Service transmission path determination method, system and apparatus, and electronic device

Abstract

Provided in the present invention are a service transmission path determination method, system and apparatus, and an electronic device. The method comprises: on the basis of a network topology structure, determining at least one end-to-end transmission path corresponding to a target service; sequentially traversing the at least one end-to-end transmission path, and on the basis of time information related to the currently traversed end-to-end transmission path and edge shaping parameters of the currently traversed end-to-end transmission path, determining an upper delay bound corresponding to the currently traversed end-to-end transmission path; and on the basis of the upper delay bound corresponding to the currently traversed end-to-end transmission path, determining a target end-to-end transmission path corresponding to the target service. By means of the method, a deterministic networking orchestration and scheduling process for a target service in a scenario where a plurality of local-area sub-networks / wide-area sub-networks are mutually connected is implemented, an upper delay bound of data transmission in a target end-to-end transmission path for the target service can be ensured, and the transmission delay requirements of the target service for data can be met.

Description

Method, system, device and electronic device for determining service transmission path

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on October 25, 2023, with application number 202311396503.0 and invention name “Method, system, device and electronic device for determining business transmission path”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present invention relates to the field of business transmission technology, and in particular to a method, system, device and electronic equipment for determining a business transmission path. Background Art

[0003] With the popularization and development of the Internet, target services in specific fields (such as time-sensitive services) have put forward higher requirements on the stability, reliability and timeliness of the network, such as telemedicine and high-precision industrial manufacturing. In order to meet these requirements, academia and industry have proposed deterministic networking (DetNet) technology to achieve deterministic bandwidth guarantee, bounded transmission delay guarantee and zero packet loss during transmission for time-sensitive services. At present, this deterministic network technology is mainly limited to deterministic service quality guarantee within a single network domain, such as a local area network or a wide area network. For local area network scenarios, the Institute of Electrical and Electronics Engineers (IEEE) Time-Sensitive Networking (TSN) Working Group (WG) provides a series of mechanisms, including clock synchronization, traffic scheduling and link reservation, to ensure delay-constrained Layer 2 transmission. For wide area network scenarios, the Internet Engineering Task Force (IETF) DetNet WG proposed the Scalable Deterministic Forwarding (SDF) mechanism to support quality of service (QoS) assurance for long-distance links.

[0004] However, in real-world scenarios, time-sensitive services often require cross-domain, long-distance, end-to-end service quality assurance. Currently, there has been some research on end-to-end deterministic cross-domain transmission, primarily considering a relatively simple end-to-end network scenario: a deterministic local subnet at the source, a long-distance link with guaranteed quality (a wide-area deterministic subnet), and a peer deterministic subnet. Time-sensitive services access a deterministic local subnet, flow through a wide-area link with guaranteed quality of service (possibly a deterministic wide-area subnet), reach the deterministic local subnet where the peer service resides, and ultimately reach the peer service end.

[0005] These scenarios typically involve larger networks, encompassing multiple interconnected local and / or wide area subnets, and involving more time-sensitive services, forwarding nodes, and network links. When allocating end-to-end transmission paths for time-sensitive services, the complexities of scheduling and orchestrating these services often result in the finalized target end-to-end transmission paths failing to meet the data transmission latency requirements of these services.

[0006] Summary of the Invention

[0007] The present invention provides a method, system, device and electronic device for determining a service transmission path, which realizes a deterministic network orchestration and scheduling process for a target service in a scenario of multiple interconnected local area subnets / wide area subnets, can ensure the upper bound of the delay when the target service transmits data in the target end-to-end transmission path, and can meet the target service's data transmission delay requirements.

[0008] The present invention provides a method for determining a service transmission path, the method comprising:

[0009] Determine at least one end-to-end transmission path corresponding to the target service based on the network topology;

[0010] traversing the at least one end-to-end transmission path in sequence, and determining an upper bound on the delay corresponding to the currently traversed end-to-end transmission path based on time information associated with the currently traversed end-to-end transmission path and an edge shaping parameter of the currently traversed end-to-end transmission path, wherein the edge shaping parameter is used to represent bandwidth resources reserved per unit time by an edge node corresponding to a wide area subnet for the target service;

[0011] If the delay upper bound corresponding to the currently traversed end-to-end transmission path is greater than a preset delay upper bound threshold, determining the next end-to-end transmission path after the currently traversed end-to-end transmission path as the new currently traversed end-to-end transmission path until the delay upper bound corresponding to the new currently traversed end-to-end transmission path is less than or equal to the preset delay upper bound threshold;

[0012] When the delay upper bound corresponding to the currently traversed end-to-end transmission path is less than or equal to a preset delay upper bound threshold, the currently traversed end-to-end transmission path is determined as the target end-to-end transmission path.

[0013] According to a method for determining a service transmission path provided by the present invention, the edge shaping parameters are used by at least one local area subnet controller involved in the currently traversed end-to-end transmission path to respectively determine a first service curve for the corresponding local subnet; the edge shaping parameters are also used by at least one wide area subnet controller involved in the currently traversed end-to-end transmission path to respectively determine a second service curve for the corresponding wide area subnet; and determining the upper delay bound corresponding to the currently traversed end-to-end transmission path based on time information involved in the currently traversed end-to-end transmission path and the edge shaping parameters of the end-to-end transmission path includes: determining the upper delay bound corresponding to the currently traversed end-to-end transmission path based on all first service curves and all second service curves.

[0014] According to a method for determining a service transmission path provided by the present invention, for any first service curve, the time information includes the closing time of the gated queue corresponding to the target service, and the first service curve is obtained based on the following steps: obtaining the closing time of the gated queue corresponding to the target service; determining the first link bandwidth corresponding to the local subnet controller based on the edge shaping parameter; and determining the first service curve based on the closing time and the first link bandwidth.

[0015] According to a method for determining a service transmission path provided by the present invention, for any second service curve, the time information includes the end time of the time slot occupied by the egress port of the starting node of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the egress port of the ending node of the transmission link of the target service in the wide area subnet. The second service curve is obtained based on the following steps: determining the maximum intra-domain delay of the wide area subnet corresponding to the wide area subnet controller based on the end time of the time slot occupied by the egress port of the starting node of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the egress port of the ending node of the transmission link of the target service in the wide area subnet; determining the second link bandwidth corresponding to the wide area subnet controller based on the edge shaping parameter; and determining the second service curve based on the maximum intra-domain delay and the second link bandwidth.

[0016] According to a method for determining a service transmission path provided by the present invention, determining the first service curve according to the shutdown time and the first link bandwidth includes: determining the first service curve according to a first function formula; wherein the first function formula is: β e(t) represents the first service curve of the local area subnet controller corresponding to the mth local area subnet at time t; W e represents the first link bandwidth of the transmission link e in the m-th local area subnet; Indicates the closing time; [·] + Indicates a positive number; U m represents the local area subnet set corresponding to the m-th local area subnet; Represents the transmission link of the mth local area subnet.

[0017] According to a method for determining a service transmission path provided by the present invention, determining the second service curve based on the maximum delay within the domain and the second link bandwidth includes: determining the second service curve based on a second function formula; wherein the second function formula is: represents a second service curve of the wide area subnet controller corresponding to the nth wide area subnet for target service f at time t; represents the second link bandwidth of the target service f in the nth wide area subnet; τ n,f It represents the maximum delay of the target service f in the nth wide area subnet.

