Deterministic forwarding path reservation method and apparatus

Abstract

A deterministic forwarding path reservation method and apparatus. The method comprises: a controller acquires deterministic forwarding resource information of all links in a deterministic network; on the basis of the deterministic forwarding resource information, the controller calculates, in the deterministic network and for a service flow, a deterministic forwarding path that satisfies a quality of service (QOS) requirement of the service flow; and on the deterministic forwarding path, the controller reserves, for the service flow, a deterministic forwarding resource that satisfies the QOS requirement of the service flow. The method solves the problems of how a controller collects deterministic forwarding resources of a network and which types of resources are collected.

Description

Method and apparatus for reserving deterministic forwarding path

[0001] Cross-reference to Related Applications

[0002] The present disclosure is based on and claims priority from Chinese Patent Application No. CN202410740181.5 entitled “Method and apparatus for reserving deterministic forwarding path” filed on June 7, 2024, the disclosure of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0003] Embodiments of the present disclosure relate to the field of communications, and in particular, to a method and apparatus for reserving deterministic forwarding path. BACKGROUND

[0004] Generally, a deterministic path is a strict explicit path calculated by a centralized controller, and resources are reserved on nodes along the path to meet the Service Level Agreement (SLA) requirements of deterministic services. To calculate the deterministic forwarding path, the controller needs to collect not only the general topology state of the network but also the deterministic forwarding resource information of the network.

[0005] With the gradual maturity of the technology for establishing deterministic forwarding paths in Internet Protocol / Multi-Protocol Label Switching (IP / MPLS) networks, especially the queue scheduling mechanism of the forwarding plane, how the controller collects the deterministic forwarding resources of the network and which types of resources to collect have become problems that need to be solved urgently in current deterministic networks. SUMMARY

[0006] Embodiments of the present disclosure provide a method and apparatus for reserving deterministic forwarding path to at least solve the problem of how the controller collects the deterministic forwarding resources of the network and which types of resources to collect in the related art.

[0007] According to an embodiment of the present disclosure, a method for reserving deterministic forwarding path is provided, including: a controller obtaining deterministic forwarding resource information of all links in a deterministic network, the controller calculating a deterministic forwarding path that meets the Quality of Service (QoS) requirements of a service flow in the deterministic network based on the deterministic forwarding resource information, and the controller reserving deterministic forwarding resources that meet the QoS requirements of the service flow for the service flow on the deterministic forwarding path.

[0008] According to another embodiment of the present disclosure, a reservation method of a deterministic forwarding path is provided, comprising: a node in a deterministic network announcing deterministic forwarding resource information of each link to a controller, so that the controller calculates a deterministic forwarding path satisfying quality of service (QoS) requirements of a service flow in the deterministic network based on the deterministic forwarding resource information; the node reserving deterministic forwarding resources satisfying the QoS requirements of the service flow on the deterministic forwarding path for the service flow according to the deterministic forwarding path announced by the controller; and the node announcing remaining deterministic forwarding resources of each link after the deterministic forwarding path is reserved to the controller, so that the controller updates a database maintained locally by the controller.

[0009] According to another embodiment of the present disclosure, a reservation device of a deterministic forwarding path is provided, applied to a controller, comprising: an acquisition module configured to acquire deterministic forwarding resource information of all links in a deterministic network; a calculation module configured to calculate a deterministic forwarding path satisfying quality of service (QoS) requirements of a service flow in the deterministic network based on the deterministic forwarding resource information, the deterministic forwarding path being of a specified scheduling mechanism type; and a reservation module configured to reserve deterministic forwarding resources satisfying the QoS requirements of the service flow on the deterministic forwarding path for the service flow.

[0010] According to another embodiment of the present disclosure, a computer program product is also provided, comprising a computer program, the computer program being executed by a processor to implement the steps in any of the above method embodiments.

[0011] According to another embodiment of the present disclosure, a computer readable storage medium is also provided, the computer readable storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

[0012] According to another embodiment of the present disclosure, an electronic device is also provided, comprising a memory and a processor, the memory storing a computer program, and the processor being configured to execute the computer program to perform the steps in any of the above method embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0013] FIG. 1 is a hardware structure block diagram of a mobile terminal of a reservation method of a deterministic forwarding path according to an embodiment of the present disclosure;

[0014] FIG. 2 is a flowchart of a reservation method of a deterministic forwarding path according to an embodiment of the present disclosure (I);

[0015] FIG. 3 is a flowchart of a reservation method of a deterministic forwarding path according to an embodiment of the present disclosure (II);

[0016] FIG. 4 is a flowchart of a reservation method of a deterministic forwarding path according to an embodiment of the present disclosure (III);

[0017] FIG. 5 is a TLV structure diagram of a deterministic forwarding scheduling capability of a link according to an embodiment of the present disclosure;

[0018] FIG. 6 is a TLV structure diagram of a maximum reservable bandwidth of each capability level of a specific scheduling mechanism type of a link according to an embodiment of the present disclosure;

[0019] FIG. 7 is a TLV structure diagram of an idle bandwidth of each capability level of a specific scheduling mechanism type of a link according to an embodiment of the present disclosure;

[0020] FIG. 8 is a TLV structure diagram of a maximum reservable burst size of each capability level of a specific scheduling mechanism type of a link according to an embodiment of the present disclosure;

[0021] FIG. 9 is a TLV structure diagram of an idle burst size of each capability level of a specific scheduling mechanism type of a link according to an embodiment of the present disclosure;

[0022] FIG. 10 is a structure diagram of a deterministic network according to an embodiment of the present disclosure;

[0023] FIG. 11 is a structure diagram of a reservation device of a deterministic forwarding path according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0024] Hereinafter, the embodiments of the present disclosure will be described in detail with reference to the accompanying drawings and in conjunction with embodiments.

[0025] It should be noted that the terms "first", "second", and the like in the description and claims of the present disclosure and the above-described accompanying drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence.

[0026] The method embodiments provided in the embodiments of the present disclosure can be executed in a mobile terminal, a computer terminal, or a similar computing device. Taking an example of running on a computer terminal, FIG. 1 is a hardware structure block diagram of a mobile terminal of a reservation method of a deterministic forwarding path according to an embodiment of the present disclosure. As shown in FIG. 1, the computer terminal can include one or more (only one is shown in FIG. 1) processors 102 (the processor 102 can include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the above-mentioned computer terminal can further include a transmission device 106 for communication function and an input and output device 108. Those skilled in the art can understand that the structure shown in FIG. 1 is only schematic, which does not limit the structure of the above-mentioned computer terminal. For example, the computer terminal can further include more or less components than those shown in FIG. 1, or have a different configuration from that shown in FIG. 1.

[0027] The memory 104 can be used to store computer programs, such as software programs of application software and modules, such as the computer program corresponding to the method for reserving deterministic forwarding path in the embodiments of the present disclosure. The processor 102 can execute various functional applications and data processing, i.e., implement the above method, by running the computer programs stored in the memory 104. The memory 104 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include memories remotely arranged with respect to the processor 102, which can be connected to the mobile terminal through a deterministic network. Examples of the deterministic network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0028] The transmission device 106 is configured to receive or send data through a deterministic network. The specific examples of the deterministic network can include a wireless deterministic network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network interface controller (NIC) which can be connected to other deterministic network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module configured to communicate with the Internet in a wireless manner.

[0029] In the embodiments, a method for reserving a deterministic forwarding path running on the computer terminal is provided. FIG. 2 is a flowchart of the method for reserving a deterministic forwarding path according to the embodiments of the present disclosure (one), as shown in FIG. 2, the flowchart includes the following steps:

[0030] In step S202, the controller acquires deterministic forwarding resource information of all links in the deterministic network.

[0031] In step S204, the controller calculates a deterministic forwarding path satisfying the quality of service (QoS) requirement of the service flow in the deterministic network based on the deterministic forwarding resource information.

[0032] In the embodiments, the free resources of each link included in the deterministic forwarding path calculated by the controller should be able to satisfy the resource requirement of the service flow, i.e., the free bandwidth resource satisfies the bandwidth requirement of the service flow, and the free burst resource satisfies the burst requirement of the service flow.

[0033] In step S206, the controller reserves deterministic forwarding resources satisfying the QoS requirement of the service flow for the service flow on the deterministic forwarding path.

[0034] Through the above steps, the controller obtains the deterministic forwarding resource information of all links in the deterministic network, and the controller calculates a deterministic forwarding path for the service flow in the deterministic network based on the deterministic forwarding resource information, which meets the quality of service (QoS) requirement of the service flow, and the controller reserves deterministic forwarding resources for the service flow on the deterministic forwarding path, which meet the QoS requirement of the service flow, thereby solving the problem of how the controller collects the deterministic forwarding resources of the network and which types of resources are collected, and further realizing the extension of a suitable protocol for the controller to collect the deterministic forwarding resources from the network and for the controller to calculate the deterministic forwarding path.

[0035] In one embodiment, before step S202, one or more scheduling mechanisms can be configured on each link of each node in the deterministic network, each scheduling mechanism can support a single or multiple capability levels, and each capability level has corresponding deterministic forwarding resources (the deterministic forwarding resources can include one or both of the maximum reservable bandwidth resource or the maximum reservable burst resource). The types of scheduling mechanisms can include at least one of the following:

[0036] default type, asynchronous traffic shaping (ATS), credit-based shaper (CBS), combination of asynchronous traffic shaping and credit-based shaper (ATS+CBS), cyclic queuing and forwarding (CQF), bin cyclic queuing and forwarding (BCQF), earliest deadline first (EDF), timeslot queueing and forwarding (TQF), fair queue (FQ), and guaranteed latency based forwarding (gLBF).

[0037] In the embodiment, for the resource of each capability level of a certain scheduling mechanism on a link of a node, the node can further maintain the corresponding free resource or used resource of each capability level, such as unused bandwidth resource (UBan) and unused burst resource (UBur), according to the reservation of the resource by the traffic flow. Since the free resource is equal to the maximum reservable resource minus the used resource, one of the free resource and the used resource can be selected, and in the initial case, the free resource is equal to the maximum reservable resource.

[0038] In the embodiment, the node can flood the resource information of each capability level of each scheduling mechanism of each link in the deterministic network through an interior gateway protocol (IGP), such as maximum reservable bandwidth (MRBan), UBan, maximum reservable burst (MRBur), and UBur, so that each node in the network obtains the resource information of each capability level of each scheduling mechanism of each link in the network.

[0039] Since the free resource is equal to the maximum reservable resource in the initial case, only MRBan and MRBur need to be announced in the initial case, and UBan and UBur do not need to be announced, so as to reduce the announcement overhead.

