Dynamic bandwidth allocation method and device for distributed time-sensitive network

Abstract

The invention discloses a dynamic bandwidth allocation method and device for a distributed time-sensitive network, the distributed time-sensitive network comprises a sending end set, a switch set and a receiving end set, and a transmission link of an information flow in the distributed time-sensitive network is started from a sending end, passes through a switch and is connected with the receiving end set. The method comprises the following steps: acquiring attribute information of an information stream to be transmitted in the distributed time sensitive network and configuration information of an initial output port; based on the attribute information of the information flow to be transmitted in the distributed time sensitive network and the configuration information of the initial output port, constructing a TA tuple; transmitting the TA tuple based on a transmission link of an information stream; the receiving end constructs an LA tuple based on the TA tuple; and transmitting the LA tuple based on the reverse direction of the transmission link of the information flow, and performing dynamic bandwidth allocation of the distributed time-sensitive network.

Description

Technical Field

[0001] The present invention relates to the technical fields of time-sensitive networks and distributed time-sensitive networks, and in particular to a method and device for dynamic bandwidth allocation in a distributed time-sensitive network. Background Art

[0002] Time-Sensitive Networking (TSN) is a promising real-time communication standard that integrates multiple scheduling mechanisms for deterministic transmission of traffic. Recently, distributed dynamic admission control with real-time guarantees has become a hot research area in TSN networks. By distributing configuration and management among nodes, distributed admission control not only improves scalability and robustness, but also effectively avoids the single point failure risk brought by centralized controllers. The Resource Allocation Protocol (RAP) is a distributed flow reservation protocol based on the Link-local Registration Protocol (LRP) and is currently being standardized in the IEEE P802.1Qdd standard. Although RAP provides a standardized protocol interface, how to cope with the dynamic changes in network traffic while ensuring real-time performance remains a key issue that needs to be addressed. Event-triggered (ET) traffic is a key type of traffic that supports deterministic communication in TSN. Its scheduling typically relies on mechanisms such as strict priority (SP), credit-based shapers (CBS), and asynchronous traffic shapers (ATS). Compared to time-triggered (TT) traffic, which requires a precise schedule, ET traffic requires only a few key parameters to configure, offering greater flexibility.

[0003] Several centralized admission control methods have been developed for time-critical ET traffic. Methods based on the SP architecture and TSN / CBS architecture significantly reduce admission time compared to traditional offline configuration methods and only require reconfiguration of ports along the path of newly admitted flows, eliminating the need for reconfiguration of the entire network. Furthermore, methods based on the TSN / ATS+CBS architecture further leverage the ATS mechanism to reduce disruption to existing flows. They also employ deadline-based resource adjustment strategies during admission control to adaptively balance remaining bandwidth resources, thereby improving overall network resource utilization. However, these centralized admission control methods rely on a global controller to maintain network state and execute optimization decisions, making them difficult to adapt to distributed deployment scenarios. For distributed admission control of time-critical ET traffic, existing methods typically ensure real-time performance by preconfiguring port delay and bandwidth budgets and checking whether these budgets are exceeded during the admission process. However, these methods rely on fixed and non-adjustable delay and bandwidth budgets, which limits the adaptability and stability of traffic admission and fails to fully utilize remaining bandwidth for non-time-critical traffic.

[0004] How to perform admission control on dynamic traffic in distributed time-sensitive network deployment scenarios without relying on prior information about the traffic is an urgent problem that needs to be solved. Summary of the Invention

[0005] The present invention mainly solves the problem of how to perform admission control on dynamic traffic in a distributed deployment scenario without relying on prior information of the traffic. The present invention discloses a dynamic bandwidth allocation method and device for a distributed time-sensitive network.

[0006] In a first aspect, embodiments of the present invention disclose a method for dynamic bandwidth allocation in a distributed time-sensitive network. The distributed time-sensitive network includes a transmitter set, a switch set, and a receiver set. The transmitter set includes several transmitters, the switch set includes several switches, and the receiver set includes several receivers. A transmission link for an information flow in the distributed time-sensitive network is an information transmission link that starts from a transmitter, passes through a switch, and reaches a receiver. The transmitters, switches, and receivers constitute nodes in the transmission link in the order of information transmission of the information flow. The method includes:

[0007] S1, obtaining attribute information of the information flow to be transmitted in the distributed time-sensitive network and configuration information of the initial output port;

[0008] S2, constructing a TA tuple based on attribute information of the information flow to be transmitted in the distributed time-sensitive network and configuration information of the initial output port;

[0009] S3, transmitting the TA tuple based on the transmission link of the information flow;

[0010] S4, the receiving end constructs an LA tuple based on the TA tuple;

[0011] S5, based on the reverse direction of the transmission link of the information flow, transmit the LA tuple to perform dynamic bandwidth allocation of the distributed time-sensitive network.