[0018] According to a method for determining a service transmission path provided by the present invention, determining the upper bound of the delay corresponding to the currently traversed end-to-end transmission path based on all first service curves and all second service curves includes: determining the total service curve corresponding to the currently traversed end-to-end transmission path based on a service curve function; determining the upper bound of the delay corresponding to the currently traversed end-to-end transmission path based on the total service curve; wherein the service curve function is β * e W (t) represents the total service curve corresponding to the end-to-end transmission path currently traversed at time t; e represents the first link bandwidth of transmission link e in the mth local area subnet; Indicates the closing time; [·] + Indicates a positive number; represents the local area subnet set corresponding to the m-th local area subnet; M represents the total number of local area subnets; represents the second link bandwidth of the target service f in the nth wide area subnet; τ n,f represents the maximum intra-domain delay of the target service f in the nth wide area subnet; v uo Represents a node set, including m local subnet nodes and n virtual nodes Each of the virtual nodes n wide area subnets Jointly labeled with the target business f; Represents the node set v up In addition to the m local subnet nodes All nodes except .

[0019] The present invention also provides a system for determining a service transmission path, comprising: a global resource orchestrator, a local area subnet controller, and a wide area subnet controller;

[0020] The global resource orchestrator is configured to determine, based on the network topology, at least one end-to-end transmission path corresponding to a target service; sequentially traverse the at least one end-to-end transmission path, and for a currently traversed end-to-end transmission path, send time information and edge shaping parameters related to the currently traversed end-to-end transmission path to at least one local area subnet controller and at least one wide area subnet controller involved in the currently traversed end-to-end transmission path, wherein the edge shaping parameters are used to represent bandwidth resources reserved per unit time by an edge node corresponding to the wide area subnet for the target service;

[0021] For each local area subnet controller, the local area subnet controller is configured to determine a first service curve of the local area subnet corresponding to the local area subnet controller based on the time information and the edge shaping parameter;

[0022] For each wide area subnetwork controller, the wide area subnetwork controller is configured to determine a second service curve corresponding to the wide area subnetwork of the wide area subnetwork controller based on the time information and the edge shaping parameter;

[0023] The global resource orchestrator is further configured to determine an upper delay bound corresponding to the currently traversed end-to-end transmission path based on all first service curves and all second service curves; if the upper delay bound corresponding to the currently traversed end-to-end transmission path is greater than a preset upper delay bound threshold, determine the next end-to-end transmission path after the currently traversed end-to-end transmission path as the new currently traversed end-to-end transmission path, until the upper delay bound corresponding to the new currently traversed end-to-end transmission path is less than or equal to the preset upper delay bound threshold; if the upper delay bound corresponding to the currently traversed end-to-end transmission path is less than or equal to the preset upper delay bound threshold, determine the currently traversed end-to-end transmission path as the target end-to-end transmission path.

[0024] The present invention also provides a device for determining a service transmission path, the device comprising:

[0025] a transmission path and bandwidth determination module configured to determine at least one end-to-end transmission path corresponding to the target service based on the network topology;

[0026] The target transmission path determination module is configured to traverse the at least one end-to-end transmission path in sequence, and determine the delay upper bound corresponding to the currently traversed end-to-end transmission path based on the time information involved in the currently traversed end-to-end transmission path and the edge shaping parameter of the currently traversed end-to-end transmission path; when the delay upper bound corresponding to the currently traversed end-to-end transmission path is greater than a preset delay upper bound threshold, the next end-to-end transmission path of the currently traversed end-to-end transmission path is determined as the new currently traversed end-to-end transmission path, until the delay upper bound corresponding to the new currently traversed end-to-end transmission path is less than or equal to the preset delay upper bound threshold; when the delay upper bound corresponding to the currently traversed end-to-end transmission path is less than or equal to the preset delay upper bound threshold, the currently traversed end-to-end transmission path is determined as the target end-to-end transmission path.

[0027] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for determining a service transmission path as described above is implemented.

[0028] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for determining a service transmission path.

[0029] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for determining a service transmission path.

[0030] The method, system, device and electronic device for determining a service transmission path provided by the present invention determine at least one end-to-end transmission path corresponding to a target service based on a network topology; sequentially traverse the at least one end-to-end transmission path, and determine the upper bound of the delay corresponding to the currently traversed end-to-end transmission path based on time information involved in the currently traversed end-to-end transmission path and an edge shaping parameter of the currently traversed end-to-end transmission path, wherein the edge shaping parameter is used to characterize the bandwidth resources reserved per unit time by the edge node corresponding to the wide area subnet for the target service; and when the time When the upper bound of the delay corresponding to the previously traversed end-to-end transmission path is greater than the preset upper bound of the delay, the next end-to-end transmission path of the currently traversed end-to-end transmission path is determined as the new end-to-end transmission path currently traversed, until the upper bound of the delay corresponding to the new end-to-end transmission path currently traversed is less than or equal to the preset upper bound of the delay; when the upper bound of the delay corresponding to the currently traversed end-to-end transmission path is less than or equal to the preset upper bound of the delay, the currently traversed end-to-end transmission path is determined as the target end-to-end transmission path. This method implements a deterministic network orchestration and scheduling process for target services in multiple interconnected local area subnets / wide area subnets scenarios, can ensure the upper bound of the delay when the target service transmits data in the target end-to-end transmission path, and can meet the target service's data transmission delay requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0032] FIG1 is a schematic diagram of an end-to-end deterministic network orchestration and scheduling framework provided by the present invention;

[0033] FIG2 is a schematic diagram of the structure of an end-to-end deterministic multi-domain orchestration network provided by the present invention;

[0034] 3 is a schematic flow chart of a method for determining a service transmission path provided by the present invention;

[0035] FIG4 is a schematic diagram of the structure of an end-to-end deterministic multi-domain orchestration network provided by the present invention;

[0036] 5 is a schematic structural diagram of a system for determining a service transmission path provided by the present invention;

[0037] 6 is a schematic structural diagram of a device for determining a service transmission path provided by the present invention;

[0038] FIG7 is a schematic structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0039] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0040] First, to fully understand the embodiments of the present invention, the background technology may be described in detail:

[0041] Currently, there are two methods for determining end-to-end transmission paths: a method of establishing a hypercycle-based time slot mapping method between adjacent subnets, and a method of using network calculation.

[0042] The former uses per-flow edge shaping and time slot offsetting to establish time-slot-based hard slicing for cross-domain end-to-end transmission, achieving high-precision end-to-end delay and jitter control. This approach eliminates forwarding uncertainty introduced by heterogeneous cross-domain devices and mechanisms through time mapping and shaping, ensuring the transmission quality of time-sensitive services. The latter calculates the upper bound of the delay for time-sensitive services at each forwarding node along the way on a hop-by-hop basis, and then analyzes and derives the upper bound of end-to-end transmission delay. This approach allows for pre-assessment of end-to-end transmission delay and network optimization to meet the transmission requirements of time-sensitive services.

[0043] In the prior art, there may be multiple deterministic network connection modes between the source host (i.e., source node) and the target host (i.e., target node). The source node can select a transmission path with different subnet combinations to connect to the target node to improve the throughput and network efficiency of end-to-end time-sensitive services.

[0044] However, such scenarios generally involve larger-scale networks, including multiple interconnected local area subnets and / or wide area subnets, involving more time-sensitive services, forwarding nodes and network links, making the scheduling and orchestration of time-sensitive services more complicated.

[0045] In this case, using a hypercycle-based time slot mapping method will result in excessively long hypercycles and an overly large end-to-end orchestration problem, making it impractical to directly apply. Using a network calculus-based method, the massive number of nodes and service flows will dramatically increase the complexity of end-to-end delay analysis. Furthermore, static network calculus resource scheduling methods are unable to adapt to the dynamic nature of backbone subnets. In summary, when traditional end-to-end deterministic service scheduling mechanisms allocate end-to-end transmission paths for time-sensitive services, the resulting target end-to-end transmission paths often fail to meet the data transmission delay requirements of time-sensitive services due to the complex scheduling and orchestration issues associated with these services.