[0040] In the embodiment, a border gateway protocol link-state (BGP-LS) session can be established between a first node and a controller in the deterministic network, and the first node can announce the resource information of each capability level of each scheduling mechanism of each link in the deterministic network to the controller through the BGP-LS, such as MRBan, UBan, MRBur, and UBur. After receiving the resource information, the controller updates a database locally maintained by the controller, for subsequent path computation.

[0041] Since the free resource is equal to the maximum reservable resource in the initial case, only MRBan and MRBur need to be announced in the initial case, and UBan and UBur do not need to be announced, so as to reduce the announcement overhead.

[0042] In the embodiment, the BGP-LS can be extended in at least one of the following manners:

[0043] A first link attribute, i.e., a DetNet Scheduling Capability TLV, is added in the BGP-LS attribute attribute to represent the deterministic forwarding scheduling capability of a link, and the TLV can appear multiple times depending on how many scheduling mechanism types are enabled on the corresponding link, wherein the first link attribute includes at least one of the following fields: a scheduling mechanism type (ST) supported by the link, a flag Flag-I (In-time Mode) indicating whether the in-time scheduling mode is supported, a flag Flag-O (On-time Mode) indicating whether the on-time scheduling mode is supported, scheduling capability level information corresponding to a specific scheduling mechanism type, and the like. When the scheduling mechanism type is asynchronous traffic shaping (ATS), credit-based shaping (CBS), or a combination of asynchronous traffic shaping and credit-based shaping (ATS+CBS), the capability level information includes one or more traffic classes. When the scheduling mechanism type is cyclic queuing and forwarding (CQF) or bin cyclic queuing and forwarding (BCQF), the capability level information includes one or more cycle lengths. When the scheduling mechanism type is earliest deadline first (EDF), the capability level information includes one or more delay levels. When the scheduling mechanism type is time-slot queuing and forwarding (TQF), the capability level information includes one or more TQF instances, and each TQF instance includes a corresponding arrangement cycle length, a number N of time slots contained in the arrangement cycle, and a number M of time slots contained in the scheduling cycle. When the scheduling mechanism type is a default type or fair queuing (FQ), the capability level information is empty.

[0044] A second link attribute, i.e., a DetNet Maximum Reservable Bandwidth TLV, is added in the BGP-LS attribute attribute to represent the maximum reservable bandwidth of each capability level of a specific scheduling mechanism type of a link, and the TLV can appear multiple times depending on how many scheduling mechanism types are enabled on the corresponding link, wherein the second link attribute includes at least one of the following fields: a scheduling mechanism type ST supported by the link, and maximum reservable bandwidth resources of a capability level of a specific scheduling mechanism type.

[0045] A third link attribute, DetNet Uneserved Bandwidth TLV, is added in the BGP-LS attribute attribute to indicate the free bandwidth of each capability level of a specific scheduling mechanism type of the link. The TLV can appear multiple times depending on how many scheduling mechanism types are enabled on the corresponding link, wherein the third link attribute includes at least one of the following fields: scheduling mechanism type ST supported by the link, free bandwidth Uneserved Bandwidth Resources of each capability level of a specific scheduling mechanism type;

[0046] A fourth link attribute, DetNet Maximum Reservable Burst TLV, is added in the BGP-LS attribute attribute to indicate the maximum reservable burst of each capability level of a specific scheduling mechanism type of the link. The TLV can appear multiple times depending on how many scheduling mechanism types are enabled on the corresponding link, wherein the fourth link attribute includes at least one of the following fields: scheduling mechanism type ST supported by the link, maximum reservable burst Maximum Reservable Burst Resources of each capability level of a specific scheduling mechanism type.

[0047] A fifth link attribute, DetNet Uneserved Burst TLV, is added in the BGP-LS attribute attribute to indicate the free burst of each capability level of a specific scheduling mechanism type of the link. The TLV can appear multiple times depending on how many scheduling mechanism types are enabled on the corresponding link, wherein the fifth link attribute includes at least one of the following fields: scheduling mechanism type ST supported by the link, free burst Uneserved Burst Resources of each capability level of a specific scheduling mechanism type.

[0048] In the embodiment, the controller reserving deterministic forwarding resources satisfying the QoS requirement of the service flow on the deterministic forwarding path for the service flow can include one of the following:

[0049] The controller reserves the bandwidth requirement and the burst requirement of the service flow in the free bandwidth resources and the free burst resources of the specified capability level of the specified scheduling mechanism of each link contained in the deterministic forwarding path in the database maintained locally by the controller;

[0050] The controller is notified of the head node of the deterministic forwarding path to initiate distributed signaling by the head node, reserve bandwidth and burst requirements of the service flow in the idle bandwidth resources and the idle burst resources of the specified scheduling mechanism of the specified capability level of each link contained in the deterministic forwarding path, and trigger the head node to notify the controller of the idle bandwidth resources and the idle burst resources of the specified scheduling mechanism of the specified capability level after reserving each link, so that the controller updates the database maintained locally.

[0051] Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the disclosure can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes a plurality of instructions for causing a terminal device (which can be a mobile phone, computer, server, or deterministic network device) to execute the method described in various embodiments of the disclosure.

[0052] Fig. 3 is a flowchart of a reservation method of a deterministic forwarding path according to an embodiment of the disclosure (two), as shown in Fig. 3, the flowchart includes the following steps:

[0053] Step S302, the nodes in the deterministic network notify the controller of the deterministic forwarding resource information of each link, so that the controller calculates a deterministic forwarding path that meets the quality of service (QoS) requirements of the service flow in the deterministic network based on the deterministic forwarding resource information.

[0054] Step S304, the nodes reserve deterministic forwarding resources that meet the QoS requirements of the service flow on the deterministic forwarding path according to the deterministic forwarding path notified by the controller.

[0055] Step S306, the nodes notify the controller of the remaining deterministic forwarding resources of each link after the deterministic forwarding path is reserved, so that the controller updates the database maintained locally.

[0056] In this embodiment, one or more scheduling mechanisms are configured on each link of the node, each scheduling mechanism supports one or more capability levels, and each capability level has corresponding deterministic forwarding resources.

[0057] Through the above steps, the nodes in the deterministic network advertise the deterministic forwarding resource information of each link to the controller, so that the controller calculates the deterministic forwarding path that meets the quality of service (QoS) requirement of the service flow in the deterministic network based on the deterministic forwarding resource information, and the nodes reserve the deterministic forwarding resource that meets the QoS requirement of the service flow on the deterministic forwarding path according to the deterministic forwarding path announced by the controller. The nodes advertise the deterministic forwarding resource remaining after the deterministic forwarding path is reserved to the controller, so that the controller updates the database maintained locally by the controller, solves the problem of how the controller collects the deterministic forwarding resource of the network and which types of resources are collected, and further implements the extension of a suitable protocol for the controller to collect the deterministic forwarding resource from the network and for the controller to calculate the deterministic forwarding path.

[0058] FIG. 4 is another flowchart (three) of a deterministic forwarding path reservation method according to an embodiment of the present disclosure, as shown in FIG. 4, the flowchart includes the following steps:

[0059] Step S401: configuring one or more scheduling mechanisms on each link of each node in the deterministic network.

[0060] Each scheduling mechanism can support a single or multiple capability levels, and each capability level has corresponding deterministic forwarding resources (including one or both of maximum reservable bandwidth resources or maximum reservable burst resources).

[0061] The configured scheduling mechanisms include but are not limited to the following:

[0062] 1. Default type, such as strict priority (SP): configurable to contain n priorities, for example, n equals 8, and to configure maximum reservable bandwidth (MRBan) resources for each priority. Excess subscription bandwidth is allowed, that is, the sum of MRBan of all priorities or MRBan of a certain priority is allowed to exceed the link bandwidth. It should be noted that SP is a queuing algorithm commonly used in current networks, which generally cannot guarantee deterministic forwarding QoS unless the network is particularly lightly loaded.

[0063] 2. Time Sensitive Networking (TSN) Asynchronous Traffic Shaping (ATS) (referring to IEEE 802.1Qbv): n traffic classes are configurable, e.g. n equals 8, and maximum reservable bandwidth and maximum reservable burst (MRBur) resources are configured for each traffic class. Over-subscription of bandwidth is not allowed, i.e. the sum of MRBan of all traffic classes is not allowed to exceed the link bandwidth. The MRBan and MRBur of a certain traffic class are the main determinants of the worst-case per-hop latency of that traffic class.

[0064] 3. TSN Credit-based shaper (CBS) (referring to IEEE 802.1Qav): n traffic classes are configurable, e.g. n equals 8, and maximum reservable bandwidth and maximum reservable burst resources are configured for each traffic class. Over-subscription of bandwidth is not allowed, i.e. the sum of MRBan of all traffic classes is not allowed to exceed the link bandwidth. The MRBan and MRBur of a certain traffic class are the main determinants of the worst-case per-hop latency of that traffic class.

[0065] 4. TSN ATS+CBS: n traffic classes are configurable, e.g. n equals 8, and maximum reservable bandwidth and maximum reservable burst resources are configured for each traffic class. Over-subscription of bandwidth is not allowed, i.e. the sum of MRBan of all traffic classes is not allowed to exceed the link bandwidth. The MRBan and MRBur of a certain traffic class are the main determinants of the worst-case per-hop latency of that traffic class. Note that ATS and CBS can be used in combination, and this disclosure considers this combination as a separate scheduling mechanism type as well, which is different from ATS or CBS in that the worst-case per-hop latency evaluation formula is different.

[0066] 5. TSN Cyclic Queuing and Forwarding (CQF) (referring to IEEE 802.1Qch): n CQF instances are configured, e.g., n equals 8, different CQF instances have different length of cycle duration (e.g., 10us). And maximum reservable bandwidth resource and maximum reservable burst resource are configured for each CQF instance. No over-subscription of bandwidth is allowed, i.e., the sum of MRBan of all CQF instances is not allowed to exceed the link bandwidth. For a certain CQF instance, its M RBur equals to the MRBan of the CQF instance multiplied by the cycle duration configured for the CQF instance.

[0067] 6. TSN Bin Cyclic Queuing and Forwarding (BCQF) (referring to IEEE 802.1Qdv): n BCQF instances are configured, e.g., n equals 8, different BCQF instances have different length of cycle duration (e.g., 10us). And maximum reservable bandwidth resource and maximum reservable burst resource are configured for each BCQF instance. No over-subscription of bandwidth is allowed, i.e., the sum of MRBan of all BCQF instances is not allowed to exceed the link bandwidth. For a certain BCQF instance, its M RBur equals to the MRBan of the BCQF instance multiplied by the cycle duration configured for the BCQF instance. Note that the difference between BCQF and CQF mainly lies in the cycle phase misalignment between adjacent nodes, and thus the per-hop latency evaluation formula is different.

