] calculating worst transmission intervals of each RC flow between each two nodes of network links based on the preferred routing scheme, comprising:

[0024]

[0025] wherein, represents the worst time interval for transmission of any frame of the RC flow from network node v α to network node v β ; and respectively represent the time for transmission of any frame of the RC flow to network node v β and v α ; represents the technical delay introduced by each data frame; represents the delay impact of TT flows on the RC flow, wherein F TT represents a set of other TT flows except the new TT flow, f k represents other TT flows except the new TT flow; represents the time offset between network node v α and network node v β for transmission of any frame of the RC flow, l k represents the data frame time length of the kth new TT flow, p k represents the flow period of the kth new TT flow; represents the delay impact of other RC flows on the RC flow, wherein F RC represents a set of RC flows, f l represents the new TT flow, F TTRec represents a set of the new TT flows, l l represents the frame time length of the lth other RC flow, p k represents the flow period of the lth other RC flow; represents the additional delay of the RC flow due to the preemption integration strategy of TT flows; l i represents the frame time length of the RC flow;

[0026] calculating the worst case delay of the corresponding RC flow based on all worst transmission intervals of each RC flow.

[0027] Further, the elastic reconstruction module comprises:

[0028] a TT flow planning and distribution unit, configured to, when the first routing scheme fails to be solved or the first scheduling scheme fails, select, based on the multiple preferred routing schemes iteratively obtained by the routing distribution unit, a preferred routing scheme in which the number of RC flows whose end-to-end worst case delay exceeds the corresponding end-to-end delay deadline is as few as possible as a second routing scheme, and call the scheduling solving unit to solve a second scheduling scheme based on the second routing scheme.

[0029] a RC flow compression solving unit configured to solve a partial RC flow compression scheme based on the second scheduling scheme, wherein the partial RC flow compression scheme satisfies all RC flows whose end-to-end worst-case latency does not exceed the corresponding end-to-end latency deadline;

[0030] a reconstruction scheme dispatching unit configured to send the partial RC flow compression scheme and the second scheduling scheme to the integration interface module as a second configuration scheme when the solving of the second routing scheme and the partial RC flow compression scheme is successful, or send the second scheduling scheme to the integration interface module as a second configuration scheme when the solving of the second routing scheme is successful but the solving of the partial RC flow compression scheme fails, and implement a traditional RC flow reconstruction method for the RC flow whose end-to-end worst-case latency exceeds the corresponding end-to-end latency deadline, or send a second configuration scheme solving failure result to the integration interface module when the solving of the second routing scheme fails.

[0031] Further, the RC flow compression solving unit solving a partial RC flow compression scheme based on the second scheduling scheme, wherein the partial RC flow compression scheme satisfies all RC flows whose end-to-end worst-case latency does not exceed the corresponding end-to-end latency deadline, comprises:

[0032] determining a set of RC flows to be solved based on the RC flow whose end-to-end worst-case latency exceeds the corresponding end-to-end latency deadline and the RC flow whose routing shares the same directed link with the RC flow;

[0033] numbering the corresponding compression scheme of each RC flow in the set of RC flows to be solved;

[0034] combining each corresponding compression scheme as a particle, combining the corresponding numbers of each corresponding compression scheme in each particle to describe the particle position, and randomly initializing the particle swarm based on the corresponding compression scheme of each RC flow in the set of RC flows to be solved;

[0035] constructing a fitness function based on the end-to-end latency and the maximum bandwidth constraint of all affected RC flows;

[0036] solving the optimal solution of the particle swarm based on the fitness function, and obtaining the partial RC flow compression scheme.

[0037] Further, the fitness function of the RC flow compression solving unit is expressed as:

[0038] F(X) = Q(X) + γE(X);

[0039] wherein Q(X) is an objective function, Q(X) = max{D ave -min{δ}}; D avedenotes the average delay time ratio of all RC flows; min{δ} denotes the minimum value of the delay of all affected RC flows caused by compression that can meet the end-to-end delay not exceeding the corresponding end-to-end delay deadline, δ satisfies where F RC denotes the set of RC flows to be solved, f h denotes the hth RC flow in the set, denotes the offset of the hth RC flow to the destination, denotes the initial offset of the data stream route vl h of the kth RC flow, dl h denotes the end-to-end delay deadline of the hth RC flow;

[0040] γ is a penalty weight;

[0041] E(X) is a penalty function,

[0042]

[0043] where, denotes the number of affected RC flows; vl j denotes the route of the jth said to-be-admitted TT flow or RC flow in the network; e αβ denotes the directed edge connecting the network nodes v α , v β ; denotes the worst time interval for transmission of any frame of the RC flow from the network node v α to the network node v β ; dl i denotes the end-to-end delay deadline of the ith RC flow; E denotes the set of directed communication links between the nodes of the network topology; e denotes the directed communication link between the nodes of the network topology; l j denotes the data stream frame length of the jth to-be-admitted TT flow or RC flow in the network; p j denotes the data stream period of the jth to-be-admitted TT flow or RC flow in the network; C j denotes the maximum bandwidth service rate of the directed link; denotes the expected to-be-admitted TT flow and RC flow in the network to meet the maximum bandwidth constraint.

[0044] Further, the elastic reconstruction module further comprises:

[0045] a decompression RC flow identification unit, configured to determine, based on the set of compressed RC flows existing in the network and the set of RC flows involved in the decompression scheme, the set of RC flows to be solved;

[0046] An RC flow decompression solving unit is configured to solve, based on a set of to-be-solved decompression schemes of RC flows, the decompression schemes of the RC flows that satisfy the end-to-end worst-case delay of each RC flow not exceeding a corresponding end-to-end delay deadline;

[0047] A decompression scheme dispatching unit is configured to send the decompression scheme of the RC flow that is successfully solved as a third configuration scheme to the integration interface module.

[0048] The present application can achieve at least one of the following advantages:

[0049] By considering the influence of the new TT flow on the RC flow in the network, the configuration scheme of the plurality of new TT flows that satisfies the end-to-end worst-case delay of each RC flow not exceeding a corresponding end-to-end delay deadline is solved based on the admission request of the new TT flow, and the second configuration scheme of the plurality of new TT flows and RC flows that satisfies the end-to-end worst-case delay of each RC flow not exceeding a corresponding end-to-end delay deadline is solved when the solving fails, thereby reducing the influence on the existing RC flow in the network and avoiding the problem of packet loss and communication interruption of the RC flow caused by secondary configuration from the source.