[0046] To address this issue, we propose an end-to-end deterministic network orchestration and scheduling framework. This framework implements deterministic network orchestration and scheduling for target services (such as time-sensitive services) across multiple interconnected local and wide area subnets. This framework ensures an upper bound on the latency of target services during data transmission along the target end-to-end transmission path, meeting their data transmission latency requirements and effectively implementing end-to-end delay-bounded scheduling for time-sensitive services. Target services are services with strict upper bounds on transmission latency.

[0047] The above end-to-end deterministic network orchestration and scheduling framework is applicable to a deterministic network, which may include multiple deterministic local area subnets and deterministic wide area subnets. The local area subnets may enable the IEEE 802.1Qbv mechanism, while the wide area subnet may enable the DetNet SDF mechanism.

[0048] As shown in Figure 1, it is a schematic diagram of the end-to-end deterministic network orchestration and scheduling framework provided by the present invention. As can be seen from Figure 1, the end-to-end deterministic network orchestration and scheduling framework can also be called the system architecture corresponding to the determination system of the service transmission path, which can include a global resource orchestration layer and a physical resource abstraction layer. Among them, the global resource orchestration layer is provided with a global resource orchestrator (abbreviated as: orchestrator), and the physical resource abstraction layer is provided with at least one local subnet controller and at least one wide area subnet controller. Each local subnet controller corresponds to at least one local resource, and each wide area subnet controller corresponds to at least one wide area resource, and exists in the host that interacts with the host.

[0049] Users can apply for service access requests to the orchestrator to enter the system that determines the service transmission path, and then obtain end-to-end deterministic service quality assurance between the source node and the target node.

[0050] The physical resource abstraction layer is primarily responsible for collecting network and / or service information and executing the transmission path configuration policies of the global resource orchestration layer. Furthermore, each local area subnet controller and each wide area subnet controller in the physical resource abstraction layer can use service curves to provide a unified resource description to the global resource orchestration layer, supporting end-to-end resource orchestration and service scheduling optimization for heterogeneous cross-domain interconnection.

[0051] The global resource orchestration layer, based on the service curve (i.e., unified resource abstraction) provided by the physical resource abstraction layer, determines the upper bound of the delay of network information and / or business information, and searches for a better resource allocation method (i.e., transmission path configuration strategy) when the upper bound of the delay satisfies the end-to-end delay constraint of time-sensitive services. During the search process, to reduce the complexity of data analysis, the wide area subnet can be abstracted as a virtual switching node. In addition, in order to more accurately analyze the maximum end-to-end delay information of the service, the one-time burst criterion analysis method is often used in the network calculation process to ensure that the selected transmission path configuration strategy meets the delay requirements of the network information and / or business information. Among them, converting the wide area subnet into a virtual node can effectively achieve low-complexity end-to-end orchestration and scheduling, greatly reducing the number of nodes involved in the end-to-end orchestration and scheduling process, reducing the problem scale of end-to-end delay analysis in network calculation, and improving the scalability of the orchestration and scheduling mechanism.

[0052] As shown in Figure 2, it is a structural diagram of the end-to-end deterministic multi-domain orchestration network provided by the present invention. As can be seen from Figure 2, the end-to-end deterministic multi-domain orchestration network consists of a plurality of interconnected local subnets and wide area subnets. Each subnet is equipped with a corresponding subnet controller, such as a local subnet controller for the local subnet and a wide area subnet controller for the wide area subnet. In the entire end-to-end deterministic multi-domain orchestration network, the orchestrator plays a key role and is responsible for end-to-end resource management and control. Through this end-to-end deterministic multi-domain orchestration network, the orchestration and scheduling of target services across different domains can be achieved, ensuring the deterministic maximum delay guarantee of network transmission.

[0053] 1 and 2 , the network topology information corresponding to all local subnets and all wide area subnets in the data plane can be represented by G = {v, ε}. The data plane includes M local subnets and N wide area subnets, where M ≥ 1 and N ≥ 1.

[0054] Among them, the mth local area subnet is available express, represents the switches and routers of the mth local subnet, Represents the transmission link of the mth local area subnet.

[0055] The nth wide area subnet is available express, represents the switches and routers of the nth wide area subnet, Indicates the transmission link of the nth wide area subnet.

[0056] The target service (such as time-sensitive service) f∈F can be represented by the five-tuple in the above end-to-end deterministic multi-domain orchestration network. Indicates that r f,0 represents the average arrival rate of the target service, b f,0 Indicates the maximum burst capacity of the target service, Indicates the source node of the target business, Indicates the destination node of the target business, Γ f Indicates the maximum end-to-end transmission delay information of the target service. and destination node The location of the time-sensitive service can be divided into long-distance time-sensitive service and local time-sensitive service: the source node of the long-distance time-sensitive service and destination node Usually located in two different local area subnets, the source node of the local time-sensitive service and destination node It should be noted that compared with the hypercycle-based solution, the use of arrival curves (i.e., quintuples) to describe service flow characteristics expands the applicable scenarios of current deterministic cross-domain orchestration and scheduling, and is no longer limited to periodic service flows.

[0057] The mth local area subnet In the above end-to-end deterministic multi-domain orchestration network, delay-bounded transmission within a preset number of hops (e.g., 7 hops) can be achieved. For local time-sensitive services, the mth local subnet It can provide end-to-end delay-bounded transmission; for long-distance time-sensitive services, the mth local subnet It provides a deterministic access service with bounded delay. In addition, the mth local area subnet Any output port h of each device in v,e Equipped with the Time Aware Shaper (TAS) shaping mechanism that supports the IEEE 802.1Qbv definition, where h v,e Indicates the output port of the switch or router v connected to the link e. The TAS shaping mechanism only considers two different service priorities, corresponding to time-sensitive services and best effort (BE) services. Among them, the scheduling period of the gated queue corresponding to the TAS shaping mechanism is a specific time length, which can be used Indicates that the opening time of the gated queue corresponding to the time-sensitive service is available Indicates that the closing time of the gated queue corresponding to the time-sensitive service is available express.

[0058] nth wide area subnet Provides end-to-end deterministic quality of service for long-distance transmission of time-sensitive services. Time-sensitive services enter the wide area subnet from one endpoint (called the ingress node) and leave from another endpoint (called the egress node). At the ingress node, edge shaping based on bandwidth guarantee is deployed to assign the first link bandwidth to each arriving time-sensitive service. When allocating link bandwidth for target services, the nth wide area subnet can only select some discrete optional link bandwidths, such as 1Mbps, 5Mbps, 10Mbps, 50Mbps, etc., and the available set B det Indicates the nth wide area subnet The output port h of each device v,e All are equipped with the scalable deterministic forwarding SDF shaping function that supports the IETF DetNet SDF definition. The length of the SDF time slot is a fixed value and can be used express.

[0059] The embodiment of the present invention is described below with reference to FIG1 and FIG2 .

[0060] FIG3 is a flow chart of a method for determining a service transmission path provided by the present invention, which may include:

[0061] 301. Determine at least one end-to-end transmission path corresponding to a target service according to a network topology.

[0062] The network topology structure may include: a local area subnet topology structure and a wide area subnet topology structure.

[0063] The M local area subnets included in the data plane can constitute a local area subnet topology structure, and the N wide area subnets can constitute a wide area subnet topology structure.

[0064] Optionally, the target service F may include a time-sensitive service f, and the number of the time-sensitive services is not limited.

[0065] The target service can be forwarded by the source node via the local subnet controller; then, after receiving the target service, the end-to-end transmission path corresponding to the target service can be determined based on the network topology in the data plane. Since the network topology is relatively complex, the number of end-to-end transmission paths finally determined is at least one, providing data support for the subsequent determination of the target end-to-end transmission path.