[0068] 7. Earliest Deadline First (EDF) (referring to draft-peng-detnet-deadline-based-forwarding-09): n delay levels are configured, e.g., n equals 8, and maximum reservable bandwidth resource and maximum reservable burst resource are configured for each delay level, and it is required to guarantee that the MRBan and M RBur of all delay levels together satisfy the schedulability condition formula of EDF.

[0069] 8. Timeslot Queueing and Forwarding (TQF) (refer to draft-peng-detnet-packet-timeslot-mechanism-06): n TQF instances can be configured, e.g., n equals 8, different TQF instances have different length of orchestration period, and the orchestration period contains N (e.g., 1000) fixed length (e.g., 10us) round-robin timeslots, the first timeslot is numbered as #0, and the Nth timeslot is numbered as #N-1. In addition, each TQF instance also has a scheduling period implemented in hardware, and the scheduling period contains M (e.g., 10) timeslots. A maximum reservable bandwidth resource is configured for each TQF instance. Over-subscription of bandwidth is not allowed, i.e., the sum of MRBan of all TQF instances is not allowed to exceed the link bandwidth. For a certain TQF instance, each timeslot contained in its orchestration period has the same MRBur value, which is equal to the MRBan of the TQF instance multiplied by the length of the timeslot.

[0070] 9. Fair Queue (FQ) (refer to draft-joung-detnet-stateless-fair-queuing-02): The performance indicator of FQ scheduling is related to specific traffic flows. When implemented, a single MRBan and MRBur can be configured for the FQ scheduling mechanism as a whole (in order to have a unified expression of the resource model with other scheduling mechanisms, we regard this configuration of FQ as a single preset capability level), so that the total bandwidth consumption of all traffic flows using the FQ scheduling mechanism through the link does not exceed MRBan, and the total burst aggregation does not exceed MRBur. The setting of MRBur is mainly limited by the length of the FQ exclusive queue or buffer.

[0071] 10. Guaranteed Latency Based Forwarding (gLBF) (refer to draft-eckert-detnet-glbf-02): n traffic classes are configured, e.g. n equals 8, and maximum reservable bandwidth resources and maximum reservable burst resources are configured for each traffic class. No over-subscription of bandwidth is allowed, i.e. the sum of MRBan of all traffic classes is not allowed to exceed the link bandwidth. The MRBan and MRRur corresponding to a traffic class are the main determinants of the worst-case per-hop latency corresponding to the traffic class. Note that the difference between gLBF and ATS is that gLBF guarantees that the packet experiences the worst-case latency at each hop.

[0072] Step S402: For each resource of each capability level of a specific scheduling mechanism on the link of each node, the node further maintains the corresponding free resource or used resource of each capability level, such as unused bandwidth resource (denoted as UBan) and unused burst resource (denoted as UBur), according to the reservation of the resource by the traffic flow. Since the free resource is equal to the maximum reservable resource minus the used resource, only one of the free resource and the used resource is selected. Initially, the free resource is equal to the maximum reservable resource.

[0073] Step S403: The node can flood the resource information of each capability level of each scheduling mechanism of each link in the deterministic network, such as MRBan, UBan, MRRur and UBur, to each node in the deterministic network through an Interior Gateway Protocol (IGP), so that each node in the deterministic network obtains the resource information of each capability level of each scheduling mechanism of each link in the deterministic network.

[0074] Since the free resource is equal to the maximum reservable resource initially, only MRBan and MRRur need to be announced initially, and UBan and UBur do not need to be announced, so as to reduce the announcement overhead.

[0075] Step S404: In the deterministic network, a first node (e.g., node A) is selected to establish a Border Gateway Protocol Link-State (BGP-LS) session with the controller, so that node A can advertise the resource information of each link in the network, each scheduling mechanism and each level of capability through BGP-LS to the controller, such as MRBan, UBan, MRBur and UBur. After receiving the information, the controller updates the database maintained locally by the controller for subsequent path computation.

[0076] Since the idle resource is equal to the maximum reservable resource in the initial case, only MRBan and MRBur need to be advertised in the initial case, and UBan and UBur do not need to be advertised, so as to reduce the advertising overhead.

[0077] BGP-LS (refer to RFC9552) mainly defines a network layer reachability information (NLRI) type (i.e., link-state NLRI, which is further subdivided into node NLRI, link NLRI and prefix NLRI) and a BGP path attribute (i.e., BGP-LS attribute, which further includes various node attributes, link attributes and prefix attributes, which are respectively advertised with node NLRI, link NLRI and prefix NLRI). Generally, a network administrator will specify a node in each IGP domain to establish a BGP-LS session with the controller, and the node will advertise the link state data learned from the IGP domain to the controller.

[0078] The BGP-LS is extended in at least one of the following ways:

[0079] 1. A first link attribute, i.e., a deterministic forwarding scheduling capability DetNet Scheduling Capability type-length-value (TLV) of a link, is added to the BGP-LS attribute, and FIG. 5 is a TLV structure diagram of the deterministic forwarding scheduling capability of a link according to an embodiment of the present disclosure. As shown in FIG. 5, the TLV can appear multiple times, depending on how many types of scheduling mechanisms are enabled on the corresponding link.

[0080] Type: 2 bytes, whose value is assigned by Internet Assigned Numbers Authority (IANA) to indicate that the TLV is DetNet Scheduling Capability TLV.

[0081] Length: 2 bytes, indicating the length of the content of the TLV (excluding the Type and Length fields themselves).

[0082] Scheduling Type (ST): 1 byte, indicating the scheduling type supported by the link, and the value is as follows:

[0083] 0: indicating the default or unspecified scheduling mechanism, such as the general SP scheduling mechanism;

[0084] 1: indicating the TSN ATS scheduling mechanism (refer to IEEE 802.1Qbv);

[0085] 2: indicating the TSN CBS scheduling mechanism (refer to IEEE 802.1Qav);

[0086] 3: indicating the TSN ATS+CBS combined scheduling mechanism;

[0087] 4: indicating the TSN CQF scheduling mechanism (refer to IEEE 802.1Qch);

[0088] 5: indicating the TSN BCQF scheduling mechanism (refer to IEEE 802.1Qdv);

[0089] 6: indicating the EDF scheduling mechanism (refer to draft-peng-detnet-deadline-based-forwarding-09);

[0090] 7: indicating the TQF scheduling mechanism (refer to draft-peng-detnet-packet-timeslot-mechanism-06);

[0091] 8: indicating the FQ scheduling mechanism (refer to draft-joung-detnet-stateless-fair-queuing-02);

[0092] 9: indicating the gLBF scheduling mechanism (refer to draft-eckert-detnet-glbf-02);

[0093] 10-255: not defined.

[0094] Flags: occupies 1 byte, contains some flags, the following two flags are currently defined:

[0095] Flag-I (In-time Mode): occupies 1 bit, indicates whether the specific scheduling mechanism type supports in-time scheduling mode. When the flag is set to 1, it indicates that in-time scheduling mode is supported, and when the flag is set to 0, it indicates that in-time scheduling mode is not supported. In-time scheduling mode can be understood as sending messages as soon as possible before the delay limit.

[0096] Flag-O (On-time Mode): occupies 1 bit, indicates whether the specific scheduling mechanism type supports on-time scheduling mode. When the flag is set to 1, it indicates that on-time scheduling mode is supported, and when the flag is set to 0, it indicates that on-time scheduling mode is not supported. On-time scheduling mode can be understood as sending messages on time at the delay limit.

[0097] Scheduling Capability Levels: a variable-length field, which contains the capability level information corresponding to the specific scheduling mechanism type. Specifically as follows:

[0098] When ST = 0, the length of this field is 0, and no additional capability level information needs to be specified for the default scheduling mechanism.

[0099] When ST = 1, this field occupies 1 byte, which contains the number n of traffic classes supported by the TSN ATS scheduling mechanism. Let the first traffic class be 0, the last traffic class be n-1, and the scheduling priority of the traffic class from 0 to n-1 increase in turn.

[0100] When ST = 2, this field occupies 1 byte, which contains the number n of traffic classes supported by the TSN CBS scheduling mechanism. Let the first traffic class be 0, the last traffic class be n-1, and the scheduling priority of the traffic class from 0 to n-1 increase in turn.

[0101] When ST = 3, this field occupies 1 byte, and contains the number n of traffic classes supported by the TSN ATS+CBS hybrid scheduling mechanism. Let the first traffic class be 0, the last traffic class be n-1, and the scheduling priority of traffic classes from 0 to n-1 increase in turn.

[0102] When ST = 4, this field occupies n*2 bytes, and contains n cycle durations supported by the TSN CQF scheduling mechanism, each cycle duration occupying 2 bytes and being in units of us. For example, the cycle duration can be 10us, or 20us, etc.

[0103] When ST = 5, this field occupies n*2 bytes, and contains n cycle durations supported by the TSN BCQF scheduling mechanism, each cycle duration occupying 2 bytes and being in units of us. For example, the cycle duration can be 10us, or 20us, etc.

[0104] When ST = 6, this field occupies 6 bytes, and contains the minimum delay level (occupying 2 bytes, in units of us), the maximum delay level (occupying 2 bytes, in units of us), and the delay level interval (occupying 2 bytes, in units of us) supported by the EDF scheduling mechanism, i.e., the number n of delay levels supported = (the maximum delay level - the minimum delay level) / the delay level interval + 1. For example, the minimum delay level can be 10us, the maximum delay level can be 100us, and the delay level interval can be 10us.

[0105] When ST = 7, the field occupies n*8 bytes, containing n TQF instances supported by the TQF scheduling mechanism, each TQF instance occupying 8 bytes, and the 8 bytes specifically include an orchestration period length (occupying 4 bytes, in units of us), an orchestration period containing a number of time slots N (occupying 2 bytes), and a scheduling period containing a number of time slots M (occupying 2 bytes). For example, the orchestration period length (Orchestration Periold Length) can be 1000us, and the orchestration period Orchestration Periold can contain 100 time slots (that is, the length of each time slot can be derived as 10us), and the scheduling period Scheduling Periold can contain a smaller number of time slots, such as 10.

[0106] When ST = 8, the length of the field is 0, and no additional capability level information needs to be specified for the FQ scheduling mechanism.

[0107] When ST = 9, the field occupies 1 byte, containing the number n of traffic classes supported by the gLBF scheduling mechanism. The first traffic class is 0, the last traffic class is n-1, and the scheduling priority of the traffic classes from 0 to n-1 increases in turn.

[0108] 2. A second link attribute, i.e., DetNet Maximum Reservable Bandwidth TLV, is added to the BGP-LS attribute attribute, indicating the maximum reservable bandwidth of each capability level of a specific scheduling mechanism type of a link. FIG. 6 is a TLV structure diagram of the maximum reservable bandwidth of each capability level of a specific scheduling mechanism type of a link according to an embodiment of the present disclosure. As shown in FIG. 6, the TLV can appear multiple times, depending on how many scheduling mechanism types are enabled on the corresponding link.