[0050] By introducing the feasibility prediction unit, the new TT flow admission configuration scheme that does not affect the end-to-end delay constraint of the RC flow in the network is preferentially selected, the bandwidth utilization of the RC flow is flexibly adjusted by introducing the elastic reconstruction module to weaken the influence of the new TT flow on the quality of service of the RC flow, the change of the bandwidth is exchanged for the uninterrupted communication of the RC flow, the robustness and determinacy of the network are ensured, the secondary reconstruction of the RC flow is avoided as much as possible, and the problems of packet loss and communication interruption of the RC flow are solved. By decompressing the RC flow when receiving the TT flow network exit notification, the network bandwidth resources are fully utilized, and the influence on the transmission quality of the RC flow is further reduced.

[0051] By constructing the fitness function based on the delay time and the maximum bandwidth constraint of the RC flow, the particle swarm algorithm is used to solve the first routing scheme, the second routing scheme, the partial RC flow compression scheme and the decompression scheme of the RC flow, respectively, so as to ensure that the optimal solution of the corresponding scheme is obtained at each link when each module of the system solves the configuration scheme.

[0052] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art upon examination of the following or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and attained by the instrumentalities expressed in the specification, claims and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0053] The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings:

[0054] Figure 1 Fig. 1 is a schematic diagram of a system architecture of the present application;

[0055] Figure 2 Fig. 2 is a flowchart of a process for solving the first and second configuration schemes of the present application. DETAILED DESCRIPTION

[0056] The preferred embodiments of the present application will be described in detail with reference to the drawings, in which:

[0057] One specific embodiment of the present application discloses a bus network dynamic reconfiguration system supporting multi-priority traffic, as shown in Fig. 1, which specifically comprises: Figure 1

[0058] a network perception and management module for acquiring topology information and network traffic information of the bus network in real time;

[0059] an end node demand processing module for identifying a demand type of a source end node, the demand type including a plurality of new TT flow admission requests of the source end node and a traffic off-network notification of a TT flow, and for calling a control center subsystem to solve a configuration scheme corresponding to each demand type based on the demand type, the topology information and the network traffic information, and feeding back the corresponding configuration scheme to the source end node if the solution is successful, and feeding back an admission request failure to the source end node if the solution fails and the demand type is the plurality of new TT flow admission requests, and feeding back only a failure result to the end node if the demand type is the traffic off-network notification of the TT flow;

[0060] a control center subsystem for solving the configuration scheme and dispatching the configuration scheme to corresponding switches and target end nodes if the solution is successful, and for solving the configuration scheme by executing the following process: when the demand type is the plurality of new TT flow admission requests, solving a first configuration scheme of the plurality of new TT flows satisfying that the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay deadline; if the first configuration scheme fails to be solved, solving a second configuration scheme of the plurality of new TT flows and RC flows satisfying that the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay deadline; and when the demand type is the traffic off-network notification of the TT flow, if there is an RC flow in a compressed state in the network, solving an RC flow decompression scheme satisfying that the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay deadline as a third configuration scheme.

[0061] ​The embodiment discloses a bus network dynamic reconstruction system supporting multi-priority flow, solves a configuration scheme of a plurality of new TT flows meeting the end-to-end worst case delay of each RC flow in the network not exceeding the corresponding end-to-end delay deadline based on a plurality of new TT flow admission requests, and solves a second configuration scheme of the plurality of new TT flows and RC flows meeting the end-to-end worst case delay of each RC flow not exceeding the corresponding end-to-end delay deadline when the solving fails, so as to reduce the influence on the existing RC flow in the network as much as possible, and avoid the problem of packet loss communication interruption of the RC flow caused by secondary configuration from the source.

[0062] Network perception and management module

[0063] Specifically, the network perception and management module comprises an OFDP and ARP data packet analysis unit, a Flowtable unit, a health management unit and a network state storage unit.

[0064] Further, the OFDP and ARP data packet analysis unit is used for obtaining network topology information in real time based on the OFDP data packet and the copy of the ARP data packet containing the switch ID of the switch at each network port of the switch;

[0065] The Flowtable unit is used for communicating with the switch of the data plane to obtain network flow information;

[0066] The health management unit periodically guarantees the continuity of the communication between the network perception and management module and the data plane, the real-time of the topology information and the network flow information through the handshaking protocol;

[0067] The network state storage unit is used for updating and storing the current topology information and network flow information in real time.

[0068] It should be noted that the admission control and planning of the multi-priority flow in the bus network dynamic reconstruction system supporting multi-priority flow adopt the TAS scheduling model based on IEEE 802.1Qbv, and the flow comprises time triggered (TT) flow, rate limited (RC) flow (such as audio and video AVB flow) and traditional Ethernet best effort (BE) flow. In the application, only the TT flow and the RC flow are considered. In order to be not lost in generality, the RC flow in the application is only periodically sent at the source end system, and no additional scheduling algorithm is executed in the port of the passing switch, and only the basic first-in-first-out principle (FIFO) is followed. This assumption is reasonable, and is downward compatible with more complex scheduling algorithms, and therefore has the most extensive application range.

[0069] Specifically, the TSN network topology is denoted as a directed graph G(V, E), where V represents the set of end system (ES) and switch (SW) network nodes, and E represents the set of directed communication links between network nodes. Let v1 and v2 represent two adjacent nodes, and satisfy v1, v2 ∈ V, then [v1, v2] and [v2, v1] can be used to represent a bidirectional data flow link. The edge e st is described as e st = [v s ,v t ] ∈ E, v s ,v t ∈ V. The multi-priority traffic in time sensitive network (TSN) mainly includes TT, rate-constrained (RC) and best effort (BE) traffic. TT traffic is forwarded according to an offline predetermined schedule, and has the highest priority. RC traffic is periodically transmitted at the source end system, and is rate-limited by a bandwidth allocation gap (BAG), and has a lower priority than TT traffic. BE traffic is a traditional Ethernet traffic without any QoS guarantee, and has the lowest priority, which is not discussed in the present application.