[0012] The attribute information of the information flow to be transmitted in the distributed time-sensitive network includes in, Represents the information flow f + The frame length, Represents the information flow f + The frame interval, Indicates the sender's response to the information flow f + Required end-to-end deadlines;

[0013] The initial output port refers to the output port of the sending end in the transmission link of the information flow;

[0014] The configuration information of the initial output port includes in, Represents the information flow f + The minimum available local deadline at the output port h of the transmission link, Represents the information flow f + The cumulative minimum available local deadline in the transmission link up to output port h, Represents the information flow f + The cumulative path length in the transmission link up to the output port h, Represents the information flow f + The status flag of the corresponding output port h;

[0015] The value type of the status flag is None or Failed, where None means that the communication resources of the output port h meet the requirements of the information flow f. + transmission requirements, Failed means that the communication resources of output port h do not meet the transmission requirements of information flow f. + transmission requirements;

[0016] The TA tuple is a sender announcement tuple; a sender announcement tuple corresponds to an information flow to be transmitted in a distributed time-sensitive network;

[0017] For the information flow f + , the TA tuple at the output port h is represented as

[0018] The information flow-based transmission link transmits the TA tuple, including:

[0019] S31, sending the TA tuple to the next switch in the transmission link of the information flow;

[0020] S32, the switch defines the received TA tuple as the first TA tuple; the first TA tuple The expression is:

[0021]

[0022] in, Represents the received information flow f + The frame length, Represents the received information flow f + The frame interval, Indicates the received information flow f + The required end-to-end deadline, Represents the received information flow f + The minimum available local deadline at output port h, Represents the received information flow f + The cumulative minimum available local deadline as of output port h, Represents the received information flow f + The cumulative path length up to the output port h, Represents the received information flow f + Corresponding TA status flag;

[0023] S33, for the first TA tuple The information flow attribute information in the inheritance process is processed to obtain the updated TA tuple TA1 (f + ,h) in the information flow attribute information, the updated TA tuple TA1 (f + ,h) The expression of the information flow attribute information is:

[0024]

[0025] in, and Represent the updated information flow f + Frame length, information flow f + The frame interval of the transmitter is the information flow f + Required end-to-end deadlines;

[0026] S34, obtaining a bandwidth allocation information set of the output port of the switch; the bandwidth allocation information set includes: the output port h has been allocated to each traffic class j∈[1,NAVB ] bandwidth where N AVB Indicates the total number of traffic classes; the total bandwidth upper limit idSl allocated to all traffic classes on the output port max ;The maximum frame length of traffic in the network l maax ; Link transmission rate C of the output port;

[0027] S35, based on the first TA tuple and bandwidth allocation information set, construct the updated TA tuple TA1(f + ,h), the updated TA tuple TA1(f + ,h) for transmission.

[0028] Based on the first TA tuple and bandwidth allocation information set, construct the updated TA tuple TA1(f + ,h), the updated TA tuple TA1(f + ,h) transmission, including:

[0029] S351, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + The minimum available local deadline at output port h Its calculation expression is:

[0030]

[0031] Where i is the flow f + The sequence number of the traffic class to which it belongs, for The subset of traffic class i in R h is the residual bandwidth of output port h and

[0032] S352, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + The cumulative minimum available local deadline at output port h Its calculation expression is:

[0033]

[0034] S353, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + Cumulative path length at output port h Its calculation expression is:

[0035]

[0036] S354, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + TA status flag at output port h The calculation expression is:

[0037]

[0038] S355, construct the updated TA tuple, whose expression is:

[0039]

[0040] S356, storing the updated TA tuple in the output port of the current switch, and sending the updated TA tuple to the next node in the transmission link;

[0041] S357, determine whether the next node in the transmission link is a receiving end, and obtain a first determination result; if the first determination result is yes, execute S4; if the first determination result is no, execute S31.

[0042] The receiving end constructs an LA tuple based on the TA tuple, including:

[0043] S41, the receiving end receives and obtains the TA tuple;

[0044] S42, obtain the output port of the receiving end for the information flow f + Required end-to-end deadline

[0045] S43, based on the end-to-end deadline Determining whether the TA tuple received by the receiving end meets any condition in the receiving condition set;

[0046] S44: If any of the conditions in the receiving condition set is met, confirm that the receiving end cannot subscribe to the information flow f + , terminate the information flow f + The sending process ends the dynamic bandwidth allocation method for the distributed time-sensitive network; if the receiving condition set is not met, executing S45;

[0047] S45, obtain information flow f + The local deadline at the output port h of the receiver

[0048] S46, the receiving end constructs an LA tuple based on the received TA tuple;

[0049] The expression of the LA tuple is:

[0050]

[0051] in, Represents the information flow f + The end-to-end deadline on path P, Represents the information flow f + The cumulative minimum available local deadline on path P is equal to the number of TA tuples received by the receiver. Represents the information flow f + The cumulative path length at intersection P is equal to the number of TA tuples received by the receiver. Represents the information flow f + The local deadline at the output port h, Represents the information flow f + The corresponding LA status flag has an initial value of Failed.

[0052] The receiving condition set includes:

[0053] The TA status flag of the TA tuple is Failed;

[0054] The cumulative minimum available local deadline in the TA tuple does not meet the end-to-end deadline requirement, i.e. Where P is the information flow f corresponding to the receiving end + The transmission path, is the information flow f + End-to-end deadline on path P.

[0055] The method of transmitting the LA tuple in the reverse direction of the transmission link of the information flow to perform dynamic bandwidth allocation of the distributed time-sensitive network includes:

[0056] S51, according to the reverse direction of the information transmission link starting from the transmitting end, passing through the switch, and reaching the receiving end, the receiving end sends an LA tuple to the switch;

[0057] S52, the output port of the node that receives the LA tuple represents the received LA tuple as:

[0058]

[0059] in, Represents the received information flow f + The end-to-end deadline on path P, Represents the received information flow f + The cumulative minimum available local deadline on path P, Represents the received information flow f + The cumulative path length at intersection P, Represents the received information flow f + The local deadline at the output port h, Represents the received information flow f + The corresponding LA status flag;

[0060] S53, generating an updated LA tuple according to the received LA tuple;

[0061] S54, for the updated LA tuple, determine whether Get the third judgment result; if the third judgment result is not satisfied, for each traffic class j∈[1,N AVB ] is configured as Execute S55; if the third judgment result is satisfied, update the LA tuple If the value is Failed, TA information is sent from the output port of the switch to the receiving end according to the information transmission link, and the bandwidth resources allocated to the information flow on the output port are recovered, and S55 is executed.