[0066] 302. Traverse at least one end-to-end transmission path in sequence, and determine a delay upper bound corresponding to the currently traversed end-to-end transmission path based on time information involved in the currently traversed end-to-end transmission path and an edge shaping parameter of the currently traversed end-to-end transmission path.

[0067] Among them, since the mechanism for enabling the wide area subnet is different from the mechanism for enabling the local area subnet, the time information involved in the wide area subnet corresponding mechanism is also different from the time information involved in the local area subnet corresponding mechanism. Specifically, the time information involved in the wide area subnet corresponding mechanism may include the closing time of the gated queue corresponding to the target service; the time information involved in the local area subnet corresponding mechanism may include the end time of the time slot occupied by the starting node egress port of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the ending node egress port of the transmission link of the target service in the wide area subnet.

[0068] The physical meaning of the edge shaping parameter is the bandwidth resources reserved by the edge node corresponding to the wide area subnet for different services (such as target services) within a unit time.

[0069] The upper delay bound refers to the maximum delay assigned to the end-to-end transmission path.

[0070] After determining at least one end-to-end transmission path corresponding to the target service, while sequentially traversing the at least one end-to-end transmission path, the time information and edge shaping parameters involved in the currently traversed end-to-end transmission path can be first obtained, and then these two data can be calculated to obtain the upper bound of the delay corresponding to the currently traversed end-to-end transmission path, thereby providing data support for the subsequent determination of the target end-to-end transmission path.

[0071] It should be noted that traditional bandwidth allocation refers to the bandwidth upper bound allocated by a network service provider for a service at the access point. During traditional service scheduling, the network service provider cannot determine the precise traffic flow model for the service. Therefore, if the service transmission rate exceeds the given bandwidth transmission upper bound, the network service provider will implement flow control within the allocated bandwidth upper bound. In contrast, in the embodiments of the present invention, within the deterministic network domain, the precise traffic flow model can be determined during service scheduling. Therefore, the bandwidth allocation function of traditional network services is no longer required. The edge shaping parameters involved in the embodiments of the present invention physically represent the bandwidth resources reserved per unit time for different services at the edge node of the wide area subnet. These edge shaping parameters are deployed to prevent uncontrollable microbursts in downstream links for multiple time-sensitive service flows, which in turn lead to uncontrollable latency and jitter. Specifically, when a target service enters a different wide area subnet, the shaping function deployed at the edge of the wide area subnet performs shaping operations based on the shaping parameters (bandwidth). When the data arrival rate of the target service exceeds the corresponding bandwidth, the edge shaping function caches the corresponding data to "smooth out" the traffic. In summary, the two have different purposes and different approaches.

[0072] Example 1, shown in Figure 4, is a schematic diagram of the end-to-end deterministic multi-domain orchestration network provided by the present invention. As shown in Figure 4, the source node is host A, and the destination nodes are hosts B and C. Host A has two time-sensitive services f: f1 and f2. Host A can send f1 to host B and f2 to host C.

[0073] In the process of sending these two time-sensitive services, host A can first initiate a service access request and report the relevant data of the two time-sensitive services; then, after receiving the service access request initiated by host A, the eligibility of the time-sensitive service f1 and the time-sensitive service f2 can be checked based on the service access request, that is, whether the time-sensitive service f1 and the time-sensitive service f2 have been registered in advance: if it is determined that they have not been registered, no operation is performed on the time-sensitive service f1 and the time-sensitive service f2; if it is determined that they have been registered, the first end-to-end transmission path corresponding to the time-sensitive service f1 is enumerated and the edge shaping parameters of the wide area subnet corresponding to the first end-to-end transmission path are determined; the second end-to-end transmission path corresponding to the time-sensitive service f2 is enumerated and the edge shaping parameters of the wide area subnet corresponding to the second end-to-end transmission path are determined.

[0074] The end-to-end transmission path and the edge shaping parameters of the wide area subnet corresponding to the end-to-end transmission path can be regarded as a pair of resource allocation combinations. express.

[0075] First end-to-end transmission path available Indicates that it can include: {1,2,5,7,9,10} or {1,3,4,6,8,10}. In this case, the edge shaping parameters can be used Indicates, n1 indicates wide area subnet A; In this case, the edge shaping parameters can be used n2 represents the wide area subnet B. The two pairs of resource allocation combinations corresponding to the time-sensitive service f1 can be determined, which are and

[0076] Second end-to-end transmission path available Indicates that it can include: {1,2,5,7,9,11} or {1,3,4,6,8,11}. In this case, the edge shaping parameters can be used Indicates; In this case, the edge shaping parameters can be used Represented. The two pairs of resource allocation combinations corresponding to the time-sensitive service f2 can be determined, which are and

[0077] Exemplarily, the process of enumerating the end-to-end transmission paths corresponding to the target service is as follows:

[0078] 1 Input: target service F, network topology information G corresponding to the network topology structure, and the set of feasible link bandwidths B of the wide area subnet det ;

[0079] 2 Output: feasible resource allocation;

[0080] 3 For {any time-sensitive service f∈F}do

[0081] 4 Use the breadth-first traversal algorithm to obtain all feasible transmission paths P in the network topology information G f ;

[0082] 5 End

[0083] 6 For {any feasible link bandwidth }do

[0084] 7 Take any set of feasible transmission paths and get

[0085] 8 End

[0086] Based on the above enumeration process, all end-to-end transmission paths corresponding to the target service can be accurately determined.

[0087] Optionally, when the local subnet controller reports the target service, it also reports the local subnet information, which may include: local subnet topology The TAS shaping mechanism corresponds to a gated queue with a scheduling period of a specific length of time. The opening time of the gated queue corresponding to the target business The link bandwidth W of each transmission link e e and the overhead cost of the transmission link e.

[0088] Optionally, the wide area subnet controller may also report wide area subnet information, which may include: wide area subnet topology Wide area subnet feasible link bandwidth set B det , the link bandwidth W of each transmission link e e and the overhead cost of the transmission link e.

[0089] Optionally, after step 302, the method may further include: sending configuration execution information to at least one target local area subnet controller and at least one target wide area subnet controller involved in the target end-to-end transmission path, wherein the configuration execution information is used to inform the source node corresponding to the target service to transmit the target service in accordance with at least one end-to-end transmission path carried by the configuration execution information and the respective edge shaping parameters of the at least one end-to-end transmission path.

[0090] In some embodiments, the edge shaping parameters are used for each of the at least one local area subnet controllers involved in the currently traversed end-to-end transmission path to determine the first service curve of the corresponding local subnet; the edge shaping parameters are also used for each of the at least one wide area subnet controllers involved in the currently traversed end-to-end transmission path to determine the second service curve of the corresponding wide area subnet; based on the time information involved in the currently traversed end-to-end transmission path and the edge shaping parameters of the currently traversed end-to-end transmission path, determining the upper bound of the delay corresponding to the currently traversed end-to-end transmission path may include: determining the upper bound of the delay corresponding to the currently traversed end-to-end transmission path based on all the first service curves and all the second service curves.

[0091] Wherein, each local area subnet controller determines that the first service curve of the corresponding local area subnet is available β e (t) represents the second service curve corresponding to the n-th wide area subnet. n,f (t) indicates.

[0092] The first service curve is used to describe the service capability provided by the switch node to the target service convergence flow; the second service curve is used to describe the service capability provided by the wide area subnet to the target service.

[0093] For the currently traversed end-to-end transmission path, the electronic device, in the process of determining the delay upper bound corresponding to the currently traversed end-to-end transmission path, may first send the edge shaping parameters corresponding to the currently traversed end-to-end transmission path to at least one local subnet controller and at least one wide area subnet controller involved in the currently traversed end transmission path.

[0094] For each local area subnet controller, a first service curve corresponding to the local area subnet can be determined and reported; for each wide area subnet controller, a second service curve corresponding to the wide area subnet can be determined and reported.