[0070] Further, all communication tasks in the network are transmitted through data flows, including data frame instances. Let F = F TT ∪ F RC represent the set of all data flows. Any TT flow f m ∈ F TT and any RC flow f n ∈ F RC can be represented by the following seven-tuple and six-tuple, respectively:

[0071] f m = {s m ,d m ,l m ,p m ,dl m ,vl m ,O m} and f n = {s n ,d n ,l n ,p n ,dl n ,vl n};

[0072] where s m , d m , l m , p m , dl m and vl mrespectively represent source ES, destination ES, frame length of time dimension (frame length divided by link bandwidth), traffic period, end-to-end (ETE) latency deadline and routing information. m TT flow only, representing the offset set between the first data frame instance on each directed link and the period start time, which collectively constitute the time schedule of the network. In short, the admission control and dynamic planning of TT flow means re-allocating the routing information and corresponding offset time of TT flow

[0073] End node demand processing module

[0074] Specifically, the end node demand processing module includes a user demand identification unit, a user demand processing unit and a user demand response unit.

[0075] Further, the user identification demand unit is used to identify the demand type of the source end node. As mentioned before, only TT flow and RC flow are considered in the present application, and the demand type of the source end node in the present application includes multiple new TT flow admission requests of the source end node and traffic off-network notification of TT flow.

[0076] The user demand processing unit is used to call the configuration scheme corresponding to each demand type solved by the control center subsystem based on the demand type, the topology information and the network flow information. It should be noted that the calling sequence of each unit in the control center subsystem is different based on different demand types. When the demand type is multiple new TT flow admission requests, the user demand processing unit first calls the routing solving module of the control center subsystem, and when the demand type is traffic off-network notification of TT flow, the user demand processing unit first calls the elastic reconstruction module of the control center subsystem. The present application will be further described in detail when introducing each module of the control center subsystem (the calling sequence between modules is shown in Figure 2 Further, the user demand processing unit obtains the topology information and network flow information from the network state storage unit of the network perception and management module when calling the control center subsystem.

[0077] The user demand response unit is used to feed back the configuration scheme corresponding to each demand type solved by the control center subsystem to the source end node, and if the solving fails, when the demand type is multiple new TT flow admission requests, the user demand response unit feeds back the admission request to the source end node, and when the demand type is traffic off-network notification of TT flow, the user demand response unit only feeds back the failure result to the end node.

[0078] Control center subsystem

[0079] Specifically, the control center subsystem includes a routing solving module, a scheduling solving module, an elastic reconstruction module and a comprehensive interface module.

[0080] Route solving module

[0081] The routing solution module is used to solve a first routing scheme for the multiple new TT flow admission requests, ensuring that the worst-case end-to-end latency of each RC flow does not exceed the corresponding end-to-end latency period. That is, when the user identification request module in the end node request processing module identifies a request type of multiple new TT flow admission requests, the user request processing unit first calls the routing solution module in the control center subsystem to solve for the first routing scheme (e.g., ...). Figure 2 (As shown).

[0082] Specifically, the routing solution module includes a routing pool solution unit, a routing allocation unit, and a feasibility prediction unit.

[0083] Furthermore, the routing pool solving unit is used to calculate multiple alternative routing schemes based on multiple new TT flow admission requests and topology information using the Dijkstra algorithm.

[0084] The routing allocation unit is used to optimize the latency rate of each RC flow affected by the new TT flow based on the network traffic information, and to iteratively solve for the optimal routing scheme using the particle swarm optimization algorithm within the maximum number of iterations.

[0085] Specifically, the process of using the particle swarm optimization algorithm to find the optimal route includes processes s1 to s4.

[0086] s1, Constructing particles.

[0087] Specifically, the available routing schemes are numbered from 1 to k.

[0088] Furthermore, the combination of routing schemes for each TT flow to be admitted is used as a particle, and the particle position is described by the combination of the corresponding numbers of the routing schemes for each TT flow in each particle. The position of each particle is represented as an M-dimensional vector X = [x1, x2, ..., x...]. M ], where x m (m∈[1,M]) represents the pending TT flow f. m The routing scheme number is determined based on the range of available routes, and its value is an integer from 1 to k. The particle's velocity is represented as V = [v1, v2, ..., v...]. M ], where v m This represents the distance the particle moves in the m-th dimension, and its value ranges from -(k-1) to (k-1).

[0089] s2. Construct the objective function and penalty function.

[0090] Specifically, the objective function is constructed based on the delay rate of the affected RC stream.

[0091] The objective function is expressed as:

[0092]

[0093] where Q(X) is the objective function for calculating the delay time rate of affected RC flows; X is a c-dimensional vector representing particle positions, X = [x1, x2,..., x m ]; c is the number of TT flows to be admitted; represents the number of affected RC flows; represents the worst end-to-end delay of the i-th RC flow, where vl i represents the route e αβ of the i-th RC flow, α β represents the directed edge connecting network nodes v represents the worst time interval for the RC flow to transmit any frame from network node v α to network node v β ; dl i represents the end-to-end delay limit of the i-th RC flow; D ave represents the average delay time rate of all RC flows, F RC | represents the total number of RC flows in the network.

[0094] Further, a penalty function is constructed based on the maximum bandwidth constraint.

[0095] The penalty function is expressed as:

[0096]

[0097] where E represents a set of directed communication links between network topology nodes; vl j represents the route of the j-th TT flow to be admitted or RC flow in the network; e represents a directed communication link between network topology nodes; l j represents the data flow frame length of the j-th TT flow to be admitted or RC flow in the network; p j represents the data flow period of the j-th TT flow to be admitted or RC flow in the network; C j represents the maximum bandwidth service rate of the directed link; represents the expectation that the TT flow to be admitted and the RC flow in the network satisfy the maximum bandwidth constraint.

[0098] s3, construct a fitness function.

[0099] Specifically, the fitness function is constructed based on the objective function, the penalty function, and the weight coefficient of the penalty function, for evaluating the performance of the routing scheme represented by the particle position. The lower the fitness function value, the better the performance of the selected route. ​

[0100] The fitness function is expressed as: F(X)=Q(X)+γE(X); wherein, γ is a penalty weight, which is different according to different topologies and flows, and is preferably 300.

[0101] s4, iteratively solving.

[0102] Specifically, in each iteration, the optimal solution of the particle swarm is obtained based on the fitness function, and the preferred routing scheme of the iteration is obtained and recorded.