[0109] Type: 2 bytes, whose value is assigned by the Internet Assigned Numbers Authority (IANA) to indicate that the TLV is a maximum reservable bandwidth TLV of each capability level of a specific scheduling mechanism type of a link.

[0110] Length: 2 bytes, indicating the length of the content of the TLV (excluding the Type and Length fields themselves).

[0111] ST: 1 byte, indicates the type of scheduling mechanism supported by the link, with the following values:

[0112] 0: indicates the default or unspecified scheduling mechanism, such as the general SP scheduling mechanism;

[0113] 1: indicates the TSN ATS scheduling mechanism (refer to IEEE 802.1Qbv);

[0114] 2: indicates the TSN CBS scheduling mechanism (refer to IEEE 802.1Qav);

[0115] 3: indicates the TSN ATS+CBS combined scheduling mechanism;

[0116] 4: indicates the TSN CQF scheduling mechanism (refer to IEEE 802.1Qch);

[0117] 5: indicates the TSN BCQF scheduling mechanism (refer to IEEE 802.1Qdv);

[0118] 6: indicates the EDF scheduling mechanism (refer to draft-peng-detnet-deadline-based-forwarding-09);

[0119] 7: indicates the TQF scheduling mechanism (refer to draft-peng-detnet-packet-timeslot-mechanism-06);

[0120] 8: indicates the FQ scheduling mechanism (refer to draft-joung-detnet-stateless-fair-queuing-02);

[0121] 9: indicates the gLBF scheduling mechanism (refer to draft-eckert-detnet-glbf-02);

[0122] 10-255: not defined.

[0123] Maximum Reservable Bandwidth Resources: a variable-length field containing the maximum reservable bandwidth for each capability level of a specific scheduling mechanism type. Details are as follows:

[0124] When ST = 0, the length of this field is 0, and no additional maximum reservable bandwidth information needs to be specified for the default scheduling mechanism.

[0125] When ST = 1, the field occupies n*4 bytes, containing the maximum reserved bandwidth MRBan (4 bytes, unit: bytes / s) of each of the traffic classes (n) of the TSN ATS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0126] When ST = 2, the field occupies n*4 bytes, containing the maximum reserved bandwidth MRBan (4 bytes, unit: bytes / s) of each of the traffic classes (n) of the TSN CBS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0127] When ST = 3, the field occupies n*4 bytes, containing the maximum reserved bandwidth MRBan (4 bytes, unit: bytes / s) of each of the traffic classes (n) of the TSN ATS+CBS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0128] When ST = 4, the field occupies n*6 bytes, containing the maximum reserved bandwidth information of each of the cycle durations (n) of the TSN CQF scheduling mechanism. The maximum reserved bandwidth information corresponding to each cycle duration occupies 6 bytes, which specifically contains the cycle duration (2 bytes, unit: us) and the maximum reserved bandwidth MRBan (4 bytes, unit: bytes / s) corresponding to the cycle duration.

[0129] When ST = 5, the field occupies n*6 bytes, containing the maximum reserved bandwidth information of each of the cycle durations (n) of the TSN BCQF scheduling mechanism. The maximum reserved bandwidth information corresponding to each cycle duration occupies 6 bytes, which specifically contains the cycle duration (2 bytes, unit: us) and the maximum reserved bandwidth MRBan (4 bytes, unit: bytes / s) corresponding to the cycle duration.

[0130] When ST = 6, the field occupies n*6 bytes, containing the maximum reserved bandwidth information of each delay level (n) of the EDF scheduling mechanism. The maximum reserved bandwidth information corresponding to each delay level occupies 6 bytes, and the 6 bytes specifically contain the delay level (occupying 2 bytes, in units of us) and the maximum reserved bandwidth MRBan (occupying 4 bytes, in units of bytes / s) corresponding to the delay level.

[0131] When ST = 7, the field occupies n*8 bytes, containing the maximum reserved bandwidth information of each TQF instance (n) of the TQF scheduling mechanism. The maximum reserved bandwidth information corresponding to each TQF instance occupies 8 bytes, and the 8 bytes specifically contain the orchestration period length (OPL) (occupying 4 bytes, in units of us) and the maximum reserved bandwidth MRBan (occupying 4 bytes, in units of bytes / s) corresponding to the OPL.

[0132] When ST = 8, the field occupies 4 bytes, containing the maximum reserved bandwidth MRBan of the FQ scheduling mechanism, in units of bytes / s.

[0133] When ST = 9, the field occupies n*4 bytes, containing the maximum reserved bandwidth MRBan (occupying 4 bytes, in units of bytes / s) corresponding to each traffic class (n) of the gLBF scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0134] (n) of the gLBF scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0135] 3. A third link attribute, i.e., DetNet Unreserved Bandwidth TLV, is added to the BGP-LS attribute attribute, indicating the idle bandwidth of each capability level of a specific scheduling mechanism type of a link. FIG. 7 is a TLV structure diagram of the idle bandwidth of each capability level of a specific scheduling mechanism type of a link according to an embodiment of the present disclosure. As shown in FIG. 7, the TLV can appear multiple times, depending on how many scheduling mechanism types are enabled on the corresponding link.

[0136] Type: 2 bytes, whose value is to be assigned by IANA to indicate that the TLV is the idle bandwidth TLV of each capability level of a specific scheduling mechanism type of a link.

[0137] Length: 2 bytes, indicating the length of the content of the TLV (excluding the Type and Length fields itself).

[0138] ST: 1 byte, indicating the type of the scheduling mechanism supported by the link, taking the following values:

[0139] 0: indicating the default or unspecified scheduling mechanism, such as the general SP scheduling mechanism;

[0140] 1: indicating the TSN ATS scheduling mechanism (refer to IEEE 802.1Qbv);

[0141] 2: indicating the TSN CBS scheduling mechanism (refer to IEEE 802.1Qav);

[0142] 3: indicating the TSN ATS+CBS combined scheduling mechanism;

[0143] 4: indicating the TSN CQF scheduling mechanism (refer to IEEE 802.1Qch);

[0144] 5: indicating the TSN BCQF scheduling mechanism (refer to IEEE 802.1Qdv);

[0145] 6: indicating the EDF scheduling mechanism (refer to draft-peng-detnet-deadline-based-forwarding-09);

[0146] 7: indicating the TQF scheduling mechanism (refer to draft-peng-detnet-packet-timeslot-mechanism-06);

[0147] 8: indicating the FQ scheduling mechanism (refer to draft-joung-detnet-stateless-fair-queuing-02);

[0148] 9: indicating the gLBF scheduling mechanism (refer to draft-eckert-detnet-glbf-02);

[0149] 10-255: not defined.

[0150] Uneserved Bandwidth Resources of each capability level of the specific scheduling mechanism type: a variable-length field containing the free bandwidth of each capability level of the specific scheduling mechanism type. Details are as follows:

[0151] When ST = 0, the length of this field is 0, and no additional free bandwidth information needs to be specified for the default scheduling mechanism.

[0152] When ST = 1, the field occupies n*4 bytes, containing the idle bandwidth UBan (4 bytes, unit: bytes / s) corresponding to each of the traffic classes (n) of the TSN ATS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0153] When ST = 2, the field occupies n*4 bytes, containing the idle bandwidth UBan (4 bytes, unit: bytes / s) corresponding to each of the traffic classes (n) of the TSN CBS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0154] When ST = 3, the field occupies n*4 bytes, containing the idle bandwidth UBan (4 bytes, unit: bytes / s) corresponding to each of the traffic classes (n) of the TSN ATS+CBS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0155] When ST = 4, the field occupies n*6 bytes, containing the idle bandwidth information of each of the cycle durations (n) of the TSN CQF scheduling mechanism, and the idle bandwidth information corresponding to each cycle duration occupies 6 bytes, which specifically contains the cycle duration (2 bytes, unit: us) and the idle bandwidth UBan (4 bytes, unit: bytes / s) corresponding to the cycle duration.

[0156] When ST = 5, the field occupies n*6 bytes, containing the idle bandwidth information of each of the cycle durations (n) of the TSN BCQF scheduling mechanism, and the idle bandwidth information corresponding to each cycle duration occupies 6 bytes, which specifically contains the cycle duration (2 bytes, unit: us) and the idle bandwidth UBan (4 bytes, unit: bytes / s) corresponding to the cycle duration.

[0157] When ST = 6, the field occupies n*6 bytes, containing the idle bandwidth information of all delay levels (n) of the EDF scheduling mechanism. The idle bandwidth information corresponding to each delay level occupies 6 bytes, and the 6 bytes specifically contain the delay level (occupying 2 bytes, in units of us) and the idle bandwidth UBan (occupying 4 bytes, in units of bytes / s) corresponding to the delay level.

[0158] When ST = 7, the field occupies n*8 bytes, containing the idle bandwidth information of all TQF instances (n) of the TQF scheduling mechanism. The idle bandwidth information corresponding to each TQF instance occupies 8 bytes, and the 8 bytes specifically contain the OPL (occupying 4 bytes, in units of us) and the idle bandwidth UBan (occupying 4 bytes, in units of bytes / s) corresponding to the OPL.

[0159] When ST = 8, the field occupies 4 bytes, containing the idle bandwidth UBan of the FQ scheduling mechanism, in units of bytes / s.

[0160] When ST = 9, the field occupies n*4 bytes, containing the idle bandwidth UBan (occupying 4 bytes, in units of bytes / s) corresponding to each traffic class of the gLBF scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0161] 4. A fourth link attribute, i.e., DetNet Maximum Reservable Burst TLV, is added in the BGP-LS attribute attribute, indicating the maximum reservable burst of each capability level of a specific scheduling mechanism type of a link. FIG. 8 is a TLV structure diagram of the maximum reservable burst of each capability level of a specific scheduling mechanism type of a link according to an embodiment of the present disclosure. As shown in FIG. 8, the TLV can appear multiple times, depending on how many scheduling mechanism types are enabled on the corresponding link.

[0162] Type: 2 bytes, the value is to be assigned by IANA, to indicate that the TLV is the maximum reservable burst TLV of each capability level of a specific scheduling mechanism type of a link.

[0163] Length: 2 bytes, indicating the length of the content of the TLV (excluding the Type and Length fields themselves).