[0062] S55, determining whether the node generating the updated LA tuple is a switch, and obtaining a fourth determination result;

[0063] If the fourth determination result is yes, the updated LA tuple is further transmitted along the reverse direction of the information transmission link, and the node receiving the updated LA tuple stores the updated LA tuple in its output port and executes S51;

[0064] If the fourth judgment result is no, Is it Ready or Partial Failed, get the fifth judgment result, if the fifth judgment result is yes, the information flow f + Forward along the output ports of the nodes marked as Ready or Partial Failed in the LA tuple;

[0065] If the fifth judgment result is no, terminate the information flow f + The sending process ends the dynamic bandwidth allocation method for the distributed time-sensitive network.

[0066] According to a second aspect of an embodiment of the present invention, a dynamic bandwidth allocation device for a distributed time-sensitive network is disclosed, the device comprising:

[0067] a memory storing executable program code;

[0068] a processor coupled to the memory;

[0069] The processor calls the executable program code stored in the memory to execute the dynamic bandwidth allocation method for a distributed time-sensitive network.

[0070] According to a third aspect of an embodiment of the present invention, a computer-storable medium is disclosed. The computer-storable medium stores computer instructions. When the computer instructions are called by a computer, the computer instructions are used to execute the dynamic bandwidth allocation method for a distributed time-sensitive network.

[0071] According to a fourth aspect of the embodiments of the present invention, an information data processing terminal is disclosed. The information data processing terminal is used to implement the dynamic bandwidth allocation method for the distributed time-sensitive network.

[0072] The beneficial effects of the present invention are:

[0073] This paper constructs a distributed adaptive admission control framework based on network calculus that dynamically adjusts the local cutoff time of each output port while meeting end-to-end deadlines. This framework allocates only the minimum bandwidth required for time-critical ET traffic, thereby maximizing the bandwidth available for non-time-critical traffic and improving overall network bandwidth utilization.

[0074] This paper designs a distributed adaptive admission control scheme based on RAP. By carrying and propagating delay information in TA and LA tuples, this scheme dynamically adjusts the local cutoff time of each node to meet end-to-end timing requirements, thereby improving the stability and adaptability of the traffic admission process. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 Flow chart for the implementation of the method of the present invention;

[0076] Figure 2 A diagram illustrating the composition of a time-sensitive network used in an embodiment of the present invention. DETAILED DESCRIPTION

[0077] In order to better understand the content of the present invention, an embodiment is given here.

[0078] Figure 1 4 is an implementation flow chart of the method of the present invention.

[0079] In a first aspect, embodiments of the present invention disclose a method for dynamic bandwidth allocation in a distributed time-sensitive network. The distributed time-sensitive network includes a transmitter set, a switch set, and a receiver set. The transmitter set includes several transmitters, the switch set includes several switches, and the receiver set includes several receivers. A transmission link for an information flow in the distributed time-sensitive network is an information transmission link that starts from a transmitter, passes through a switch, and reaches a receiver. The transmitters, switches, and receivers constitute nodes in the transmission link in the order of information transmission of the information flow. The method includes:

[0080] S1, obtaining attribute information of the information flow to be transmitted in the distributed time-sensitive network and configuration information of the initial output port;

[0081] S2, constructing a TA tuple based on attribute information of the information flow to be transmitted in the distributed time-sensitive network and configuration information of the initial output port;

[0082] S3, transmitting the TA tuple based on the transmission link of the information flow;

[0083] S4, the receiving end constructs an LA tuple based on the TA tuple;

[0084] S5, transmitting the LA tuple based on the reverse direction of the transmission link of the information flow, and performing dynamic bandwidth allocation of the distributed time-sensitive network;

[0085] The LA tuple is the receiver subscription tuple;

[0086] The attribute information of the information flow to be transmitted in the distributed time-sensitive network includes in, Represents the information flow f + The frame length, Represents the information flow f + The frame interval, Indicates the sender's response to the information flow f + Required end-to-end deadlines;

[0087] The initial output port refers to the output port of the sending end in the transmission link of the information flow;

[0088] The configuration information of the initial output port includes in, Represents the information flow f + The minimum available local deadline at the output port h of the transmission link, Represents the information flow f + The cumulative minimum available local deadline in the transmission link up to output port h, Represents the information flow f + The cumulative path length in the transmission link up to the output port h, Represents the information flow f + The status flag of the corresponding output port h;

[0089] The value type of the status flag is None or Failed, where None means that the communication resources of the output port h meet the requirements of the information flow f. + transmission requirements, Failed means that the communication resources of output port h do not meet the transmission requirements of information flow f. + transmission requirements;

[0090] The TA tuple is a sender announcement tuple; a sender announcement tuple corresponds to an information flow to be transmitted in a distributed time-sensitive network;

[0091] For the information flow f + , the TA tuple at the output port h is represented as

[0092] The information flow-based transmission link transmits the TA tuple, including:

[0093] S31, sending the TA tuple to the next switch in the transmission link of the information flow;

[0094] S32, the switch defines the received TA tuple as the first TA tuple; the first TA tuple The expression is:

[0095]

[0096] in, Represents the received information flow f + The frame length, Represents the received information flow f + The frame interval, Indicates the received information flow f + The required end-to-end deadline, Represents the received information flow f + The minimum available local deadline at output port h, Represents the received information flow f + The cumulative minimum available local deadline as of output port h, Represents the received information flow f + The cumulative path length up to the output port h, Represents the received information flow f +Corresponding TA status flag.