[0103] Specifically, in the process of searching for the optimal solution, the position and speed of the particle are updated according to two values: the current optimal solution pBest of the individual particle and the optimal solution gBest of all particles in the population; the search process is expressed as:

[0104] V t+1 =ωV t +c1r1(pBest-X t )+c2r2(gBest-X t );

[0105] X t+1 =X t +V t ;

[0106] Wherein, ω is an inertia weight, used to adjust the degree to which the particle maintains its original speed, and the value range is between 0-1, and is preferably set to 0.9; c1 and c2 represent self-learning and social learning factors, respectively, used to balance the local and global search ability, c1 controls the speed and direction of the particle in the search space to update itself, the greater the value, the more inclined to search along the direction of its own historical optimal solution, and the smaller the value, the more inclined to random exploration; c2 controls the speed of the particle in the space to update according to the optimal solution of the neighbor particle, the greater the value, the more inclined to search affected by the neighbor particle, and the smaller the value, the more inclined to independent search; the sum of c1 and c2 is not greater than 4, and preferably, c1 and c2 are set to 2 and 0.9, respectively; r1 and r2 are random numbers between 0 and 1; t represents the search number in the solving process.

[0107] Further, the feasibility prediction unit is used to determine whether the preferred routing scheme obtained by the routing allocation unit in each iteration is feasible, if the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay limit, the preferred routing scheme is determined to be feasible and is taken as the first routing scheme, otherwise it is not feasible, and the routing allocation unit is returned for the next iteration.

[0108] Specifically, the calculation method of the end-to-end worst-case delay of each RC flow is:

[0109] calculating worst transmission intervals of each RC flow between each pair of nodes in the network link based on the preferred routing scheme;

[0110] calculating worst-case delay of each RC flow based on all worst transmission intervals of each RC flow.

[0111] Further, the calculating worst transmission intervals of each RC flow between each pair of nodes in the network link based on the preferred routing scheme comprises:

[0112]

[0113] wherein, represents the worst time interval for transmission of any frame of the RC flow from network node v α to network node v β ; and respectively represent the time for transmission of any frame of the RC flow to network node v β and v α ; represents the technical delay introduced by each data frame; represents the delay impact of TT flows on the RC flow, wherein F TT represents the set of other TT flows that are not to be admitted, f k represents the other TT flows that are to be admitted, represents the time offset between transmission of any frame of the RC flow from network node v α to network node v β , l k represents the data frame time length of the kth TT flow to be admitted, p k represents the flow period of the kth TT flow to be admitted; represents the delay impact of other RC flows on the RC flow, wherein F RC represents the set of RC flows, f l represents the TT flow to be admitted, F TTRec represents the set of TT flows to be admitted, l l represents the frame time length of the lth other RC flow, p k represents the flow period of the lth other RC flow; represents the additional delay of the RC due to adoption of the preemption integration strategy; l i represents the frame time length of the RC flow.

[0114] The following explains the principle of calculating the worst transmission intervals of each RC flow between each pair of nodes in the network link:

[0115] In the embodiment method, it is reasonably assumed that all RC flows have passed the worst-case end-to-end delay verification in the design stage, i.e. wcdn ≤dl n (wcd n represents the worst case end-to-end delay of the nth RC flow). When judging whether the end-to-end worst case delay of each RC flow exceeds the corresponding end-to-end delay limit, only the RC flows intersecting with the route of the to-be-admitted TT flow need to be judged. This is because in the delay analysis, the end-to-end delay of non-intersecting RC flows may change, but the worst case end-to-end delay wcd does not change (regardless of the delay analysis method such as network calculus or trajectory method).

[0116] Further, the end-to-end delay of the RC flow is mainly affected by four aspects in addition to the transmission delay: ① related technology delay, such as delay caused by inter-node data validity and integrity check, bandwidth allocation interval error containment delay, etc.; ② delay caused by the competition of bandwidth resources between RC flows; ③ delay caused by high-priority TT flow to RC flow; ④ delay caused by TT and RC flow integration, which is mainly divided into three kinds of shuffling, preemption and timely blocking according to different integration strategies. In the method of the present application, the most common preemption strategy is adopted (this strategy is the most general, representative and practical).

[0117] Further, the end-to-end delay of the RC flow mainly consists of two parts: ① queuing delay at the source ES; ② delay of each intermediate node (link).

[0118] First, consider the queuing delay at the source ES. Specifically, the set of RC flows intersecting with the route of the to-be-admitted TT flow is denoted as F RC ′. Since the initial phase (offset) of the route corresponding to the RC flow can be allocated by a traditional offset allocation algorithm, for the initial offset of vl i is denoted as Further, the transmission time (denoted as ) of the RC flow f j leaving the end node into the network can be calculated as follows:

[0119]

[0120] where FIFO(j,i) represents that the data frame of the TT flow f j enters the buffer earlier than the data frame of the RC flow f i ; when FIFO(j,i) is an empty set, Therefore, the offset of the RC flow f i entering the network can be represented by several possible discrete values.

[0121] Secondly, the delay calculation of each intermediate node (link) is considered. Specifically, the present application adopts an iterative strategy of overall timing analysis, and uses the concept of busy period to analyze the delay of the affected RC flow in the network. For any data frame of the RC flow, its busy period on a certain link can be defined as a time interval, i.e. the time interval from the frame arriving at node v α to the frame transmitting to the next node v β is denoted as The time interval from the frame arriving at node v α to the frame transmitting to the next node v β is denoted as The end-to-end delay of the RC flow is denoted as To ensure that the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay deadline dl i , the worst-case time interval is described as (for any known ):

[0122]

[0123] wherein, represents the technical delay introduced by each data frame; represents the delay impact of the TT flow on the RC flow; represents the delay impact of other RC flows on the RC flow; represents the additional delay of the RC flow due to the adoption of the preemption integration strategy by the TT flow.

[0124] Since the above formula is a recursive function containing on both sides, the recursive tree method can be used to solve it. Thus, the obtained represents the worst-case end-to-end delay of the RC flow f αβ in link e i starting from time , which can guarantee the transmission of all TT flows and other accumulated RC flows arriving earlier in this time period. The obtained can be used as the starting time of the next link in v i to solve the busy period until is obtained.