[0164] ST: 1 byte, indicating the scheduling mechanism type supported by the link, taking the following values:

[0165] 0: indicates the default or unspecified scheduling mechanism, e.g. general SP scheduling mechanism;

[0166] 1: indicates the TSN ATS scheduling mechanism (refer to IEEE 802.1Qbv);

[0167] 2: indicates the TSN CBS scheduling mechanism (refer to IEEE 802.1Qav);

[0168] 3: indicates the TSN ATS+CBS combined scheduling mechanism;

[0169] 4: indicates the TSN CQF scheduling mechanism (refer to IEEE 802.1Qch);

[0170] 5: indicates the TSN BCQF scheduling mechanism (refer to IEEE 802.1Qdv);

[0171] 6: indicates the EDF scheduling mechanism (refer to draft-peng-detnet-deadline-based-forwarding-09);

[0172] 7: indicates the TQF scheduling mechanism (refer to draft-peng-detnet-packet-timeslot-mechanism-06);

[0173] 8: indicates the FQ scheduling mechanism (refer to draft-joung-detnet-stateless-fair-queuing-02);

[0174] 9: indicates the gLBF scheduling mechanism (refer to draft-eckert-detnet-glbf-02);

[0175] 10~255: not defined.

[0176] Maximum Reservable Burst Resources: variable length field containing the maximum reservable burst resources for each capability level of the specific scheduling mechanism type. Details are as follows:

[0177] When ST=0, the length of this field is 0, and no additional maximum reservable burst information needs to be specified for the default scheduling mechanism.

[0178] When ST = 1, the field occupies n*4 bytes, containing the maximum reservable burst amount MRBur (4 bytes, in bytes) corresponding to each of the traffic classes (n) of the TSN ATS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0179] When ST = 2, the field occupies n*4 bytes, containing the maximum reservable burst amount MRBur (4 bytes, in bytes) corresponding to each of the traffic classes (n) of the TSN CBS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0180] When ST = 3, the field occupies n*4 bytes, containing the maximum reservable burst amount MRBur (4 bytes, in bytes) corresponding to each of the traffic classes (n) of the TSN ATS+CBS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0181] When ST = 4, the field occupies n*6 bytes, containing the maximum reservable burst amount information of each of the cycle durations (n) of the TSN CQF scheduling mechanism, and the maximum reservable burst amount information corresponding to each cycle duration occupies 6 bytes, and the 6 bytes specifically contain the cycle duration (2 bytes, in us) and the maximum reservable burst amount MRBur (4 bytes, in bytes) corresponding to the cycle duration.

[0182] When ST = 5, the field occupies n*6 bytes, containing the maximum reservable burst amount information of each of the cycle durations (n) of the TSN BCQF scheduling mechanism, and the maximum reservable burst amount information corresponding to each cycle duration occupies 6 bytes, and the 6 bytes specifically contain the cycle duration (2 bytes, in us) and the maximum reservable burst amount MRBur (4 bytes, in bytes) corresponding to the cycle duration.

[0183] When ST = 6, the field occupies n*6 bytes, containing the maximum reserved burst information of each delay level (n) of the EDF scheduling mechanism. The maximum reserved burst information corresponding to each delay level occupies 6 bytes, and the 6 bytes specifically contain the delay level (occupying 2 bytes, in units of us) and the maximum reserved burst M RBur (occupying 4 bytes, in units of bytes) corresponding to the delay level.

[0184] When ST = 7, the field occupies n*8 bytes, containing the maximum reserved burst information of each TQF instance (n) of the TQF scheduling mechanism. The maximum reserved burst information corresponding to each TQF instance occupies 8 bytes, and the 8 bytes specifically contain the OPL (occupying 4 bytes, in units of us) and the maximum reserved burst M RBur (occupying 4 bytes, in units of bytes) corresponding to the single time slot of the OPL. It should be noted that the N time slots included in the orchestration period have the same M RBur, so there is no need to specify a separate M RBur for each time slot.

[0185] When ST = 8, the field occupies 4 bytes, containing the maximum reserved burst M RBur of the FQ scheduling mechanism, in units of bytes.

[0186] When ST = 9, the field occupies n*4 bytes, containing the maximum reserved burst M RBur (occupying 4 bytes, in units of bytes) corresponding to each traffic class (n) of the gLBF scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0187] 5. A fifth link attribute, i.e., DetNet Unreserved Burst TLV, is added in the BGP-LS attribute attribute, indicating the idle burst of each capability level of the specific scheduling mechanism type of the link. FIG. 9 is a TLV structure diagram of the idle burst of each capability level of the specific scheduling mechanism type of the link according to an embodiment of the present disclosure. As shown in FIG. 9, the TLV can appear multiple times, depending on how many scheduling mechanism types are enabled on the corresponding link.

[0188] Type: 2 bytes, the value is to be assigned by IANA, to indicate that the present TLV is the idle burst TLV of each capability level of the specific scheduling mechanism type of the link.

[0189] Length: 2 bytes, indicating the length of the content of this TLV (excluding the Type and Length fields themselves).

[0190] ST: 1 byte, indicating the type of scheduling mechanism supported by the link, taking the following values:

[0191] 0: indicating the default or unspecified scheduling mechanism, such as the general SP scheduling mechanism;

[0192] 1: indicating the TSN ATS scheduling mechanism (refer to IEEE 802.1Qbv);

[0193] 2: indicating the TSN CBS scheduling mechanism (refer to IEEE 802.1Qav);

[0194] 3: indicating the TSN ATS+CBS combined scheduling mechanism;

[0195] 4: indicating the TSN CQF scheduling mechanism (refer to IEEE 802.1Qch);

[0196] 5: indicating the TSN BCQF scheduling mechanism (refer to IEEE 802.1Qdv);

[0197] 6: indicating the EDF scheduling mechanism (refer to draft-peng-detnet-deadline-based-forwarding-09);

[0198] 7: indicating the TQF scheduling mechanism (refer to draft-peng-detnet-packet-timeslot-mechanism-06);

[0199] 8: indicating the FQ scheduling mechanism (refer to draft-joung-detnet-stateless-fair-queuing-02);

[0200] 9: indicating the gLBF scheduling mechanism (refer to draft-eckert-detnet-glbf-02);

[0201] 10-255: not defined.

[0202] Uneserved Burst Resources: a variable-length field containing the unreserved burst resources for each capability level of the specific scheduling mechanism type. Details are as follows:

[0203] When ST = 0, the length of this field is 0, and no additional unreserved burst resource information needs to be specified for the default scheduling mechanism.

[0204] When ST = 1, the field occupies n*4 bytes, containing the idle burst amount UBur (4 bytes, unit: bytes) corresponding to each of the traffic classes (n) of the TSN ATS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0205] When ST = 2, the field occupies n*4 bytes, containing the idle burst amount UBur (4 bytes, unit: bytes) corresponding to each of the traffic classes (n) of the TSN CBS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0206] When ST = 3, the field occupies n*4 bytes, containing the idle burst amount UBur (4 bytes, unit: bytes) corresponding to each of the traffic classes (n) of the TSN ATS+CBS scheduling mechanism, in turn from traffic class 0 to traffic class n-1.

[0207] When ST = 4, the field occupies n*6 bytes, containing the idle burst amount information of each of the cycle durations (n) of the TSN CQF scheduling mechanism, and the idle burst amount information corresponding to each cycle duration occupies 6 bytes, which specifically contains the cycle duration (2 bytes, unit: us) and the idle burst amount UBur (4 bytes, unit: bytes) corresponding to the cycle duration.

[0208] When ST = 5, the field occupies n*6 bytes, containing all cycle duration (n) of TSN BCQF scheduling mechanism respectively corresponding idle burst information, each cycle duration corresponding idle burst information occupies 6 bytes, and the 6 bytes specifically contain cycle duration (occupying 2 bytes, unit: us) and idle burst UBur (occupying 4 bytes, unit: bytes) corresponding to the cycle duration. When ST = 6, the field occupies n*6 bytes, containing all delay level (n) of EDF scheduling mechanism respectively corresponding idle burst information, each delay level corresponding idle burst information occupies 6 bytes, and the 6 bytes specifically contain delay level (occupying 2 bytes, unit: us) and idle burst UBur (occupying 4 bytes, unit: bytes) corresponding to the delay level.

[0209] When ST = 7, the field occupies 4+k*6 bytes, containing k (k≤N) time slots of specific TQF instance of TQF scheduling mechanism corresponding idle burst information, that is, specifically containing OPL (occupying 4 bytes, unit: us), and each time slot corresponding time slot number (occupying 2 bytes) and its idle burst UBur (occupying 4 bytes, unit: bytes).

[0210] When ST = 8, the field occupies 4 bytes, containing idle burst UBur of FQ scheduling mechanism, unit: bytes.

[0211] When ST = 9, the field occupies n*4 bytes, containing all traffic class (n) of gLBF scheduling mechanism respectively corresponding idle burst UBur (occupying 4 bytes, unit: bytes), in turn from traffic class 0 to traffic class n-1.

[0212] It should be noted that in specific applications, the above-mentioned BGP-LS extension is not limited to importing link state data from the IGP domain and reporting to the controller, but also can import data from nodes or links not belonging to the IGP domain, such as inter-domain links connecting two autonomous systems (Autonomous System, referred to as AS) also have the deterministic forwarding resource of the present patent and can be imported to BGP-LS and reported to the controller.

[0213] Step S405: path calculation and deterministic forwarding resource reservation.

[0214] The controller can calculate a deterministic forwarding path satisfying the QoS requirement of the traffic flow in the deterministic network based on the collected deterministic forwarding resource information of all links in the deterministic network, and the deterministic forwarding path adopts a specified scheduling mechanism type. The free resource of each link contained in the calculated deterministic forwarding path should be able to satisfy the resource requirement of the traffic flow, that is, the free bandwidth resource satisfies the bandwidth requirement of the traffic flow, and the free burst resource satisfies the burst requirement of the traffic flow.

[0215] More specifically, the controller can specify a specific capability level for each link contained in the deterministic forwarding path. Generally, all links contained in the deterministic forwarding path can use the same capability level (such as the same traffic class or the same cycle duration, etc.), and of course, different capability levels can also be specified. For each link contained in the deterministic forwarding path, it is checked whether the free bandwidth resource corresponding to the specific capability level specified for use satisfies the bandwidth requirement of the traffic flow, and whether the free burst resource satisfies the burst requirement of the traffic flow.

[0216] If the check passes, the deterministic forwarding path calculation is successful. The controller can deduct (i.e., reserve) the bandwidth requirement and the burst requirement of the traffic flow from the free bandwidth resource and the free burst resource of the specified capability level of the specified scheduling mechanism for each link contained in the deterministic forwarding path in the database maintained locally by the controller, or the controller can notify the head node of the deterministic forwarding path, and then the head node initiates distributed signaling (such as Resource Reservation Protocol Traffic Engineering (RSVP-TE)) to implement the above deduction action in the deterministic network, that is, the deduction is implemented by the node to which each link in the deterministic forwarding path belongs. For the latter case, it will cause the free resources of these links in the deterministic network to change, which will again trigger node A to notify the controller of the free resource information of the specified capability level of the specified scheduling mechanism of these links through BGP-LS, and the controller will update the database maintained locally after receiving the information.