[0097] The variable values ​​in the first TA tuple are inherited from the corresponding variables in the received TA tuple;

[0098] S33, using the first TA tuple Generate updated TA tuple TA1(f + ,h);

[0099] The use of the first TA tuple Generate updated TA tuple TA1(f + ,h), including:

[0100] Obtain the bandwidth allocation information set of the output port of the switch; the bandwidth allocation information set includes: the output port h has been allocated to each traffic class j∈[1,N AVB ] bandwidth where N AVB Indicates the total number of traffic classes; the total bandwidth upper limit idSl allocated to all traffic classes on the output port max ;The maximum frame length of traffic in the network l max ; Link transmission rate C of the output port;

[0101] S34, for the first TA tuple The information flow attribute information in the inheritance process is processed to obtain the updated TA tuple TA1 (f + ,h) in the information flow attribute information, the updated TA tuple TA1 (f + ,h) The expression of the information flow attribute information is:

[0102]

[0103] Among them, lf1+, p f1+ and Represent the updated information flow f + Frame length, information flow f + The frame interval of the transmitter is the information flow f + Required end-to-end deadlines;

[0104] S35, based on the first TA tuple and bandwidth allocation information set, construct the updated TA tuple TA1(f + ,h), the updated TA tuple TA1(f + ,h)transmission;

[0105] The first TA tuple (TA) based on ～(f^+, h) and the bandwidth allocation information set, constructing an updated TA tuple TA1(f^+, h), and transmitting the updated TA tuple TA1(f^+, h), including:

[0106] S351, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + The minimum available local deadline at output port h Its calculation expression is:

[0107]

[0108] Where i is the flow f + The sequence number of the traffic class to which it belongs, for The subset of traffic class i in R h is the residual bandwidth of output port h and l f is the frame length of information stream f.

[0109] S352, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + The cumulative minimum available local deadline at output port h Its calculation expression is:

[0110]

[0111] S353, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + Cumulative path length at output port h Its calculation expression is:

[0112]

[0113] S354, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + TA status flag at output port h The calculation expression is:

[0114]

[0115] S355, construct the updated TA tuple, whose expression is:

[0116]

[0117] S356, storing the updated TA tuple in the output port of the current switch, and sending the updated TA tuple to the next node in the transmission link;

[0118] S357, determining whether the next node in the transmission link is a receiving end, and obtaining a first determination result; if the first determination result is yes, executing S4; if the first determination result is no, executing S31;

[0119] The receiving end constructs an LA tuple based on the TA tuple, including:

[0120] S41, the receiving end receives and obtains the TA tuple;

[0121] S42, obtain the output port of the receiving end for the information flow f + Required end-to-end deadline

[0122] S43, based on the end-to-end deadline Determining whether the TA tuple received by the receiving end meets any condition in the receiving condition set;

[0123] S44: If any of the conditions in the receiving condition set is met, confirm that the receiving end cannot subscribe to the information flow f + , terminate the information flow f + The sending process ends the dynamic bandwidth allocation method for the distributed time-sensitive network; if the receiving condition set is not met, executing S45;

[0124] S45, obtain information flow f + The local deadline at the output port h of the receiver

[0125] S46, the receiving end constructs an LA tuple based on the received TA tuple;

[0126] The expression of the LA tuple is:

[0127]

[0128] in, Represents the information flow f + The end-to-end deadline on path P, Represents the information flow f + The cumulative minimum available local deadline on path P is equal to the number of TA tuples received by the receiver. Represents the information flow f + The cumulative path length at intersection P is equal to the number of TA tuples received by the receiver. Represents the information flow f+ The local deadline at the output port h, Represents the information flow f + The corresponding LA status flag has an initial value of Failed;

[0129] The receiving condition set includes:

[0130] The TA status flag of the TA tuple is Failed;

[0131] The cumulative minimum available local deadline in the TA tuple does not meet the end-to-end deadline requirement, i.e. Where P is the information flow f corresponding to the receiving end + The transmission path, is the information flow f + End-to-end deadline on path P.

[0132] Indicates that the end-to-end deadline requirement is not met;

[0133] The method of transmitting the LA tuple in the reverse direction of the transmission link of the information flow to perform dynamic bandwidth allocation of the distributed time-sensitive network includes:

[0134] S51, according to the reverse direction of the information transmission link starting from the transmitting end, passing through the switch, and reaching the receiving end, the receiving end sends an LA tuple to the switch;

[0135] S52, the output port of the node that receives the LA tuple represents the received LA tuple as:

[0136]

[0137] in, Represents the received information flow f + The end-to-end deadline on path P, Represents the received information flow f + The cumulative minimum available local deadline on path P, Represents the received information flow f + The cumulative path length at intersection P, Represents the received information flow f + The local deadline at the output port h, Represents the received information flow f + The corresponding LA status flag;

[0138] S53, generating an updated LA tuple according to the received LA tuple;

[0139] S54, for the updated LA tuple, determine whether Get the third judgment result; if the third judgment result is not satisfied, for each traffic class j∈[1,N AVB ] is configured as Then when When Assigned to Ready, when When it exists, Assigned to The result of merging the value and Ready according to the merging rule is executed, and S55 is executed; if the third judgment result is satisfied, the LA tuple is updated. If the value is Failed, TA information is sent from the output port of the switch to the receiving end according to the information transmission link, and the bandwidth resources allocated to the information flow on the output port are recovered, and S55 is executed.