[0125] Further, the judgment of whether the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay deadline can be expressed as:

[0126]

[0127] Further, if each RC flow end-to-end worst-case delay does not exceed the corresponding end-to-end delay deadline, it is determined that the preferred routing scheme is feasible, and the preferred routing scheme is taken as the first routing scheme, otherwise, it is determined that the preferred routing scheme is infeasible, and the routing allocation unit is returned for the next iteration, until the maximum iteration number solved by the routing allocation unit is reached, if the preferred routing scheme obtained by iteration is still infeasible, the first routing scheme fails to be solved.

[0128] The scheduling solving module

[0129] Specifically, the scheduling solving module is configured to solve a first scheduling scheme based on the first routing scheme, and take the successfully solved first scheduling scheme as the first configuration scheme.

[0130] The scheduling solving module includes a scheduling solving unit and a scheduling dispatching unit.

[0131] Further, the scheduling solving unit is configured to solve a first scheduling scheme satisfying the scheduling constraint based on the first routing scheme by using an integer linear programming solver, and the solver can be gurobi, cplex, etc. Since the solver and the solving method are prior art, they will not be discussed here.

[0132] It should be noted that, when the scheduling solving unit fails to solve the first scheduling scheme based on the first routing scheme, the routing allocation unit of the routing solving module is returned for continued iteration to solve the first routing scheme, until the maximum iteration number of the routing allocation unit is reached. It should be noted that, generally, the solving of the first scheduling scheme based on the first routing scheme is related to network load, and the failure probability is extremely small in actual application.

[0133] Further, the scheduling dispatching unit is configured to send the successfully solved first scheduling scheme to the comprehensive interface module as the first configuration scheme.

[0134] The elastic reconstruction module

[0135] Specifically, the elastic reconstruction module is configured to, when the first routing scheme or the first scheduling scheme fails to be solved, solve a second scheduling scheme of the plurality of new TT flows with the number of RC flows whose end-to-end worst-case delay exceeds the corresponding end-to-end delay deadline being as small as possible, if the solving fails, solve a second configuration scheme failure result, if the solving succeeds, solve a partial RC flow compression scheme satisfying all RC flows whose end-to-end worst-case delay does not exceed the corresponding end-to-end delay deadline based on the second scheduling scheme, and take the successfully solved partial RC flow compression scheme and the second scheduling scheme as the second configuration scheme; and when the demand type is a traffic off-network notification of a TT flow, solve a decompression scheme of the RC flow as a third configuration scheme.

[0136] The elastic reconstruction module comprises a TT flow plan allocation unit, an RC flow compression solving unit, a reconstruction scheme dispatching unit, a decompressed RC flow identifying unit, an RC flow decompression solving unit and a decompression scheme dispatching unit.

[0137] Further, the TT flow plan allocation unit is configured to, when the first routing scheme fails to be solved or the first scheduling scheme fails to be solved, select, based on the multiple optimal routing schemes obtained by the routing allocation unit, an optimal routing scheme in which the number of RC flows whose end-to-end worst-case delay exceeds the corresponding end-to-end delay deadline is as small as possible as a second routing scheme, and call the scheduling solving unit to solve a second scheduling scheme based on the second routing scheme.

[0138] It should be noted that, when the first routing scheme fails to be solved or the first scheduling scheme fails to be solved, it means that the first configuration scheme of the multiple new TT flows that satisfy the condition that the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay deadline cannot be solved, i.e., the first configuration scheme fails to be solved. In this case, the application considers compressing the RC flows already existing in the network and reducing the quality of service of part of the RC flows to exchange for their delay performance and communication continuity.

[0139] Further, the process of solving the second routing scheme by the TT flow plan allocation unit comprises:

[0140] Based on all the optimal routing schemes obtained by the routing allocation unit, the second routing scheme is solved iteratively, and in each round of iteration:

[0141] An optimal routing scheme in which the number of RC flows whose end-to-end delay exceeds the corresponding worst-case delay is the least is selected from the optimal routing schemes;

[0142] Based on the optimal routing scheme and the scheduling constraints, the scheduling solving unit is called to solve the second scheduling scheme by using an integer linear programming solver:

[0143] If successful, the optimal routing scheme is the second routing scheme, and the iteration is ended;

[0144] If failed, it is determined whether all the optimal routing schemes have been traversed, if not, the next round of iteration is continued, and if yes, it is indicated that the second scheduling scheme fails to be solved.

[0145] It should be noted that, in each round of iteration, the solving of the second scheduling scheme based on the optimal routing scheme selected in the round and the scheduling constraints is related to the network load, and in actual application, the failure probability of the solving of the second scheduling scheme is extremely small.

[0146] Further, if the second scheduling scheme fails to be solved, it is indicated that the second configuration scheme fails to be solved, and the TT flow plan allocation unit sends the failure result to the comprehensive interface module.

[0147] Further, when the second scheduling scheme is successfully solved, the RC flow compression solving unit is configured to solve a partial RC flow compression scheme based on the second scheduling scheme, wherein all RC flows in the partial RC flow compression scheme satisfy that the end-to-end worst-case latency of each RC flow does not exceed the corresponding end-to-end latency deadline.

[0148] Specifically, the process of solving the partial RC flow compression scheme by the RC flow compression solving unit includes steps s21-s25.

[0149] s21, determining a set of RC flows to be solved based on RC flows whose end-to-end worst-case latency exceeds the corresponding end-to-end latency deadline and RC flows whose routing shares the same directed link with the RC flows.

[0150] Specifically, based on the second scheduling scheme, RC flows whose corresponding end-to-end worst-case latency exceeds the corresponding end-to-end latency deadline and RC flows whose routing shares the same directed link with the RC flows are determined to form a set F RC The number of RC flows in the set F RC ' is denoted as M'.

[0151] s22, numbering the corresponding compression schemes of each RC flow in the set of RC flows to be solved, and the total number of period scheme numbers of each RC flow in the set F RC ' is denoted as k'.

[0152] s23, taking a combination of each of the corresponding compression schemes as a particle, and describing the position of the particle by a combination of the corresponding numbers of each of the corresponding compression schemes in each particle, and randomly initializing a particle group based on the corresponding compression schemes of each RC flow in the set F RC ' of RC flows to be solved.

[0153] Specifically, the position of each particle is represented as an M' dimensional vector X' = [x'1, x'2,..., x' M ], where x' m' (m' ∈ [1, M']) represents the compression scheme number of the RC flow f m' to be solved, and the moving range of the particle is -(k'-1) to (k'-1).