[0217] Embodiment I

[0218] Embodiment I enables TQF scheduling mechanism on each link in the deterministic network, the controller collects the deterministic forwarding resources related to the TQF scheduling mechanism of the deterministic network, and then the controller calculates a deterministic forwarding path adopting the TQF scheduling mechanism for the traffic flow in the deterministic network and reserves the deterministic forwarding resources for the traffic flow.

[0219] Figure 10 is a schematic diagram of a deterministic network according to an embodiment of the present disclosure, as shown in Figure 10, containing 6 nodes and one controller, and BGP-LS sessions are established between node S and the controller to advertise the link state data of the deterministic network. Assume that all the outgoing interfaces of the nodes in the deterministic network are configured with the TQF instance with an orchestration period length of 1000us (by default, the on-time mode is adopted), and the time slot length in the orchestration period is 10us, i.e., the orchestration period contains N = 100 time slots numbered 0-99. The scheduling period contains M = 10 time slots. Assume that the propagation delay of all the links is 0, the processing delay and forwarding delay of the messages in the nodes are 0, and the phases of the orchestration periods of all the outgoing interfaces are aligned.

[0220] Assume that the maximum reservable bandwidth MRBan = 1 Gbps is configured on the TQF instance on all the outgoing interfaces of the nodes, then the maximum reservable burst quantity MRBur = 1 Gbps * 10us = 10000 bits of a single time slot of the TQF. Initially, the idle burst quantity UBur is equal to MRBur. That is, initially, node S will advertise the following information of each unidirectional link to the controller through BGP-LS:

[0221] Assume that the controller or the head node S needs to calculate a candidate path path_SD of a periodic service flow (whose source is src and destination is dst) with a path S-A-C-E-D, and the expected end-to-end delay is 200us, and the node resident delay budget of each hop will be 200 / 4 = 50us. The bandwidth required by the service flow is 1 Mbps, and the burst quantity of the service flow is 1000 bits.

[0222] Assume that the burst of the service flow always arrives in time slot 0 of node S periodically, then the controller can reserve burst quantity resources in time slot 5 on port-1 of node S, in time slot 10 on port-4 of node A, in time slot 15 on port-10 of node C, and in time slot 20 on port-14 of node E for the service flow on port-1 of node S, so as to achieve the node resident delay of 50us at each hop. In embodiment one, the reservation actions on all the outgoing interfaces are successful, because the UBur of the reserved time slots of each outgoing interface of the OPL 1000us instance is greater than the burst quantity requirement of the service flow, so the path path_SD can be successfully established.

[0223] The controller performs the operation of deducting the resource requirement of the service flow from the free resource in the local data maintained by it for each outgoing interface contained in the path_SID. For example:

[0224] For port-1, the UBan of the OPL 1000us instance of the TQF scheduling mechanism will be updated to 999Mbps after deduction, and the UBur of the time slot 5 of the OPL 1000us instance of the TQF scheduling mechanism will be updated to 9000 bits after deduction;

[0225] For port-4, the UBan of the OPL 1000us instance of the TQF scheduling mechanism will be updated to 999Mbps after deduction, and the UBur of the time slot 10 of the OPL 1000us instance of the TQF scheduling mechanism will be updated to 9000 bits after deduction;

[0226] For port-10, the UBan of the OPL 1000us instance of the TQF scheduling mechanism will be updated to 999Mbps after deduction, and the UBur of the time slot 15 of the OPL 1000us instance of the TQF scheduling mechanism will be updated to 9000 bits after deduction;

[0227] For port-14, the UBan of the OPL 1000us instance of the TQF scheduling mechanism will be updated to 999Mbps after deduction, and the UBur of the time slot 20 of the OPL 1000us instance of the TQF scheduling mechanism will be updated to 9000 bits after deduction;

[0228] Alternatively, the controller can also advertise the deterministic forwarding path to the head node S of the path, and then the head node S initiates distributed signaling (such as RSVP-TE) to implement the above deduction action in the deterministic network, i.e. the deduction is implemented by the home node of each link in the deterministic forwarding path (i.e. S, A, C, E) respectively. In this case, the node S will again advertise the updated free resource information of these links to the controller through BGP-LS, for example, the following information related to the outgoing interface port-1 is advertised:

[0229] The advertisement for other outgoing interfaces is similar. The controller will update the database maintained by it after receiving.

[0230] Embodiment Two

[0231] Embodiment two is to enable TSN ATS scheduling mechanism on each link in deterministic network (embodiments for TSN CBS, TSN ATS+CBS, gLBF are similar), the controller collects deterministic forwarding resources related to TSN ATS scheduling mechanism of the deterministic network, and then the controller calculates a deterministic forwarding path for the service flow in the deterministic network using TSN ATS scheduling mechanism and reserves specific deterministic forwarding resources for the service flow.

[0232] The deterministic network as shown in FIG. 10 assumes that all the outgoing interfaces of the nodes in the deterministic network are configured to use TSN ATS scheduling mechanism, and support 8 traffic classes (from 0 to 7, the scheduling priority increases in turn). It is assumed that the propagation delay of all links is 0, and the processing delay and forwarding delay of the message in the node are both 0.

[0233] It is assumed that the maximum reservable bandwidth MRBan=1Gbps and the maximum reservable burst MRBur=10000bits are configured for each traffic class on the outgoing interface of all nodes. Initially, the idle burst UBur is equal to MRBur. That is, initially, node S will advertise the following information of each unidirectional link to the controller through BGP-LS:

[0234] It is assumed that the controller or the head node S needs to calculate a candidate path path_SD for a periodic service flow (its source is src and its destination is dst) passing through S-A-C-E-D, and the expected end-to-end delay is 200us, and the average per-hop will share 200 / 4=50us of the node resident delay budget. The bandwidth required by the service flow is 1Mbps, and the burst of the service flow is 1000bits.

[0235] According to the per-hop delay evaluation formula of TSN ATS, it is assumed that the controller reserves deterministic forwarding resources of traffic class 3 at each hop of the candidate path for the service flow to achieve 50us of node resident delay at each hop. In embodiment two, the reservation action on all the above outgoing interfaces is successful, because the UBan or UBur of traffic class 3 of each outgoing interface is greater than the bandwidth or burst requirement of the service flow, so the path path_SD can be successfully established.

[0236] The controller performs the operation of deducting the resource requirement of the service flow from the idle resources in the local data it maintains for each outgoing interface included in path_SID. For example:

[0237] For port-1, the UBan of traffic class 3 of its TSN ATS scheduling mechanism will be updated to 999Mbps after deduction, and the UBur of traffic class 3 of its TSN ATS scheduling mechanism will be updated to 9000bits after deduction.

[0238] The same is true for port-4, port-10, and port-14.

[0239] Alternatively, the controller can also advertise the deterministic forwarding path to the head node S of the path, and then the head node S initiates distributed signaling (such as RSVP-TE) to implement the above deduction action in the deterministic network, i.e., each link in the deterministic forwarding path is implemented by the home node of the link (i.e., S, A, C, E) respectively. In this case, node S will again advertise the updated free resource information of these links to the controller through BGP-LS, for example, the following information related to interface port-1 is advertised:

[0240] The announcement for other outgoing interfaces is similar. After receiving, the controller will update the database maintained locally.

[0241] Embodiment Three

[0242] Embodiment three is to enable FQ scheduling mechanism on each link in the deterministic network. The controller collects the deterministic forwarding resources related to the FQ scheduling mechanism of the deterministic network, and then the controller calculates a deterministic forwarding path with FQ scheduling mechanism for the traffic flow in the deterministic network and reserves the specific process of deterministic forwarding resources for the traffic flow.

[0243] As shown in FIG. 10, the deterministic network, assuming that all outgoing interfaces of nodes in the deterministic network are configured to use FQ scheduling mechanism, and are all configured with FQ maximum reservation bandwidth MRBan = 1Gbps and maximum reservation burst MRBur = 10000bits. Initially, the free burst UBur is equal to MRBur. That is, initially, node S will advertise the following information of each unidirectional link to the controller through BGP-LS:

[0244] Suppose the controller or head node S needs to compute a candidate path path_SD for a periodic traffic flow (whose source is src and destination is dst) with a path S-A-C-E-D, and the expected end-to-end latency is 4ms, and the average per-hop will share 1ms of the in-node resident latency budget. The bandwidth required by the traffic flow is 1Mbps, and the burst of the traffic flow is 1000bits. According to the per-hop latency evaluation formula of FQ, the per-hop latency is related to the burst of the traffic flow and the service rate, which is about 1000bits / 1Mbps = 1ms. In embodiment three, the reservation actions on all the egress interfaces above can be successful, because the UBan or UBur of the FQ scheduling mechanism of each egress interface is greater than the bandwidth or burst requirement of the traffic flow, so the path path_SD can be successfully established.

[0245] The controller performs the operation of deducting the resource requirement of the traffic flow from the free resources in the local data maintained by the controller for each egress interface contained in path_SID. For example:

[0246] For port-1, the UBan of the FQ scheduling mechanism will be updated to 999Mbps after deduction, and the UBur of the FQ scheduling mechanism will be updated to 9000bits after deduction;

[0247] The same is true for port-4, port-10, and port-14.

[0248] Alternatively, the controller can also advertise the deterministic forwarding path to the head node S of the path, and then the head node S initiates distributed signaling (such as RSVP-TE) to implement the above deduction action in the deterministic network, that is, each link in the deterministic forwarding path is implemented by the node (that is, S, A, C, E) belonging to the link to deduct. In this case, the node S will again advertise the updated free resource information of the link to the controller through BGP-LS, for example, the following information related to egress interface port-1 is advertised:

[0249] The advertisement for other egress interfaces is similar. The controller will update the database maintained locally after receiving it.

[0250] Embodiment four

[0251] Embodiment four is the specific process of enabling TSN CQF scheduling mechanism on each link in the deterministic network (the embodiment for TSN BCQF is similar), the controller collects the deterministic forwarding resources related to the TSN CQF scheduling mechanism of the deterministic network, and then the controller computes a deterministic forwarding path using the TSN CQF scheduling mechanism for the traffic flow in the deterministic network and reserves deterministic forwarding resources for the traffic flow.

[0252] The deterministic network as shown in FIG. 10 assumes that all the egress interfaces of all the nodes in the deterministic network are configured to use the TSN CQF scheduling mechanism, and support 8 cycle durations (e.g., 10us, 20us,..., 80us). It is assumed that the propagation delay of all the links is 0, and the processing delay and the forwarding delay of the packet in the node are both 0.