[0140] S55, determining whether the node generating the updated LA tuple is a switch, and obtaining a fourth determination result;

[0141] If the fourth determination result is yes, the updated LA tuple is further transmitted along the reverse direction of the information transmission link, and the node receiving the updated LA tuple stores the updated LA tuple in its output port and executes S51;

[0142] If the fourth judgment result is no, Is it Ready or Partial Failed, get the fifth judgment result, if the fifth judgment result is yes, the information flow f + Forward along the output ports of the nodes marked as Ready or Partial Failed in the LA tuple;

[0143] If the fifth judgment result is no, terminate the information flow f + The sending process ends the dynamic bandwidth allocation method for the distributed time-sensitive network.

[0144] for The value is Ready, indicating that the traffic flow transmission resources of the node are ready; the value is Failed, indicating that the information flow transmission resources do not meet the requirements; the value is PartialFailed, indicating that the information flow resources partially meet the requirements.

[0145] Generating an updated LA tuple according to the received LA tuple includes:

[0146] Determine whether the output port already stores an existing LA tuple, and obtain a second determination result;

[0147] If the second judgment result is yes, read the existing LA tuple and express it as

[0148] in Indicates the existing information flow f + The end-to-end deadline on path P, Indicates the existing information flow f + The cumulative minimum available local deadline on path P, Indicates the existing information flow f + The cumulative path length at intersection P, Indicates the existing information flow f + The local deadline at the output port h, Indicates the existing information flow f + The corresponding LA status flag;

[0149] Get information flow f + The minimum available local deadline in the TA tuple at the corresponding output port of the switch The set of flows that already exist at the output port Each of these flows With frame length l f , frame interval p f and local deadlines at the output ports

[0150] If the second judgment result is no, define Get the existing configuration information of the output port; get the information flow f + The minimum available local deadline in the TA tuple at the corresponding output port of the switch The set of flows that already exist at the output port Each of these flows With frame length l f , frame interval p f and local deadlines at the output ports

[0151] Get the Status flags in like The value is Failed. Perform the first assignment operation; if The value is not Failed. Perform the second assignment operation;

[0152] The first assignment operation includes:

[0153] like The received LA tuple Assign values ​​to and will The value is assigned to Failed.

[0154] like Existence, will Assign values ​​to and will The value of and Failed, according to the merging rules to get the result value, assign the result value to

[0155] The second assignment operation includes:

[0156] Calculate the information flow f + The local deadline corresponding to the received LA tuple on the output port for

[0157]

[0158] If both meet exist, Not for Failed and If the following three conditions are met, no further steps will be executed. Assign values ​​to And for each traffic class j∈[1,N AVB ], calculate the required bandwidth The calculation expression is:

[0159]

[0160] in is a collection With the set {f +} belongs to the subset of class j, like Then: When When Assign the value to Ready. When it exists, Assigned to and "Ready" is the result of merging according to the merge rules.

[0161] For status flags The value range includes Ready, Partial Failed, and Failed;

[0162] The merging rule includes: when both states are Ready, the merging result is Ready; if both states are Failed, the merging result is Failed; in other cases, the merging result is Partial Failed.

[0163] LA tuple after replacement The meaning of the elements, according to The meaning of the corresponding element is determined.

[0164] The meaning of the variables of the received LA tuple is determined according to the meaning of the corresponding variables of the LA tuple sent by the receiving end to the switch.

[0165] According to a second aspect of an embodiment of the present invention, a dynamic bandwidth allocation device for a distributed time-sensitive network is disclosed, the device comprising:

[0166] a memory storing executable program code;

[0167] a processor coupled to the memory;

[0168] The processor calls the executable program code stored in the memory to execute the dynamic bandwidth allocation method for a distributed time-sensitive network.

[0169] According to a third aspect of an embodiment of the present invention, a computer-storable medium is disclosed. The computer-storable medium stores computer instructions. When the computer instructions are called by a computer, the computer instructions are used to execute the dynamic bandwidth allocation method for a distributed time-sensitive network.

[0170] According to a fourth aspect of the embodiments of the present invention, an information data processing terminal is disclosed. The information data processing terminal is used to implement the dynamic bandwidth allocation method for the distributed time-sensitive network.

[0171] Figure 2 A diagram illustrating the composition of a time-sensitive network used in an embodiment of the present invention.

[0172] In a fifth aspect of the present invention, a method for dynamic bandwidth allocation in a distributed time-sensitive network is disclosed, wherein the time-sensitive network used is composed of Figure 2 As shown, including:

[0173] Step 1: The sending system constructs the TA tuple

[0174] Step 1-1: The sending system builds TA

[0175] The sending system constructs a TA tuple to announce the new flow f + The TA tuple sent at the output port h0 of the sending end system is

[0176]

[0177] Step 2: End systems and switches propagate TA

[0178] This step is repeated in a distributed manner on the sending system and switches in the network and includes the following substeps:

[0179] The sending system executes the following process:

[0180] Step 2-1: The sending system receives TA

[0181] The sending end system receives the TA tuple constructed by itself, and the received TA tuple at its output port h0 is

[0182]

[0183] Step 2-2: The sending system processes and propagates TA

[0184] The sending end system reads the existing configuration information at the output port h0 based on the received TA tuple. The existing flow set at the output port h0 Among them, the frame length of stream f1 is The frame interval is p f1 =4000us, and its local deadline at the output port h0 is The frame length of stream f2 is The frame interval is The local deadline at the output port h0 is The bandwidth allocated to traffic class 1 at output port h0 is The total bandwidth limit idSl allocated to all traffic classes of the output port max The maximum frame length of the traffic in the network is 75Mbit / s. max The link transmission rate of the output port is C = 100Mbit / s.