[0154] s24, constructing a fitness function based on the end-to-end latency and the maximum bandwidth constraint of all affected RC flows.

[0155] Specifically, the objective function is constructed based on the end-to-end latency of all affected RC flows. The objective function is represented as:

[0156] Q(X) = max{D ave -min{δ}};

[0157] Wherein, D avedenotes the average delay time ratio of all RC flows; min{δ} denotes the minimum value of all affected RC flows' delay caused by compression that can meet the end-to-end delay not exceeding the corresponding end-to-end delay deadline, δ satisfies where F RC denotes the set of RC flows to be solved, f h denotes the hth RC flow in the set, denotes the offset of the hth RC flow to the destination, the calculation method and principle of which have been described above and will not be discussed here; denotes the initial offset of the data stream route vl h of the kth RC flow, dl h denotes the end-to-end delay deadline of the hth RC flow.

[0158] Further, a penalty function is constructed based on the maximum bandwidth constraint. The penalty function is represented as:

[0159]

[0160] wherein, denotes the number of affected RC flows; vl j denotes the route of the jth new TT flow or RC flow in the network; e αβ denotes the directed edge connecting the network nodes v α , v β ; denotes the worst time interval for transmission of any frame of the RC flow from the network node v α to the network node v β ; dl i denotes the end-to-end delay deadline of the ith RC flow; E denotes the set of directed communication links between the nodes of the network topology; e denotes the directed communication link between the nodes of the network topology; l j denotes the data stream frame length of the jth new TT flow or RC flow in the network; p j denotes the data stream period of the jth new TT flow or RC flow in the network; C j denotes the maximum bandwidth service rate of the directed link; denotes the expectation that the TT flow to be admitted and the RC flow in the network satisfy the maximum bandwidth constraint.

[0161] Further, a fitness function is constructed based on the objective function, the penalty function and the weight coefficient of the penalty function. The fitness function is represented as:

[0162] F(X) = Q(X) + γE(X);

[0163] wherein γ is the penalty weight; preferably, the value is 300.

[0164] s25, obtaining the partial RC flow compression scheme based on the fitness function.

[0165] Specifically, in each iteration of the particle swarm reaching the maximum number of iterations, the current optimal solution pBest of a single particle and the optimal solution gBest of all particles in the population are obtained based on the fitness function. According to the fitness function constructed in s24, the obtained solution needs to satisfy Further, in the desired iteration reaching the maximum number of iterations, the optimal solution is determined based on the best fitness function value, and the partial RC flow compression scheme is obtained.

[0166] Further, if the RC flow compression solving unit successfully solves the partial RC flow compression scheme, the reconstructed scheme dispatching unit sends the partial RC flow compression scheme and the second scheduling scheme to the integrated interface module as the second configuration scheme.

[0167] If the RC flow compression solving unit fails to solve the partial RC flow compression scheme, the reconstructed scheme dispatching unit sends the second scheduling scheme to the integrated interface module as the second configuration scheme, and implements a traditional RC flow reconstruction method on the RC flow whose worst-case delay exceeds the corresponding end-to-end delay deadline; or when the second routing scheme fails to be solved, sends the second configuration scheme solving failure result to the integrated interface module.

[0168] It should be noted that the traditional RC flow reconstruction method is a method for processing RC flows in the network when a new TT flow is admitted in the prior art. Specifically, the traditional RC flow reconstruction method is a post-hoc flow reconfiguration method, that is, only the routing and scheduling scheme of the newly added TT flow is considered to meet the related delay and jitter constraints in the design phase; after the newly added TT flow is dispatched to the node, whether the specified RC flow needs to be reconfigured is determined according to the specific frame loss situation (continuous periodic frame loss or communication timeout) in the network node; at this time, the communication of these timeout RC flows (i.e., stop data transmission) is interrupted, and the routing scheme is re-solved and dispatched according to the link load information; the process is repeated until there is no RC flow in the network that appears communication timeout, or after a sufficient number of iterations and the problem cannot be solved, gradually try to restore a subset of the set of RC flows; after further iteration for a sufficient number of times and still unable to solve the problem, gradually try to reduce the subset until the subset is empty (empty set means that all specified RC flows are not restored and do not communicate at all). The strategy for reducing the subset is determined based on expert experience and the importance ranking of RC flows, and is predefined at the initial design time.

[0169] It is further needed to be explained that in practical application, the probability of failure of the solving of the partial RC flow compression scheme in steps s21-s25 is extremely small, and the service quality of the partial RC flow is reduced to the maximum extent to exchange for the communication continuity of the whole RC flow, so as to avoid the packet loss and communication interruption of the RC flow.

[0170] Further, when the demand type identified by the user demand identification unit is the traffic off-network notification of the TT flow, the user demand processing unit first calls the decompression RC flow identification unit of the elastic reconstruction module.

[0171] Further, the decompression RC flow identification unit determines the set of RC flows to be solved in the decompression scheme based on the intersection of the set of RC flows in the compression state existing in the network and the set of RC flows involved in the off-network traffic route. RC ”.

[0172] Further, the RC flow decompression solving unit is configured to solve the decompression scheme of the RC flow based on the set of RC flows to be solved in the decompression scheme, so that the end-to-end worst-case delay of each RC flow does not exceed the corresponding end-to-end delay deadline; specifically, the solving process is the same as that of the RC flow compression solving unit for solving the partial RC flow compression scheme, which will not be described here.

[0173] Further, if the decompression scheme of the RC flow is successfully solved, the decompression scheme dispatching unit sends the successfully solved decompression scheme of the RC flow to the integrated interface module as the third configuration scheme; if the decompression scheme of the RC flow is successfully solved, the decompression scheme dispatching unit sends the failure result to the integrated interface module.

[0174] The integrated interface module

[0175] The integrated interface module is configured to dispatch the successfully solved first, second or third configuration scheme to the corresponding switch, target end node and end node demand processing module respectively; and when the second configuration scheme fails to be solved, feedback the admission request failure to the end node demand processing module, and when the third configuration scheme fails to be solved, feedback the failure result to the end node demand processing module.

[0176] Specifically, the integrated interface module has a northbound interface and a southbound interface.

[0177] Further, the southbound interface is configured to communicate between the centralized control subsystem and the switch, and is configured to dispatch the successfully solved first, second or third configuration scheme to the corresponding switch and target end node respectively.