[0095] In this way, after receiving the first service curve sent by each local area subnet controller and the second service curve sent by each wide area subnet controller, the electronic device can first convert the wide area subnet into a virtual node. At this time, the second service curve of the time-sensitive service on the virtual node can be used e,f (t) represents; finally, based on all the first service curves and all the second service curves obtained, the delay upper bound corresponding to the currently traversed end-to-end transmission path is determined.

[0096] In some embodiments, for any first service curve, the time information includes the closing time of the gated queue corresponding to the target service, and the first service curve is obtained based on the following steps: the local subnet controller obtains the closing time of the gated queue corresponding to the target service; the local subnet controller determines the first link bandwidth corresponding to the local subnet controller based on the edge shaping parameter; the local subnet controller determines the first service curve based on the closing time and the first link bandwidth.

[0097] In the process of determining the corresponding first service curve, each local subnet controller can first obtain the closing time of the gated queue corresponding to the target service, and determine the first link bandwidth corresponding to each local subnet controller based on the edge shaping parameters issued, and then determine the first service curve corresponding to all switches in the local subnet corresponding to each local subnet controller based on each closing time and each first link bandwidth.

[0098] In some embodiments, the local area subnet controller determines the first service curve according to the shutdown time and the first link bandwidth, which may include: the local area subnet controller determines the first service curve according to a first function formula.

[0099] Among them, the first function formula is:

[0100] β e (t) represents the first service curve of the local area subnet controller corresponding to the mth local area subnet at time t; W e represents the first link bandwidth of transmission link e in the mth local area subnet; Indicates closing time; [·] + Indicates a positive number; U m represents the local subnet set corresponding to the mth local subnet; Represents the transmission link of the mth local area subnet.

[0101] The local subnet controller can accurately determine the first service curves corresponding to all switches in the local subnet according to the first function formula.

[0102] In some embodiments, for any second service curve, the time information includes the end time of the time slot occupied by the starting node egress port of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the ending node egress port of the transmission link of the target service in the wide area subnet. The second service curve is obtained based on the following steps: the wide area subnet controller determines the maximum intra-domain delay of the wide area subnet corresponding to the wide area subnet controller based on the end time of the time slot occupied by the starting node egress port of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the ending node egress port of the transmission link of the target service in the wide area subnet; the wide area subnet controller determines the second link bandwidth corresponding to the wide area subnet controller based on the edge shaping parameter; the wide area subnet controller determines the second service curve based on the maximum intra-domain delay and the second link bandwidth.

[0103] In the process of determining the corresponding second service curve, each wide area subnet controller can first determine the maximum delay within the corresponding wide area subnet, and determine the second link bandwidth corresponding to each wide area subnet controller based on the edge shaping parameters issued, and then determine the second service curve corresponding to each wide area subnet controller based on the maximum delay within each domain and each second link bandwidth.

[0104] In some embodiments, determining the second service curve according to the maximum delay and the second link bandwidth within the wide area subnet controller domain may include: the wide area subnet controller determining the second service curve according to a second function formula.

[0105] Among them, the second function formula is:

[0106] represents the second service curve of the wide area subnet controller corresponding to the nth wide area subnet at time t for the target service f; represents the second link bandwidth of the target service f in the nth wide area subnet; τ n,f It represents the maximum intra-domain delay of target service f in the nth wide area subnet.

[0107] The wide area subnet controller can accurately determine the second service curve corresponding to the wide area subnet according to the first function formula.

[0108] Optionally, the wide area subnet controller may use a rate-delay service curve model β(t) = R(t-τ n,f ). Where R represents the average service rate of the nth wide area subnet, τ n,f The average service rate R and the maximum intra-domain delay τ are n,f The value of is related to the configuration of the target service f on the nth wide area subnet.

[0109] Given a resource allocation combination In the case of The maximum delay in the domain τ n,f It is related to the following three factors: 1. The time slot length of the nth wide area subnet 2. Transmission path of target service f within the nth wide area subnet I represents the intersection; 3. The topology and clock synchronization properties of the nth wide area subnet.

[0110] Given a resource allocation combination In the case of in, represents the number of hops, Δ h Indicates the hth hop time slot and The time interval between Indicates the end time of the time slot occupied by the starting node egress port of the transmission link e of the target service f in the nth wide sub-area network, Indicates the end time of the time slot occupied by the target service f at the egress port of the end node of the transmission link e in the nth wide area subnet.

[0111] In this way, end-to-end scheduling and optimization of target services can be effectively achieved.

[0112] In some embodiments, determining the upper bound of the delay corresponding to the currently traversed end-to-end transmission path based on all first service curves and all second service curves may include: determining the total service curve corresponding to the currently traversed end-to-end transmission path based on the service curve function; and determining the upper bound of the delay corresponding to the currently traversed end-to-end transmission path based on the total service curve.

[0113] Among them, the service curve function is:

[0114] β * e (t) represents the total service curve corresponding to the end-to-end transmission path currently traversed at time t; W e represents the first link bandwidth of transmission link e in the mth local area subnet; Indicates closing time; [·]+ Indicates a positive number; represents the local subnet set corresponding to the mth local subnet; M represents the total number of local subnets; represents the second link bandwidth of the target service f in the nth wide area subnet; τ n,f represents the maximum intra-domain delay of target service f in the nth wide area subnet; v up Represents a node set, including m local subnet nodes and n virtual nodes Each virtual node n wide area subnets Jointly labeled with the target business f; Represents the node set v up Except for m local subnet nodes All nodes except .

[0115] Based on the above process, network calculus theory can be used to eliminate the differences between the wide area subnet and the local area subnet in terms of service forwarding mechanism and resource reuse mode, which can obtain a better end-to-end and unified logical abstraction.

[0116] The network calculation process is as follows: Given a resource allocation combination In the case of up ={v up ,ε up}. Among them, v up Represents a node set, which can include m local subnet nodes and n virtual nodes Each virtual node n wide area subnets Labeled together with the target business f.

[0117] Node v∈v up The output port is exist In case of The first service curve is composed of the above first curve function β e (t)=We e [t-Tm,ctsn+ available; in In the case of hv,eup, the second service curve of the output port is given by the second curve function Available.

[0118] In summary, in the target network G up ={v up ,ε up}, based on the output port The total service curve of is used, and the Separate Flow Analysis (SFA) algorithm is used to obtain the upper bound of the delay corresponding to the currently traversed end-to-end transmission path.

[0119] Based on the above process, converting the wide area subnet into a virtual node can effectively achieve low-complexity end-to-end orchestration and scheduling.

[0120] The wide-area subnet utilizes an incremental / online deployment approach, adapting to its massive connections and dynamic environment. Because the wide-area subnet employs an orthogonal resource allocation approach, time-sensitive services are isolated from each other. Therefore, network calculations enable individual analysis of target services and support incremental / online deployment.

[0121] The SFA algorithm refers to the use of the above-mentioned network calculation principles to calculate the upper bound of the delay corresponding to the end-to-end transmission path.

[0122] Using the SFA algorithm, the upper bound of the delay corresponding to the end-to-end transmission path can be obtained with high accuracy.

[0123] 303. When the delay upper bound corresponding to the currently traversed end-to-end transmission path is greater than a preset delay upper bound threshold, the next end-to-end transmission path after the currently traversed end-to-end transmission path is determined as the new currently traversed end-to-end transmission path, until the delay upper bound corresponding to the new currently traversed end-to-end transmission path is less than or equal to the preset delay upper bound threshold.

[0124] The preset upper delay threshold can be understood as the maximum end-to-end transmission delay requirement for the target service (e.g., a time-sensitive service), i.e., the maximum transmission delay required for data transmission by the target service. The preset upper delay thresholds for different target services can be the same or different, and are not specifically limited here.