[0253] It is assumed that the CQF instance corresponding to each length of cycle duration on the egress interface of all the nodes is configured with a maximum reservable bandwidth MRBan = 1 Gbps and a maximum reservable burst MRBur = 10000 bits. Initially, the idle burst UBur is equal to MRBur. That is, initially, the node S will advertise the following information of each unidirectional link to the controller through BGP-LS:

[0254] It is assumed that the controller or the head node S needs to calculate a candidate path path_SD of a periodic service flow (whose source is src and whose destination is dst) with an approach S-A-C-E-D, and the expected end-to-end delay is 200us. The average per-hop will share 200 / 4 = 50us of the node resident delay budget. The bandwidth required by the service flow is 1 Mbps, and the burst of the service flow is 1000 bits.

[0255] According to the per-hop delay evaluation formula of TSN CQF, the controller reserves deterministic forwarding resources with a cycle duration of 50us at each hop of the candidate path of the service flow to achieve a node resident delay of 50us at each hop. In embodiment four, the reservation action on all the egress interfaces described above is successful because the UBan or UBur of each egress interface with a cycle duration of 50us is greater than the bandwidth or burst requirement of the service flow, so the path path_SD can be successfully established.

[0256] The controller performs the operation of deducting the resource requirement of the service flow from the idle resource in the local data it maintains for each egress interface included in path_SID. For example:

[0257] For port-1, the UBan of the TSN CQF scheduling mechanism with a cycle duration of 50us will be updated to 999Mbps after deduction, and the UBur of the TSN CQF scheduling mechanism with a cycle duration of 50us will be updated to 9000 bits after deduction;

[0258] The same is true for port-4, port-10, and port-14.

[0259] Alternatively, the controller can also advertise the deterministic forwarding path to the head node S of the path, and then the head node S initiates distributed signaling (such as RSVP-TE) to implement the above deduction action in the deterministic network, that is, the deduction is implemented by the home node of each link in the deterministic forwarding path (that is, S, A, C, and E). In this case, the node S will again advertise the updated free resource information of these links to the controller through BGP-LS, for example, the following information related to interface port-1 is advertised:

[0260] The advertisement for other egress interfaces is similar. After receiving, the controller will update the database maintained locally.

[0261] Embodiment five

[0262] Embodiment five is to enable the EDF scheduling mechanism on each link in the deterministic network, the controller collects the deterministic forwarding resources related to the EDF scheduling mechanism of the deterministic network, and then the controller calculates a deterministic forwarding path using the EDF scheduling mechanism for the service flow in the deterministic network and reserves the specific process of the deterministic forwarding resources for the service flow.

[0263] As shown in FIG. 10, the deterministic network assumes that all egress interfaces of all nodes in the deterministic network are configured to use the EDF scheduling mechanism, and support 8 delay levels (for example: 10us, 20us,..., 80us). It is assumed that the propagation delay of all links is 0, and the processing delay and forwarding delay of the message in the node are both 0.

[0264] It is assumed that the maximum reservable bandwidth MRBan=1Gbps and the maximum reservable burst MRBur=10000bits are configured for each delay level on all egress interfaces of all nodes. Initially, the free burst UBur is equal to MRBur. That is, initially, the node S will advertise the following information of each unidirectional link to the controller through BGP-LS:

[0265] It is assumed that the controller or the head node S needs to calculate a candidate path path_SD for a periodic service flow (whose source is src and destination is dst) passing through S-A-C-E-D, and the expected end-to-end delay is 200us, and the average per-hop will share 200 / 4=50us of the node resident delay budget. The bandwidth required by the service flow is 1Mbps, and the burst of the service flow is 1000bits.

[0266] According to the per-hop latency evaluation formula of EDF, the controller reserves deterministic forwarding resources of delay level 50us for the traffic flow at each hop of the candidate path, so as to achieve a node resident latency of 50us at each hop. In Embodiment Five, the reservation actions on all the egress interfaces mentioned above are successful, because the UBan or UBur of delay level 50us of each egress interface is greater than the bandwidth or burst requirement of the traffic flow, so the path path_SD can be successfully established.

[0267] The controller performs the operation of deducting the resource requirement of the traffic flow from the idle resources in the local data it maintains, for each egress interface contained in path_SID. For example:

[0268] For port-1, the UBan of delay level 50us of the EDF scheduling mechanism will be updated to 999Mbps after deduction, and the UBur of delay level 50us of the EDF scheduling mechanism will be updated to 9000bits after deduction;

[0269] The same applies to port-4, port-10 and port-14.

[0270] Alternatively, the controller can also notify the head node S of the deterministic forwarding path, and then the head node S initiates distributed signaling (such as RSVP-TE) to implement the above deduction action in the deterministic network, i.e. each link in the deterministic forwarding path is implemented by the node belonging to it (i.e. S, A, C, E). In this case, the node S will again notify the controller of the updated idle resource information of these links through BGP-LS, for example, the following information related to egress interface port-1 is notified:

[0271] The notification for other egress interfaces is similar. After receiving, the controller will update the database it maintains locally.

[0272] Those skilled in the art can clearly understand that the method according to the above-mentioned embodiments can be realized by means of software and necessary general hardware platforms, and of course, can also be realized by hardware, but in many cases, the former is a better implementation. Based on such understanding, the technical solutions of the present disclosure can be embodied in the form of a software product, which is stored in a storage medium (such as a ROM / RAM, a magnetic disk, or an optical disk) and includes a plurality of instructions for causing a terminal device (which can be a mobile phone, a computer, a server, or a deterministic network device) to execute the methods of various embodiments of the present disclosure.

[0273] In this embodiment, a deterministic forwarding path reservation device is also provided, which is configured to implement the above-mentioned embodiments and preferred embodiments, and the description of which has been made above. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware is also possible and contemplated.

[0274] FIG. 11 is a structural block diagram of a deterministic forwarding path reservation device according to an embodiment of the present disclosure. As shown in FIG. 11, the deterministic forwarding path reservation device 1100 includes:

[0275] An obtaining module 1102 is configured to obtain deterministic forwarding resource information of all links in the deterministic network;

[0276] A calculating module 1104 is configured to calculate, based on the deterministic forwarding resource information, a deterministic forwarding path that satisfies a quality of service (QoS) requirement of a service flow and uses a specified scheduling mechanism type in the deterministic network for the service flow;

[0277] A reserving module 1106 is configured to reserve deterministic forwarding resources that satisfy the QoS requirement of the service flow on the deterministic forwarding path for the service flow.

[0278] In this embodiment, the free bandwidth resources of the deterministic forwarding path satisfy the bandwidth requirement of the service flow, and the free burst resources of the deterministic forwarding path satisfy the burst requirement of the service flow.

[0279] In one embodiment, the deterministic forwarding path reservation device 1100 further includes:

[0280] A configuring module is configured to configure one or more scheduling mechanisms on each link of each node in the deterministic network, wherein each scheduling mechanism supports one or more capability levels, and each capability level has corresponding deterministic forwarding resources.

[0281] In the embodiment, the types of the scheduling mechanism include at least one of the following:

[0282] default type, asynchronous traffic shaping (ATS), credit-based shaping (CBS), combination of asynchronous traffic shaping and credit-based shaping (ATS+CBS), cyclic queuing and forwarding (CQF), bin cyclic queuing and forwarding (BCQF), earliest deadline first (EDF), time-slot queuing and forwarding (TQF), fair queuing (FQ), guaranteed low-latency-based forwarding (gLBF).

[0283] In the embodiment, the configuration module includes:

[0284] The maintaining submodule is configured to maintain, according to the reservation of the resources by the traffic flows, the free resources or the used resources of each capability level of each scheduling mechanism on each link of each node, wherein the free resources include free bandwidth resources and free burst resources, and the free resources are equal to the maximum reservable resources in the initial case.

[0285] In the embodiment, the maintaining submodule includes:

[0286] The flooding sub-submodule is configured to cause each node in the deterministic network to obtain the resource information of each capability level of each scheduling mechanism on each link in the deterministic network by flooding, by the internal gateway protocol (IGP), the resource information of each capability level of each scheduling mechanism on each link of the node in the deterministic network.

[0287] In the embodiment, the flooding sub-submodule is further configured to, in the initial case, flood only the maximum reservable bandwidth resources and the maximum reservable burst resources of each capability level of each scheduling mechanism on each link of the node.

[0288] In the embodiment, the configuration module further includes:

[0289] The announcing submodule is configured to establish a border gateway protocol link state (BGP-LS) session between a selected first node and the controller in the deterministic network, so that the first node announces, by the BGP-LS, the resource information of each capability level of each scheduling mechanism on each link in the deterministic network to the controller, and the controller updates a locally maintained database after receiving the resource information, for subsequent path computation.

[0290] In the embodiment, the announcing submodule is further configured to, in the initial case, announce only the maximum reservable bandwidth resources and the maximum reservable burst resources.

[0291] In the embodiment, the announcing submodule is further configured to extend the BGP-LS by at least one of the following:

[0292] The first link attribute is added in the announced BGP-LS attribute to represent the deterministic forwarding scheduling capability of the link, wherein the first link attribute includes at least one of the following fields: a scheduling mechanism type supported by the link, a flag indicating whether a just-in-time scheduling mode is supported, a flag indicating whether a near-just-in-time scheduling mode is supported, capability level information corresponding to a specific scheduling mechanism type, and the like. When the scheduling mechanism type is asynchronous traffic shaping ATS, credit-based shaping CBS, asynchronous traffic shaping combined with credit-based shaping ATS+CBS, or guaranteed low- delay forwarding gLBF, the capability level information includes one or more traffic classes. When the scheduling mechanism type is cyclic queuing and forwarding CQF or bin cyclic queuing and forwarding BCQF, the capability level information includes one or more cycle lengths. When the scheduling mechanism type is earliest deadline first EDF, the capability level information includes one or more delay levels. When the scheduling mechanism type is time-slot queuing and forwarding TQF, the capability level information includes one or more TQF instances, and each TQF instance includes a corresponding arrangement cycle length, a number N of time slots contained in the arrangement cycle, and a number M of time slots contained in the scheduling cycle. When the scheduling mechanism type is a default type or fair queuing FQ, the capability level information is empty.

[0293] The second link attribute is added in the announced BGP-LS attribute to represent the maximum reservable bandwidth of each capability level of a specific scheduling mechanism type of the link, wherein the second link attribute includes at least one of the following fields: a scheduling mechanism type supported by the link and the maximum reservable bandwidth of each capability level of the specific scheduling mechanism type.

[0294] The third link attribute is added in the announced BGP-LS attribute to represent the idle bandwidth of each capability level of a specific scheduling mechanism type of the link, wherein the third link attribute includes at least one of the following fields: a scheduling mechanism type supported by the link and the idle bandwidth of each capability level of the specific scheduling mechanism type.