[0178] Further, the northbound interface is used for communication between the centralized control subsystem and the end node demand processing module, is used for feeding back the successful configuration scheme to the end node demand processing module, and feeding back the failure result to the end node demand processing module when the second configuration scheme fails, and the admission request of the TT flow is not passed; when the third configuration scheme fails, the failure result is fed back to the end node demand processing module.

[0179] The bus network dynamic reconstruction system supporting the multi-priority flow disclosed in the embodiment solves the configuration scheme of the multiple new TT flows meeting the end-to-end worst case delay of each RC flow in the network not exceeding the corresponding end-to-end delay deadline based on the admission request of the new TT flow, and solves the second configuration scheme of the multiple new TT flows and RC flows meeting the end-to-end worst case delay of each RC flow not exceeding the corresponding end-to-end delay deadline when the solving fails, reduces the influence on the existing RC flow in the network, and avoids the problem of packet loss and communication interruption of the RC flow caused by secondary configuration from the source.

[0180] In the embodiment, the feasibility prediction unit is introduced, the new TT flow admission configuration scheme not affecting the end-to-end delay constraint of the RC flow in the network is preferentially selected, the bandwidth utilization of the RC flow is elastically adjusted by introducing the elastic reconstruction module to weaken the influence of the new TT flow on the quality of service of the RC flow, the robustness and determinacy of the network are ensured, the secondary reconstruction of the RC flow is avoided as much as possible, and the problems of packet loss and communication interruption of the RC flow are solved. By decompressing the RC flow under the condition of receiving the TT flow network exit notification, the network bandwidth resources are fully utilized, and the influence on the transmission quality of the RC flow is further reduced.

[0181] Further, the fitness function is constructed based on the delay time and the maximum bandwidth constraint of the RC flow, the particle swarm algorithm is used to solve the first routing scheme, the second routing scheme, the partial RC flow compression scheme and the RC flow decompression scheme, and the optimal solution of the corresponding scheme is ensured in each link when each module of the system solves the configuration scheme.

[0182] The above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited to this, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed in the present application, which should be covered in the protection scope of the present application.

Claims

1. A bus network dynamic reconfiguration system supporting multi-priority traffic, characterized in that, include: The network awareness and management module is used to acquire real-time topology and network traffic information of the bus network. The end-node requirement processing module is used to identify the requirement types of the source end-node. These requirement types include multiple new TT flow admission requests and TT flow traffic decommissioning notifications from the source end-node. Based on the requirement types, the topology information, and network traffic information, the module calls the control center subsystem to solve for the configuration scheme corresponding to each requirement type. If the solution is successful, the corresponding configuration scheme is fed back to the source end-node. If the solution fails, when the requirement type is multiple new TT flow admission requests, the module reports an admission request failure to the source end-node; when the requirement type is a TT flow decommissioning notification, only a failure result is reported to the end-node. The TT flow is time-triggered traffic. The control center subsystem is used to solve configuration schemes and, upon successful solution, distributes the configuration schemes to the corresponding switches and target nodes. The control center subsystem executes the following process to solve configuration schemes: When the demand type is the multiple new TT flow admission request, it solves for a first configuration scheme for the multiple new TT flows that satisfies the condition that the end-to-end latency of each RC flow does not exceed the corresponding end-to-end latency period; if solving the first configuration scheme fails, it solves for a second configuration scheme for the multiple new TT flows and RC flows that satisfies the condition that the end-to-end latency of each RC flow does not exceed the corresponding end-to-end latency period; when the demand type is a TT flow traffic decommissioning notification, if... If there are RC flows in compressed state in the network, then the RC flow decompression scheme that satisfies that the end-to-end delay of each RC flow does not exceed the corresponding end-to-end delay period is used as the third configuration scheme; the configuration scheme is dispatched to the corresponding switch, including: dispatching the successfully solved first, second or third configuration scheme to the corresponding switch, target end node and end node demand processing module respectively; and when solving the second configuration scheme fails, the admission request is reported as not passed to the end node demand processing module, and when solving the third configuration scheme fails, the failure result is reported to the end node demand processing module; wherein, the RC flow is the low-priority traffic in the network.

2. The bus network dynamic reconfiguration system according to claim 1, characterized in that, The control center subsystem includes: The routing solution module is used to solve a first routing scheme for the multiple new TT flows based on the multiple new TT flow admission requests, which satisfies that the end-to-end delay of each RC flow does not exceed the corresponding end-to-end delay period; The scheduling solution module is used to solve for a first scheduling scheme based on the first routing scheme, and to use the successfully solved first scheduling scheme as the first configuration scheme. The elastic reconfiguration module is used to, when solving the first routing scheme or the first scheduling scheme fails, solve the second scheduling scheme for the plurality of new TT flows with the fewest number of RC flows whose end-to-end latency exceeds the corresponding end-to-end latency period. If the solution fails, the solution for the second configuration scheme also fails. If the solution succeeds, it solves a partial RC flow compression scheme based on the second scheduling scheme, satisfying that the end-to-end latency of all RC flows does not exceed the corresponding end-to-end latency period, and uses the successfully solved partial RC flow compression scheme and the second scheduling scheme as the second configuration scheme; and when a traffic withdrawal notification of the demand type is TT flow is received, it solves the RC flow decompression scheme as the third configuration scheme. The integrated interface module is used to dispatch the successfully solved first, second, or third configuration schemes to the corresponding switches, target end nodes, and end node requirement processing modules, respectively; and when solving the second configuration scheme fails, it reports the admission request failure to the end node requirement processing module, and when solving the third configuration scheme fails, it reports the failure result to the end node requirement processing module.

3. The bus network dynamic reconfiguration system according to claim 2, characterized in that, The elastic reconfiguration module, based on the second scheduling scheme, solves for partial RC flow compression schemes that satisfy the condition that the end-to-end delay of all RC flows does not exceed the corresponding end-to-end delay period. It also includes: if the solution for partial RC flow compression schemes fails, the second scheduling scheme is used as the second configuration scheme, and the second configuration scheme is fed back to the source node through the integrated interface module, and the traditional RC traffic reconfiguration method is implemented for RC flows whose delay exceeds the corresponding end-to-end delay period.