[0185] According to the received TA tuple and the existing configuration information at the output port h0, the updated sending TA tuple at the output port h0 is obtained as

[0186]

[0187] Switch 1 executes the following process:

[0188] Step 2-1: Switch 1 receives TA

[0189] Switch 1 receives the TA tuple transmitted by the sending end system, and the received TA tuple at its output port h1 is

[0190]

[0191] Step 2-2: Switch 1 processes and propagates TA

[0192] Switch 1 reads the existing configuration information at output port h1 based on the received TA tuple. The existing flow set at output port h1 Among them, the frame length of stream f1 is 200Bytes, frame interval is p f1 =4000us, and its local deadline at the output port g1 is The frame length of stream f2 is The frame interval is p f1 = 4000us, and the local deadline at the output port h1 is The bandwidth allocated to traffic class 1 at output port h1 is The total bandwidth limit idSl allocated to all traffic classes of the output port max The maximum frame length of the traffic in the network is 75Mbit / s. max The link transmission rate of the output port is C = 100Mbit / s.

[0193] According to the received TA tuple and the existing configuration information at the output port h1, the updated sending TA tuple at the output port h1 is obtained as

[0194]

[0195] Switch 2 executes the following process:

[0196] Step 2-1: Switch 2 receives TA

[0197] Switch 2 receives the TA tuple transmitted by switch 1, and the received TA tuple at its output port h2 is

[0198]

[0199] Step 2-2: Switch 2 processes and propagates TA

[0200] Switch 2 reads the existing configuration information at output port h2 based on the received TA tuple. The existing flow set at output port h2 Among them, the frame length of stream f1 is The frame interval is p f1 =4000us, and its local deadline at the output port h2 is The frame length of stream f2 is The frame interval is Its local deadline at port h2 is The bandwidth allocated to traffic class 1 at output port h2 is The total bandwidth limit idSl allocated to all traffic classes of the output port max The maximum frame length of the traffic in the network is 75Mbit / s. max The link transmission rate of the output port is C = 100Mbit / s.

[0201] According to the received TA tuple and the existing configuration information at the output port h2, the updated sending TA tuple at the output port h2 is obtained as

[0202]

[0203] Step 3: Build LA in the receiving system

[0204] Step 3-1: The receiving system evaluates subscription feasibility

[0205] The TA status flag in the TA tuple transmitted by switch 2 received by the receiving end system is “None”, and the minimum available local deadline accumulated on the received path is 620.31us. + End-to-end deadline requirements Therefore, the new flow f + End-to-end cutoff requirements Both conditions in step 3-1 are not met, and the receiving system has the ability to subscribe to the new stream f + conditions.

[0206] Step 3-2: Receiver system builds LA

[0207] The receiving system constructs an LA tuple to subscribe to the new stream. The LA tuple is defined as

[0208]

[0209] Step 4: Switches and end systems propagate LA

[0210] This step is repeated in a distributed manner across the switches and sending end systems in the network and includes the following substeps:

[0211] Switch 2 executes the following process:

[0212] Step 4-1: Switch 2 receives LA

[0213] Switch 2 receives the LA tuple transmitted by the receiving end system and receives the LA tuple at its output port h2.

[0214]

[0215] Step 4-2: Switch 2 processes and propagates the LA, and also reserves bandwidth.

[0216] Switch 2 generates an updated LA tuple based on the received LA tuple at the receiving output port h2 and performs the corresponding bandwidth allocation operation:

[0217] 1. Read the existing configuration information of output port h2. The current new flow f + There is no existing LA tuple at the output port h2, that is, The minimum available local deadline at output port h2 provided in its corresponding TA tuple The set of flows that already exist at output port h2 The frame length of stream f1 is The frame interval is p f1 = 4000us and the local deadline is The frame length of stream f2 is The frame interval is p f2 = 4000us and the local deadline is The bandwidth allocated to traffic class 1 at output port h2 is The total bandwidth limit idSl allocated to all traffic classes on the output port max The maximum frame length of the traffic in the network is 75Mbit / s. max The link transmission rate of the output port is C = 100Mbit / s.

[0218] 2. If it is not "Failed", the updated LA tuple is calculated as

[0219]

[0220] The bandwidth allocation for computing class 1 is This value is less than the bandwidth upper limit idSl max , so the Used to configure output port h2.

[0221] Finally, switch 2 propagates the updated LA tuple to switch 1.

[0222] Switch 1 executes the following process:

[0223] Step 4-1: Switch 1 receives LA

[0224] Switch 1 receives the LA tuple transmitted by switch 2, and the received LA tuple is at its output port h1.

[0225]

[0226] Step 4-2: Switch 1 processes and propagates the LA, and also reserves bandwidth

[0227] Switch 1 generates an updated LA tuple based on the received LA tuple at the receiving output port h1 and performs the corresponding bandwidth allocation operation:

[0228] 1. Read the existing configuration information of output port g1. The current new flow f + There is no existing LA tuple at the output port h1, that is, The minimum available local deadline at output port h1 provided in its corresponding TA tuple The set of flows that already exist at output port h1 The frame length of stream f1 is The frame interval is and the local deadline is The frame length of stream f2 is The frame interval is and the local deadline is The bandwidth allocated to traffic class 1 at output port h1 is The total bandwidth limit idSl allocated to all traffic classes on the output port max The maximum frame length of the traffic in the network is 75Mbit / s. max The link transmission rate of the output port is C = 100Mbit / s.