[0295] The fourth link attribute is added in the announced BGP-LS attribute to represent the maximum reservable burst size of each capability level of a specific scheduling mechanism type of the link, wherein the fourth link attribute includes at least one of the following fields: a scheduling mechanism type supported by the link and the maximum reservable burst size of each capability level of the specific scheduling mechanism type.

[0296] The fifth link attribute is added in the announced BGP-LS attribute to represent the idle burst size of each capability level of a specific scheduling mechanism type of the link, wherein the fifth link attribute includes at least one of the following fields: a scheduling mechanism type supported by the link and the idle burst size of each capability level of the specific scheduling mechanism type.

[0297] In the embodiment, the reservation module 1106 includes:

[0298] The first reservation submodule is configured to reserve, in the idle bandwidth resource and the idle burst resource of the specified capability level of the specified scheduling mechanism of each link contained in the deterministic forwarding path, bandwidth demand and burst demand of the service flow in a database maintained locally by the controller;

[0299] The second reservation submodule is configured to notify a head node of the deterministic forwarding path by the controller, to initiate distributed signaling by the head node, to reserve, in the idle bandwidth resource and the idle burst resource of the specified capability level of the specified scheduling mechanism of each link contained in the deterministic forwarding path, bandwidth demand and burst demand of the service flow, and to trigger the head node to notify the controller of the idle bandwidth resource and the idle burst resource of the specified capability level of the specified scheduling mechanism of each link after the reservation, so that the controller updates the database maintained locally by the controller.

[0300] It should be noted that the above modules can be implemented by software or hardware, and for the latter, the following implementation manners can be used, but are not limited thereto: the above modules are located in the same processor; or the above modules are located in different processors in any combination.

[0301] Embodiments of the present disclosure further provide a computer program product comprising computer instructions which, when executed by a processor, implement the steps in any of the above method embodiments.

[0302] Embodiments of the present disclosure further provide a computer-readable storage medium having a computer program stored therein, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

[0303] In an example embodiment, the above computer-readable storage medium can include, but is not limited to, a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0304] Embodiments of the present disclosure further provide an electronic device comprising a memory having a computer program stored therein and a processor configured to execute the computer program to perform the steps in any of the above method embodiments.

[0305] In an example embodiment, the above electronic device can further comprise a transmission device connected to the processor and an input / output device connected to the processor.

[0306] The specific examples in the present embodiment can refer to the examples described in the above embodiments and exemplary embodiments, which will not be repeated here.

[0307] Obviously, those skilled in the art should understand that the modules or steps of the present disclosure described above can be realized by general computing devices, which can be centralized on a single computing device or distributed on a deterministic network composed of multiple computing devices, which can be realized by program codes executable by the computing devices, so that they can be stored in storage devices and executed by the computing devices, and in some cases, the steps shown or described can be executed in different order, or they can be respectively manufactured into individual integrated circuit modules, or multiple modules or steps among them can be manufactured into a single integrated circuit module to realize. Thus, the present disclosure is not limited to any specific combination of hardware and software.

[0308] The above only describes the preferred embodiments of the present disclosure and is not intended to limit the present disclosure. Those skilled in the art can make various modifications and changes to the present disclosure. Any modification, equivalent replacement, improvement, etc. within the principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for reserving deterministic forwarding paths, comprising: a controller obtaining deterministic forwarding resource information of all links in a deterministic network; the controller computing a deterministic forwarding path for a traffic flow in the deterministic network based on the deterministic forwarding resource information, the deterministic forwarding path satisfying quality of service (QoS) requirements of the traffic flow; the controller reserving deterministic forwarding resources for the traffic flow on the deterministic forwarding path, the deterministic forwarding resources satisfying the QoS requirements of the traffic flow.

2. The method of claim 1, wherein, The free bandwidth resources of the deterministic forwarding path satisfy bandwidth requirements of the traffic flow, and the free burst resources of the deterministic forwarding path satisfy burst requirements of the traffic flow.

3. The method of claim 1, wherein, Before obtaining the deterministic forwarding resource information of all links in the deterministic network, further comprising: configuring one or more scheduling mechanisms on each link of each node in the deterministic network, wherein each scheduling mechanism supports one or more capability levels, and each capability level has corresponding deterministic forwarding resources.

4. The method of claim 3, wherein, The types of the scheduling mechanisms include at least one of the following: default type, asynchronous traffic shaping (ATS), credit-based shaping (CBS), ATS+CBS, cyclic queuing and forwarding (CQF), box cyclic queuing and forwarding (BCQF), earliest deadline first (EDF), time-slot queuing and forwarding (TQF), fair queuing (FQ), guaranteed low-latency-based forwarding (gLBF).

5. The method of claim 3, wherein, Further comprising: For each capability level of a specific scheduling mechanism on a link of each node, the node maintains corresponding free resources or used resources of each capability level according to the reservation of resources by the traffic flow, wherein the free resources include free bandwidth resources and free burst resources, and the free resources are equal to the maximum reservable resources in the initial case.

6. The method of claim 5, wherein, Further comprising: The node floods resource information of each capability level of each scheduling mechanism on each link of the node to the deterministic network through an interior gateway protocol (IGP), so that each node in the deterministic network obtains the resource information of each capability level of each scheduling mechanism on each link in the deterministic network.

7. The method of claim 6, wherein, In the initial case, only the maximum reservable bandwidth resources and the maximum reservable burst resources of each capability level of each scheduling mechanism on each link are flooded.

8. The method of claim 3, wherein, Further comprising: A border gateway protocol link state (BGP-LS) session is established between a selected first node in the deterministic network and the controller, so that the first node advertises the resource information of each capability level of each scheduling mechanism on each link in the deterministic network to the controller through the BGP-LS, and the controller updates a locally maintained database after receiving the resource information for subsequent path computation.

9. The method of claim 8, wherein, In the initial case, only the maximum reservable bandwidth resources and the maximum reservable burst resources are advertised.

10. The method of claim 8, wherein, The BGP-LS is extended by at least one of the following ways: The first link attribute is added in the announced BGP-LS attribute to represent the deterministic forwarding scheduling capability of the link, wherein the first link attribute comprises at least one of the following fields: a scheduling mechanism type supported by the link, a flag indicating whether a just-in-time scheduling mode is supported, a flag indicating whether a near-just-in-time scheduling mode is supported, capability level information corresponding to a specific scheduling mechanism type, and the like. When the scheduling mechanism type is asynchronous traffic shaping ATS, credit-based shaping CBS, asynchronous traffic shaping combined with credit-based shaping ATS+CBS, or guaranteed low- delay forwarding gLBF, the capability level information comprises one or more traffic classes. When the scheduling mechanism type is cyclic queuing and forwarding CQF or bin cyclic queuing and forwarding BCQF, the capability level information comprises one or more cycle lengths. When the scheduling mechanism type is earliest deadline first EDF, the capability level information comprises one or more delay levels. When the scheduling mechanism type is time-slot queuing and forwarding TQF, the capability level information comprises one or more TQF instances, and each TQF instance comprises a corresponding arrangement cycle length, a number N of time slots contained in the arrangement cycle, and a number M of time slots contained in the scheduling cycle. When the scheduling mechanism type is a default type or fair queuing FQ, the capability level information is empty. The second link attribute is added in the announced BGP-LS attribute to represent the maximum reservable bandwidth of each capability level of a specific scheduling mechanism type of the link, wherein the second link attribute comprises at least one of the following fields: a scheduling mechanism type supported by the link, and the maximum reservable bandwidth of each capability level of the specific scheduling mechanism type. The third link attribute is added in the announced BGP-LS attribute to represent the idle bandwidth of each capability level of a specific scheduling mechanism type of the link, wherein the third link attribute comprises at least one of the following fields: a scheduling mechanism type supported by the link, and the idle bandwidth of each capability level of the specific scheduling mechanism type. The fourth link attribute is added in the announced BGP-LS attribute to represent the maximum reservable burst of each capability level of a specific scheduling mechanism type of the link, wherein the fourth link attribute comprises at least one of the following fields: a scheduling mechanism type supported by the link, and the maximum reservable burst of each capability level of the specific scheduling mechanism type. The fifth link attribute is added in the announced BGP-LS attribute to represent the idle burst of each capability level of a specific scheduling mechanism type of the link, wherein the fifth link attribute comprises at least one of the following fields: a scheduling mechanism type supported by the link, and the idle burst of each capability level of the specific scheduling mechanism type.

11. The method of claim 3, wherein, The controller reserves deterministic forwarding resources satisfying the QoS requirement of the service flow on the deterministic forwarding path for the service flow, and the method comprises one of the following: The controller reserves bandwidth and burst requirements of the service flow in idle bandwidth resources and idle burst resources of a specified capability level of a specified scheduling mechanism of each link contained in the deterministic forwarding path in a database maintained locally by the controller. The controller notifies a head node of the deterministic forwarding path to initiate distributed signaling by the head node to reserve bandwidth and burst requirements of the service flow in free bandwidth resources and free burst resources of a specified capability level of a specified scheduling mechanism for each link contained in the deterministic forwarding path, and triggers the head node to notify the controller of free bandwidth resources and free burst resources of a specified capability level of a specified scheduling mechanism after reservation for each link, so that the controller updates a database maintained locally by the controller. 12.A method for reserving a deterministic forwarding path, comprising: a node in a deterministic network notifying a controller of deterministic forwarding resource information of each link, so that the controller calculates a deterministic forwarding path satisfying quality of service (QoS) requirements of a service flow in the deterministic network based on the deterministic forwarding resource information; the node reserving deterministic forwarding resources satisfying the QoS requirements of the service flow on the deterministic forwarding path according to the deterministic forwarding path notified by the controller; the node notifying the controller of remaining deterministic forwarding resources of each link after reservation of the deterministic forwarding path, so that the controller updates a database maintained locally by the controller.

13. The method of claim 12, wherein, one or more scheduling mechanisms are configured on each link of the node, wherein each scheduling mechanism supports one or more capability levels, and each capability level has corresponding deterministic forwarding resources. 14.A device for reserving a deterministic forwarding path, applied to a controller, comprising: an obtaining module configured to obtain deterministic forwarding resource information of all links in a deterministic network; a calculating module configured to calculate a deterministic forwarding path satisfying quality of service (QoS) requirements of a service flow in the deterministic network based on the deterministic forwarding resource information, and the deterministic forwarding path adopts a specified scheduling mechanism type; a reserving module configured to reserve deterministic forwarding resources satisfying the QoS requirements of the service flow on the deterministic forwarding path. 15.A computer program product, comprising a computer program which, when executed by a processor, implements the steps of the method in any one of claims 1 to 13.

16. A computer readable storage medium having stored therein a computer program, wherein, The computer program, when executed by a processor, implements the steps of the method in any one of claims 1 to 13. 17.An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method in any one of claims 1 to 13 when executing the computer program.

Images