[0125] 304. If the delay upper bound corresponding to the currently traversed end-to-end transmission path is less than or equal to a preset delay upper bound threshold, determine the currently traversed end-to-end transmission path as the target end-to-end transmission path.

[0126] The target end-to-end transmission path is a better transmission path configuration strategy, so that the target service can be transmitted from the source node to the destination node with a delay upper bound that meets a preset delay upper bound threshold.

[0127] Exemplarily, in the process of sequentially traversing the at least one end-to-end transmission path, if the currently traversed end-to-end transmission path is the first end-to-end transmission path, the time information and edge shaping parameters involved in the first end-to-end transmission path may be first obtained, and based on the time information and the edge shaping parameters, a first upper delay bound corresponding to the first end-to-end transmission path, i.e., an upper delay bound for the target service when transmitted in the first end-to-end transmission path, may be determined; then, a determination is made as to whether the first upper delay bound is less than a preset upper delay bound threshold: if the first upper delay bound is greater than or equal to the preset upper delay bound threshold, the second end-to-end transmission path is continued to be traversed, and a second upper delay bound corresponding to the second end-to-end transmission path is determined, and the second upper delay bound is determined as the current first upper delay bound, and the step of determining whether the first upper delay bound is less than the preset upper delay bound threshold is simultaneously performed, until an end-to-end transmission path corresponding to a delay upper bound less than the preset upper delay bound threshold is traversed from at least one end-to-end transmission path, and then determined as the target end-to-end transmission path corresponding to the target service. If the first delay upper bound is less than or equal to the preset delay upper bound threshold, the first end-to-end transmission path is determined as the target end-to-end transmission path corresponding to the target service.

[0128] Combined with the above example 1, example 2: for the time-sensitive service f1, the first end-to-end transmission path is traversed. When the current end-to-end transmission path is In the case of You can Issued to The first subnet controller involved. For the first subnet controller, the first service curve set corresponding to the target service calculated by the controller in each local area network is {β1(t),β2(t),β7(t),β9(t),β 10 (t)}, the wide area subnet controller calculates the second service curve of the target service in the corresponding wide area subnet A as According to all the service curves reported by the first subnet controller, the target service is determined to have a current end-to-end transmission path of The delay upper bound corresponding to the time is determined, and the delay upper bound is determined to be less than the preset delay upper bound threshold. At this time, the target end-to-end transmission path of the time-sensitive service f1 is

[0129] For the time-sensitive service f2, the second end-to-end transmission path is traversed. When the current end-to-end transmission path is In the case of You can Issued to The second subnet controller involved. For the second subnet controller, the second service curve set corresponding to the target service calculated by the controller in each local area network is {β1(t),β2(t),β7(t),β9(t),β 11 (t)}, the wide area subnet controller calculates the second service curve of the target service in the corresponding wide area subnet A as According to all the service curves reported by the second subnet controller, the target service is determined to have a current end-to-end transmission path of The corresponding delay upper bound is determined, and the delay upper bound is determined to be greater than the preset delay upper bound threshold. At this time, the next second end-to-end transmission path is continued to be traversed.

[0130] The next second end-to-end transmission path is You can Issued to The third subnet controller involved. For the third subnet controller, the second service curve set corresponding to the target service calculated by the controller in each local area network is {β1(t),β3(t),β6(t),β8(t),β 11 (t)}, the wide area subnet controller calculates the second service curve of the target service in the corresponding wide area subnet B as According to all the service curves reported by the third subnet controller, the target service is determined to have a current end-to-end transmission path of The delay upper bound corresponding to the time is determined, and the delay upper bound is determined to be less than the preset delay upper bound threshold. At this time, the target end-to-end transmission path of the time-sensitive service f2 is

[0131] Finally, for the time-sensitive service f1, the first subnet controller is informed to allow transmission and perform data plane configuration; for the time-sensitive service f2, the third subnet controller is informed to allow transmission and perform data plane configuration.

[0132] In an embodiment of the present invention, the deterministic network orchestration and scheduling process of the target service in a scenario of multiple interconnected local area subnets / wide area subnets can ensure the upper bound of the delay when the target service transmits data in the target end-to-end transmission path, and can meet the target service's data transmission delay requirements.

[0133] It should be noted that the execution subject involved in the embodiment of the present invention can be a device for determining a service transmission path or an orchestrator, which is not specifically limited here.

[0134] The following describes a system for determining a service transmission path provided by the present invention. The system for determining a service transmission path described below and the method for determining a service transmission path described above can refer to each other.

[0135] FIG5 is a schematic diagram of the structure of a system for determining a service transmission path provided by the present invention, which may include:

[0136] The global resource orchestrator 501 is configured to determine at least one end-to-end transmission path corresponding to a target service based on a network topology; sequentially traverse the at least one end-to-end transmission path, and for a currently traversed end-to-end transmission path, send time information and edge shaping parameters related to the currently traversed end-to-end transmission path to at least one local area subnet controller and at least one wide area subnet controller involved in the currently traversed end-to-end transmission path. The edge shaping parameters are used to represent bandwidth resources reserved per unit time by an edge node corresponding to the wide area subnet for the target service.

[0137] The local area subnet controller 502 is configured to determine a first service curve corresponding to the local area subnet of the local area subnet controller based on the time information and the edge shaping parameter;

[0138] The wide area subnet controller 503 is configured to determine a second service curve corresponding to the wide area subnet of the wide area subnet controller based on the time information and the edge shaping parameter;

[0139] The global resource orchestrator 501 is further configured to determine an upper delay bound corresponding to the currently traversed end-to-end transmission path based on all first service curves and all second service curves; if the upper delay bound corresponding to the currently traversed end-to-end transmission path is greater than a preset upper delay bound threshold, determine the next end-to-end transmission path after the currently traversed end-to-end transmission path as the new currently traversed end-to-end transmission path, until the upper delay bound corresponding to the new currently traversed end-to-end transmission path is less than or equal to the preset upper delay bound threshold; if the upper delay bound corresponding to the currently traversed end-to-end transmission path is less than or equal to the preset upper delay bound threshold, determine the currently traversed end-to-end transmission path as the target end-to-end transmission path.

[0140] Optionally, the global resource orchestrator 501 is specifically configured to use the edge shaping parameter for each of the at least one local area subnet controllers involved in the currently traversed end-to-end transmission path to determine a first service curve corresponding to the local area subnet; the edge shaping parameter is also used for each of the at least one wide area subnet controllers involved in the currently traversed end-to-end transmission path to determine a second service curve corresponding to the wide area subnet; and determine the upper bound of the delay corresponding to the currently traversed end-to-end transmission path based on all the first service curves and all the second service curves.

[0141] Optionally, for any first service curve, the time information includes the closing time of the gated queue corresponding to the target service, and the first service curve is obtained by the corresponding local subnet controller 502 based on the following steps: the local subnet controller 502 is specifically configured to obtain the closing time of the gated queue corresponding to the target service; determine the first link bandwidth corresponding to the local subnet controller 502 according to the edge shaping parameter; and determine the first service curve according to the closing time and the first link bandwidth.

[0142] Optionally, for any second service curve, the time information includes the end time of the time slot occupied by the starting node egress port of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the ending node egress port of the transmission link of the target service in the wide area subnet. The second service curve is obtained by the corresponding wide area subnet controller 503 based on the following steps: the wide area subnet controller 503 is specifically configured to determine the maximum intra-domain delay of the wide area subnet corresponding to the wide area subnet controller based on the end time of the time slot occupied by the starting node egress port of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the ending node egress port of the transmission link of the target service in the wide area subnet; determine the second link bandwidth corresponding to the wide area subnet controller 503 based on the edge shaping parameter; and determine the second service curve based on the maximum intra-domain delay and the second link bandwidth.