4. The bus network dynamic reconfiguration system according to claim 3, characterized in that, The routing solution module includes: The routing pool solving unit is used to calculate a set of optional routing schemes based on the multiple new TT flow admission requests and the topology information. The routing allocation unit is used to iteratively solve for the optimal routing scheme based on the available routing schemes within a maximum number of iterations, with the goal of optimizing the latency rate of each RC flow affected by the new TT flow, based on the network traffic information. The feasibility prediction unit is used to determine whether the preferred routing scheme obtained in each iteration is feasible. If the end-to-end delay of each RC flow does not exceed the corresponding end-to-end delay period, the preferred routing scheme is determined to be feasible and is used as the first routing scheme. Otherwise, it is not feasible and returns to the routing allocation unit for the next iteration.

5. The bus network dynamic reconfiguration system according to claim 4, characterized in that, The method by which the feasibility prediction unit calculates the end-to-end delay of each RC stream includes: Based on the preferred routing scheme, the transmission interval of each RC flow between every two nodes on the network link is calculated, including: ； in, This indicates that any frame in the RC stream originates from a network node. Transmitted to network node The time interval; and These represent any frame of the RC stream transmitted to the network node. and The moment; This indicates the technical latency introduced by each data frame; This indicates the delay impact of the TT stream on the RC stream, where This represents the set of other TT streams that are not the new TT stream. This indicates that the network is not the new TT stream mentioned above, but other TT streams. This indicates that any frame of the RC stream originates from a network node. Transmitted to network node The time offset between them This represents the data stream frame duration of the new TT stream described in the k-th stream. This represents the new TT data flow period described in the k-th entry; This indicates the delay impact of other RC flows on this RC flow, where Represents a set of RC streams. This refers to the new TT stream. This represents the new TT stream set. Indicates the frame duration of the lth other RC stream. Indicates the flow cycle of the lth other RC flow; This indicates the additional latency of the TT stream to the RC due to the preemptive integration strategy. This indicates the frame duration of the RC stream; The delay of the corresponding RC stream is calculated based on all transmission intervals of each RC stream.

6. The bus network dynamic reconfiguration system according to claim 3, characterized in that, The elastic reconstruction module includes: The TT flow planning and allocation unit is used to select the preferred route scheme with the fewest RC flows whose end-to-end delay exceeds the corresponding end-to-end delay period as the second route scheme when the solution of the first route scheme or the first scheduling scheme fails. The second route scheme is then selected based on the multiple preferred route schemes obtained by the route allocation unit iteratively. The scheduling solution unit is then called to solve the second scheduling scheme based on the second route scheme. The RC stream compression solution unit is used to solve a partial RC stream compression scheme based on the second scheduling scheme, which satisfies that the end-to-end delay of all RC streams does not exceed the corresponding end-to-end delay period. The reconstruction scheme dispatching unit is used to send the partial RC flow compression scheme and the second scheduling scheme as the second configuration scheme to the integrated interface module when the second routing scheme and the partial RC flow compression scheme are successfully solved; or when the second routing scheme is successfully solved but the partial RC flow compression scheme is failed to be solved, the second scheduling scheme is sent as the second configuration scheme to the integrated interface module, and the traditional RC traffic reconstruction method is implemented for RC flows with a delay exceeding the corresponding end-to-end delay period; or when the second routing scheme is failed to be solved, the unit sends the result of the second configuration scheme solution failure to the integrated interface module.

7. The bus network dynamic reconfiguration system according to claim 6, characterized in that, The RC stream compression solution unit solves for a subset of RC stream compression schemes based on the second scheduling scheme, ensuring that the end-to-end delay of all RC streams does not exceed the corresponding end-to-end delay period. These schemes include: The set of RC flows to be solved is determined based on RC flows whose end-to-end delay exceeds the corresponding end-to-end delay period and RC flows that share the same directed link with their routes. Number the compression schemes for each RC flow in the set of RC flows to be solved; Each combination of the corresponding compression schemes is used as a particle, and the particle position is described by the combination of the corresponding numbers of each compression scheme in each particle. The particle swarm is randomly initialized based on the corresponding compression schemes of each RC flow in the set of RC flows to be solved. The fitness function is constructed based on the end-to-end delay and maximum bandwidth constraints of all affected RC flows; Based on the fitness function, the optimal solution for particle swarm optimization is obtained, resulting in a partial RC flow compression scheme.

8. The bus network dynamic reconfiguration system according to claim 7, characterized in that, The fitness function of the RC flow compression solver is expressed as: ； in, Let be the objective function. ; This represents the average delay rate of all RC streams; This means that the delay caused by compression in all affected RC flows can satisfy the condition that the end-to-end delay does not exceed the minimum value of the corresponding end-to-end delay period. satisfy ,in Denotes the set of RC flows to be solved. This represents the h-th RC flow in the set. This represents the offset of the h-th RC stream as it reaches the destination. This represents the data flow route for the k-th RC flow. The initial offset, Indicates the end-to-end delay duration of the h-th RC stream; For penalty weighting; For the penalty function, ； in, Indicates the number of RC flows affected; This represents the route for the j-th pending TT flow or RC flow in the network; Indicates connection to network nodes , The directed edges; This indicates that any frame in the RC stream originates from a network node. Transmitted to network node The time interval; This represents the end-to-end delay duration of the i-th RC stream; This represents the set of directed communication links between nodes in a network topology. This represents a directed communication link between nodes in a network topology. This represents the length of the data stream frame of the j-th pending TT stream or RC stream in the network; This represents the data flow period of the j-th pending TT flow or RC flow in the network; This indicates the maximum bandwidth service rate of a directed link; This indicates that the expected TT flow and the RC flow in the network should meet the maximum bandwidth constraint.

9. The bus network dynamic reconfiguration system according to claim 6, characterized in that, The elastic reconfiguration module also includes: The decompression RC flow identification unit is used to determine the set of RC flows to be solved for the decompression scheme based on the set of RC flows in the compressed state existing in the network and the set of RC flows involved in the routing of the disconnected traffic. The RC stream decompression solution unit is used to solve the set of RC streams to be solved based on the decompression scheme, and to solve the decompression scheme of the RC streams that satisfies that the end-to-end delay of each RC stream does not exceed the corresponding end-to-end delay period. The decompression scheme dispatch unit is used to send the decompression scheme of the successfully solved RC stream as the third configuration scheme to the synthesis interface module.

Images