[0229] 2. If it is not "Failed", the updated LA tuple is calculated as

[0230]

[0231] The bandwidth allocation for computing class 1 is This value is less than the bandwidth upper limit idSl max , so the Used to configure output port h1.

[0232] Finally, switch 1 propagates the updated LA tuple to the sending end system.

[0233] The sending system executes the following process:

[0234] Step 4-1: The sending system receives LA

[0235] The sending end system receives the LA tuple transmitted by switch 1, and the received LA tuple is at its output port h0.

[0236]

[0237] Step 4-2: The sending system processes and propagates the LA and makes bandwidth reservations

[0238] The sending end system generates an updated LA tuple at the receiving output port h0 based on the received LA tuple and performs the corresponding bandwidth allocation operation:

[0239] 1. Read the existing configuration information of output port h0. The current new flow f + There is no existing LA tuple at the output port h0, that is, The minimum available local deadline at output port h0 provided in its corresponding TA tuple The set of flows that already exist at output port h0 The frame length of stream f1 is The frame interval is and the local deadline is The frame length of stream f2 is l f2 =400Bytes, frame interval is p f2 = 4000us and the local deadline is The bandwidth allocated to traffic class 1 at output port h0 is The total bandwidth limit idSl allocated to all traffic classes on the output port max The maximum frame length of the traffic in the network is 75Mbit / s. max The link transmission rate of the output port is C = 100Mbit / s.

[0240] 2. If it is not "Failed", the updated LA tuple is calculated as

[0241]

[0242] The bandwidth allocation for computing class 1 is This value is less than the bandwidth upper limit idSl max , so the Used to configure output port h0.

[0243] Finally, due to the updated is "Ready", the sending end system sends a new flow f along the output ports h0, h1, and h2 + data is sent.

[0244] The foregoing is merely an embodiment of the present invention and is not intended to limit the present invention. It will be apparent to those skilled in the art that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims of the present invention.

Claims

1. A method for dynamic bandwidth allocation in a distributed time-sensitive network, characterized in that: The distributed time-sensitive network includes a transmitter set, a switch set, and a receiver set. The transmitter set includes several transmitters, the switch set includes several switches, and the receiver set includes several receivers. The transmission link of the information flow in the distributed time-sensitive network is an information transmission link that starts from the transmitter, passes through the switch, and reaches the receiver. The transmitters, switches, and receivers constitute nodes in the transmission link according to the order of information transmission of the information flow. The method includes: S1, obtaining attribute information of the information flow to be transmitted in the distributed time-sensitive network and configuration information of the initial output port; S2, constructing a TA tuple based on attribute information of the information flow to be transmitted in the distributed time-sensitive network and configuration information of the initial output port; S3, transmitting the TA tuple based on the transmission link of the information flow; S4, the receiving end constructs an LA tuple based on the TA tuple; S5, based on the reverse direction of the transmission link of the information flow, transmit the LA tuple to perform dynamic bandwidth allocation of the distributed time-sensitive network.

2. The method for dynamic bandwidth allocation in a distributed time-sensitive network according to claim 1, wherein: The attribute information of the information flow to be transmitted in the distributed time-sensitive network includes in, Represents the information flow f + The frame length, Represents the information flow f + The frame interval, Indicates the sender's response to the information flow f + Required end-to-end deadlines; The initial output port refers to the output port of the sending end in the transmission link of the information flow; The configuration information of the initial output port includes in, Represents the information flow f + The minimum available local deadline at the output port h of the transmission link, Represents the information flow f + The cumulative minimum available local deadline in the transmission link up to output port h, Represents the information flow f + The cumulative path length in the transmission link up to the output port h, Represents the information flow f + The status flag of the corresponding output port h; The value type of the status flag is None or Failed, where None means that the communication resources of the output port h meet the requirements of the information flow f. + transmission requirements, Failed means that the communication resources of output port h do not meet the transmission requirements of information flow f. + transmission requirements; The TA tuple is a sender announcement tuple; a sender announcement tuple corresponds to an information flow to be transmitted in a distributed time-sensitive network; For the information flow f + , the TA tuple at the output port h is represented as 3. The method for dynamic bandwidth allocation in a distributed time-sensitive network according to claim 2, wherein: The information flow-based transmission link transmits the TA tuple, including: S31, sending the TA tuple to the next switch in the transmission link of the information flow; S32, the switch defines the received TA tuple as the first TA tuple; the first TA tuple The expression is: in, Represents the received information flow f + The frame length, Represents the received information flow f + The frame interval, Indicates the received information flow f + The required end-to-end deadline, Represents the received information flow f + The minimum available local deadline at output port h, Represents the received information flow f + The cumulative minimum available local deadline as of output port h, Represents the received information flow f + The cumulative path length up to the output port h, Represents the received information flow f + Corresponding TA status flag; S33, for the first TA tuple The information flow attribute information in the inheritance process is processed to obtain the updated TA tuple TA1 (f + ,h) in the information flow attribute information, the updated TA tuple TA1 (f + ,h) The expression of the information flow attribute information is: in, and Represent the updated information flow f + Frame length, information flow f + The frame interval of the transmitter is the information flow f + Required end-to-end deadlines; S34, obtaining a bandwidth allocation information set of the output port of the switch; the bandwidth allocation information set includes: the output port h has been allocated to each traffic class j∈[1,N AVB ] bandwidth where N AVB Indicates the total number of traffic classes; the total bandwidth upper limit idSl allocated to all traffic classes on the output port max ;The maximum frame length of traffic in the network l max ; Link transmission rate C of the output port; S35, based on the first TA tuple and bandwidth allocation information set, construct the updated TA tuple TA1(f + ,h), the updated TA tuple TA1(f + ,h) for transmission.