[0143] Optionally, the local subnet controller 502 is specifically configured to determine the first service curve according to a first function formula; wherein the first function formula is: β e (t) represents the first service curve of the local area subnet controller corresponding to the mth local area subnet at time t; W e represents the first link bandwidth of the transmission link e in the m-th local area subnet; Indicates the closing time; [·] + Indicates a positive number; U m Indicates the local area subnet set corresponding to the mth local area subnet; Represents the transmission link of the mth local area subnet.

[0144] Optionally, the wide area subnet controller 503 is specifically configured to determine the second service curve according to a second function formula; wherein the second function formula is: represents the second service curve of the wide area subnet controller corresponding to the nth wide area subnet at time t for the target service f; represents the second link bandwidth of the target service f in the nth wide area subnet; τ n,f Indicates the maximum intra-domain delay of the target service f in the nth wide area subnet.

[0145] Optionally, the global resource scheduler 501 is specifically configured to determine the total service curve corresponding to the currently traversed end-to-end transmission path according to the service curve function; determine the delay upper bound corresponding to the currently traversed end-to-end transmission path according to the total service curve; wherein the service curve function is β * e (t) represents the total service curve corresponding to the end-to-end transmission path currently traversed at time t; W e represents the first link bandwidth of transmission link e in the mth local area subnet; Indicates the closing time; [·] + Indicates a positive number; represents the local subnet set corresponding to the mth local subnet; M represents the total number of local subnets; represents the second link bandwidth of the target service f in the nth wide area subnet; τ n,f represents the maximum intra-domain delay of the target service f in the nth wide area subnet; v up Represents a node set, including m local subnet nodes and n virtual nodes Each virtual node n wide area subnets Jointly labeled with the target business f; Represents the node set v up Except for the m local subnet nodes All nodes except .

[0146] The following describes a device for determining a service transmission path provided by the present invention. The device for determining a service transmission path described below and the method for determining a service transmission path described above can refer to each other.

[0147] FIG6 is a schematic diagram of the structure of the device for determining a service transmission path provided by the present invention, which may include:

[0148] The transmittable path and bandwidth determination module 601 is configured to determine at least one end-to-end transmission path corresponding to the target service according to the network topology;

[0149] The target transmission path determination module 602 is configured to sequentially traverse the at least one end-to-end transmission path, determine an upper delay bound corresponding to the currently traversed end-to-end transmission path based on time information associated with the currently traversed end-to-end transmission path and an edge shaping parameter of the currently traversed end-to-end transmission path, the edge shaping parameter being used to represent bandwidth resources reserved per unit time by an edge node corresponding to the wide area subnet for the target service; determine, if the upper delay bound corresponding to the currently traversed end-to-end transmission path is greater than a preset upper delay bound threshold, the next end-to-end transmission path after the currently traversed end-to-end transmission path as the new currently traversed end-to-end transmission path, until the upper delay bound corresponding to the new currently traversed end-to-end transmission path is less than or equal to the preset upper delay bound threshold; and determine, if the upper delay bound corresponding to the currently traversed end-to-end transmission path is less than or equal to the preset upper delay bound threshold, the currently traversed end-to-end transmission path as the target end-to-end transmission path.

[0150] Optionally, the target transmission path determination module 602 is specifically configured such that the edge shaping parameter is used for each of the at least one local subnet controllers involved in the currently traversed end-to-end transmission path to determine a first service curve corresponding to the local subnet; the edge shaping parameter is also used for each of the at least one wide area subnet controllers involved in the currently traversed end-to-end transmission path to determine a second service curve corresponding to the wide area subnet; and based on all the first service curves and all the second service curves, the delay upper bound corresponding to the currently traversed end-to-end transmission path is determined.

[0151] Optionally, for any first service curve, the time information includes the closing time of the gated queue corresponding to the target service, and the first service curve is obtained based on the following steps: obtaining the closing time of the gated queue corresponding to the target service; and determining the first link bandwidth corresponding to the local subnet controller based on the edge shaping parameter.

[0152] Optionally, for any second service curve, the time information includes the end time of the time slot occupied by the starting node egress port of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the ending node egress port of the transmission link of the target service in the wide area subnet. The second service curve is obtained based on the following steps: determining the maximum intra-domain delay of the wide area subnet corresponding to the wide area subnet controller according to the end time of the time slot occupied by the starting node egress port of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the ending node egress port of the transmission link of the target service in the wide area subnet; determining the second link bandwidth corresponding to the wide area subnet controller according to the edge shaping parameter; and determining the second service curve according to the maximum intra-domain delay and the second link bandwidth.

[0153] Optionally, the first service curve is determined according to a first function formula; wherein the first function formula is: W represents the first service curve of the local subnet controller corresponding to the mth local subnet at time t; e represents the first link bandwidth of the transmission link e in the m-th local area subnet; Indicates the closing time; [·] + Indicates a positive number; U m Indicates the local area subnet set corresponding to the mth local area subnet; Represents the transmission link of the mth local area subnet.

[0154] Optionally, the second service curve is determined according to a second function formula; wherein the second function formula is: represents the second service curve of the wide area subnet controller corresponding to the nth wide area subnet at time t for the target service f; represents the second link bandwidth of the target service f in the nth wide area subnet; τ n,f Indicates the maximum intra-domain delay of the target service f in the nth wide area subnet.

[0155] Optionally, the target transmission path determination module 602 is specifically configured to determine the total service curve corresponding to the currently traversed end-to-end transmission path according to the service curve function; determine the delay upper bound corresponding to the currently traversed end-to-end transmission path according to the total service curve; wherein the service curve function is β * e (t) represents the total service curve corresponding to the end-to-end transmission path currently traversed at time t; W e represents the first link bandwidth of transmission link e in the mth local area subnet; Indicates the closing time; [·] + Indicates a positive number; represents the local subnet set corresponding to the mth local subnet; M represents the total number of local subnets; represents the second link bandwidth of the target service f in the nth wide area subnet; τ n,f represents the maximum intra-domain delay of the target service f in the nth wide area subnet; v up Represents a node set, including m local subnet nodes and n virtual nodes Each virtual node n wide area subnets Jointly labeled with the target business f; Represents the node set v up Except for the m local subnet nodes All nodes except .

[0156] As shown in FIG7 , it is a schematic diagram of the structure of the electronic device provided by the present invention, and the electronic device may include: a processor (processor) 710, a communication interface (Communications Interface) 720, a memory (memory) 730 and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other through the communication bus 740. The processor 710 can call the logic instructions in the memory 730 to execute a method for determining a service transmission path, which includes: determining at least one end-to-end transmission path corresponding to the target service according to the network topology structure; traversing the at least one end-to-end transmission path in sequence, and determining the upper bound of the delay corresponding to the end-to-end transmission path currently traversed according to the time information involved in the end-to-end transmission path currently traversed and the edge shaping parameter of the end-to-end transmission path currently traversed, wherein the edge shaping parameter is used to characterize the bandwidth resources reserved by the edge node corresponding to the wide area subnet for the target service within a unit time. source; when the upper bound of the delay corresponding to the currently traversed end-to-end transmission path is greater than a preset upper bound of the delay threshold, the next end-to-end transmission path of the currently traversed end-to-end transmission path is determined as the new end-to-end transmission path currently traversed, until the upper bound of the delay corresponding to the new currently traversed end-to-end transmission path is less than or equal to the preset upper bound of the delay threshold; when the upper bound of the delay corresponding to the currently traversed end-to-end transmission path is less than or equal to the preset upper bound of the delay threshold, the currently traversed end-to-end transmission path is determined as the target end-to-end transmission path.