4. The method for dynamic bandwidth allocation in a distributed time-sensitive network according to claim 3, wherein: Based on the first TA tuple and bandwidth allocation information set, construct the updated TA tuple TA1(f + ,h), the updated TA tuple TA1(f + ,h) transmission, including: S351, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + The minimum available local deadline at output port h Its calculation expression is: Where i is the flow f + The sequence number of the traffic class to which it belongs, for The subset of traffic class i in R h is the residual bandwidth of output port h and S352, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + The cumulative minimum available local deadline at output port h Its calculation expression is: S353, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + Cumulative path length at output port h Its calculation expression is: S354, calculate and obtain the updated TA tuple TA1 (f + ,h) in the information flow f + TA status flag at output port h The calculation expression is: S355, construct the updated TA tuple, whose expression is: S356, storing the updated TA tuple in the output port of the current switch, and sending the updated TA tuple to the next node in the transmission link; S357, determine whether the next node in the transmission link is a receiving end, and obtain a first determination result; if the first determination result is yes, execute S4; if the first determination result is no, execute S31.

5. The method for dynamic bandwidth allocation in a distributed time-sensitive network according to claim 3, wherein: The receiving end constructs an LA tuple based on the TA tuple, including: S41, the receiving end receives and obtains the TA tuple; S42, obtain the output port of the receiving end for the information flow f + Required end-to-end deadline S43, based on the end-to-end deadline Determining whether the TA tuple received by the receiving end meets any condition in the receiving condition set; S44: If any of the conditions in the receiving condition set is met, confirm that the receiving end cannot subscribe to the information flow f + , terminate the information flow f + The sending process ends the dynamic bandwidth allocation method for the distributed time-sensitive network; if the receiving condition set is not met, executing S45; S45, obtain information flow f + The local deadline at the output port h of the receiver S46, the receiving end constructs an LA tuple based on the received TA tuple; The expression of the LA tuple is: in, Represents the information flow f + The end-to-end deadline on path P, Represents the information flow f + The cumulative minimum available local deadline on path P is equal to the number of TA tuples received by the receiver. Represents the information flow f + The cumulative path length at intersection P is equal to the number of TA tuples received by the receiver. Represents the information flow f + The local deadline at the output port h, Represents the information flow f + The corresponding LA status flag has an initial value of Failed.

6. The method for dynamic bandwidth allocation in a distributed time-sensitive network according to claim 5, wherein: The receiving condition set includes: The TA status flag of the TA tuple is Failed; The cumulative minimum available local deadline in the TA tuple does not meet the end-to-end deadline requirement, i.e. Where P is the information flow f corresponding to the receiving end + The transmission path, is the information flow f + End-to-end deadline on path P.

7. The method for dynamic bandwidth allocation in a distributed time-sensitive network according to claim 1, wherein: The method of transmitting the LA tuple in the reverse direction of the transmission link of the information flow to perform dynamic bandwidth allocation of the distributed time-sensitive network includes: S51, according to the reverse direction of the information transmission link starting from the transmitting end, passing through the switch, and reaching the receiving end, the receiving end sends an LA tuple to the switch; S52, the output port of the node that receives the LA tuple represents the received LA tuple as: in, Represents the received information flow f + The end-to-end deadline on path P, Represents the received information flow f + The cumulative minimum available local deadline on path P, Represents the received information flow f + The cumulative path length at intersection P, Represents the received information flow f + The local deadline at the output port h, Represents the received information flow f + The corresponding LA status flag; S53, generating an updated LA tuple according to the received LA tuple; S54, for the updated LA tuple, determine whether Get the third judgment result; if the third judgment result is not satisfied, for each traffic class j∈[1,N AVB ] is configured as Execute S55; if the third judgment result is satisfied, update the LA tuple If the value is Failed, TA information is sent from the output port of the switch to the receiving end according to the information transmission link, and the bandwidth resources allocated to the information flow on the output port are recovered, and S55 is executed. S55, determining whether the node generating the updated LA tuple is a switch, and obtaining a fourth determination result; If the fourth determination result is yes, the updated LA tuple is further transmitted along the reverse direction of the information transmission link, and the node receiving the updated LA tuple stores the updated LA tuple in its output port and executes S51; If the fourth judgment result is no, Is it Ready or Partial Failed, get the fifth judgment result, if the fifth judgment result is yes, the information flow f + Forward along the output ports of the nodes marked as Ready or Partial Failed in the LA tuple; If the fifth judgment result is no, terminate the information flow f + The sending process ends the dynamic bandwidth allocation method for the distributed time-sensitive network.

8. A dynamic bandwidth allocation device for a distributed time-sensitive network, characterized in that: The device comprises: a memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the dynamic bandwidth allocation method for a distributed time-sensitive network according to any one of claims 1 to 7.

9. A computer storable medium, characterized in that The computer storable medium stores computer instructions, and when the computer instructions are called by a computer, the computer instructions are used to execute the dynamic bandwidth allocation method for a distributed time-sensitive network according to any one of claims 1 to 7.

10. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the dynamic bandwidth allocation method for a distributed time-sensitive network according to any one of claims 1 to 7.

Images