[0157] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0158] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the service transmission path determination method provided by the above methods, the method comprising: determining at least one end-to-end transmission path corresponding to the target service according to the network topology structure; traversing the at least one end-to-end transmission path in sequence, and determining the delay upper bound corresponding to the currently traversed end-to-end transmission path according to the time information involved in the currently traversed end-to-end transmission path and the edge shaping parameters of the currently traversed end-to-end transmission path, the edge shaping parameters Used to characterize the bandwidth resources reserved by the edge node corresponding to the wide area subnet for the target service within unit time; when the delay upper bound corresponding to the currently traversed end-to-end transmission path is greater than a preset delay upper bound threshold, the next end-to-end transmission path of the currently traversed end-to-end transmission path is determined as the new currently traversed end-to-end transmission path, until the delay upper bound corresponding to the new currently traversed end-to-end transmission path is less than or equal to the preset delay upper bound threshold; when the delay upper bound corresponding to the currently traversed end-to-end transmission path is less than or equal to the preset delay upper bound threshold, the currently traversed end-to-end transmission path is determined as the target end-to-end transmission path.

[0159] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for determining a service transmission path provided by each of the above methods, the method comprising: determining, based on a network topology, at least one end-to-end transmission path corresponding to a target service; traversing the at least one end-to-end transmission path in sequence, and determining an upper delay bound corresponding to the currently traversed end-to-end transmission path based on time information involved in the currently traversed end-to-end transmission path and an edge shaping parameter of the currently traversed end-to-end transmission path, the edge shaping parameter being used to characterize bandwidth resources reserved by an edge node corresponding to a wide area subnet for the target service per unit time; and, if the upper delay bound corresponding to the currently traversed end-to-end transmission path is greater than a preset upper delay bound threshold, determining the next end-to-end transmission path of the currently traversed end-to-end transmission path as the new currently traversed end-to-end transmission path. Until the new upper delay bound corresponding to the currently traversed end-to-end transmission path is less than or equal to the preset upper delay bound threshold; when the upper delay bound corresponding to the currently traversed end-to-end transmission path is less than or equal to the preset upper delay bound threshold, the currently traversed end-to-end transmission path is determined as the target end-to-end transmission path.

[0160] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0161] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0162] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention. Industrial Applicability

[0163] The solution provided by the embodiment of the present application can be applied to the field of service transmission technology. In the embodiment of the present application, at least one end-to-end transmission path corresponding to the target service is determined based on the network topology; the at least one end-to-end transmission path is traversed in sequence, and the upper bound of the delay corresponding to the currently traversed end-to-end transmission path is determined based on the time information involved in the currently traversed end-to-end transmission path and the edge shaping parameters of the currently traversed end-to-end transmission path; the target end-to-end transmission path corresponding to the target service is determined based on the upper bound of the delay corresponding to the currently traversed end-to-end transmission path. This method can ensure the upper bound of the delay when the target service transmits data in the target end-to-end transmission path in the deterministic network orchestration and scheduling process of the target service in the scenario of multiple interconnected local area subnets / wide area subnets, and can meet the target service's requirements for data transmission delay.

Claims

1. A method for determining a service transmission path, comprising: Determine at least one end-to-end transmission path corresponding to the target service according to the network topology; Traversing the at least one end-to-end transmission path in sequence, and determining the delay upper bound corresponding to the end-to-end transmission path currently traversed according to time information involved in the end-to-end transmission path currently traversed and an edge shaping parameter of the end-to-end transmission path currently traversed, wherein the edge shaping parameter is used to characterize the bandwidth resources reserved by the edge node corresponding to the wide area subnet for the target service within a unit time; In the case where the delay upper bound corresponding to the currently traversed end-to-end transmission path is greater than a preset delay upper bound threshold, determining the next end-to-end transmission path of the currently traversed end-to-end transmission path as the new currently traversed end-to-end transmission path, until the delay upper bound corresponding to the new currently traversed end-to-end transmission path is less than or equal to the preset delay upper bound threshold; When the delay upper bound corresponding to the currently traversed end-to-end transmission path is less than or equal to a preset delay upper bound threshold, the currently traversed end-to-end transmission path is determined as the target end-to-end transmission path.

2. The method according to claim 1, wherein: The edge shaping parameter is used for at least one local area subnet controller involved in the currently traversed end-to-end transmission path to respectively determine a first service curve of the corresponding local area subnet; the edge shaping parameter is also used for at least one wide area subnet controller involved in the currently traversed end-to-end transmission path to respectively determine a second service curve of the corresponding wide area subnet; The determining, according to time information involved in the currently traversed end-to-end transmission path and an edge shaping parameter of the currently traversed end-to-end transmission path, an upper bound of a delay corresponding to the currently traversed end-to-end transmission path includes: An upper bound of a delay corresponding to the currently traversed end-to-end transmission path is determined according to all the first service curves and all the second service curves.

3. The method according to claim 2, wherein: For any first service curve, the time information includes a closing time of a gated queue corresponding to the target service, and the first service curve is obtained based on the following steps: Obtaining the closing time of the gated queue corresponding to the target service; Determining a first link bandwidth corresponding to the local subnet controller according to the edge shaping parameter; The first service curve is determined according to the shutdown time and the first link bandwidth.

4. The method according to claim 2, wherein: For any second service curve, the time information includes the end time of the time slot occupied by the start node egress port of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the end node egress port of the transmission link of the target service in the wide area subnet, and the second service curve is obtained based on the following steps: Determine the maximum intra-domain delay of the wide area subnet controller corresponding to the wide area subnet according to the end time of the time slot occupied by the egress port of the starting node of the transmission link of the target service in the wide area subnet and the end time of the time slot occupied by the egress port of the ending node of the transmission link of the target service in the wide area subnet; Determining a second link bandwidth corresponding to the wide area subnet controller according to the edge shaping parameter; The second service curve is determined according to the maximum delay within the domain and the second link bandwidth.

5. The method according to claim 3, wherein: The determining the first service curve according to the shutdown time and the first link bandwidth includes: Determining the first service curve according to a first function formula; Wherein, the first function formula is: β e (t) represents the first service curve of the local area subnet controller corresponding to the mth local area subnet at time t; W e represents the first link bandwidth of the transmission link e in the m-th local area subnet; Indicates the closing time; [·] + Indicates a positive number; U m Indicates the mth The local subnet set corresponding to the local subnet; Represents the transmission link of the mth local area subnet.

6. The method according to claim 4, wherein: The determining the second service curve according to the maximum delay in the domain and the second link bandwidth includes: Determining the second service curve according to the second function formula; Wherein, the second function formula is: Indicates a second service curve of the wide area subnet controller corresponding to the nth wide area subnet at time t for the target service f; represents the second link bandwidth of the target service f in the nth wide area subnet; τ n，f It represents the maximum delay of the target service f in the nth wide area subnet.

7. The method according to claim 3, wherein: The determining, according to all the first service curves and all the second service curves, a delay upper bound corresponding to the currently traversed end-to-end transmission path includes: Determining, according to the service curve function, a total service curve corresponding to the currently traversed end-to-end transmission path; Determining, according to the total service curve, an upper bound of a delay corresponding to the currently traversed end-to-end transmission path; Wherein, the service curve function is: β * e W (t) represents the total service curve corresponding to the end-to-end transmission path currently traversed at time t; e represents the first link bandwidth of the transmission link e in the mth local area subnet; Indicates the closing time; [·] + Indicates a positive number; represents the local area subnet set corresponding to the m-th local area subnet; M represents the total number of local area subnets; Indicates the target industry The second link bandwidth of service f in the nth wide area subnet; τ n，f represents the maximum delay of the target service f in the nth wide area subnet; v up Represents a node set, including m local subnet nodes and n virtual nodes Each of the virtual nodes n wide area subnets Labeled together with the target business f; Represents the node set v up In addition to the m local subnet nodes All nodes except 8. A system for determining a service transmission path, comprising: Global resource orchestrator, local area subnet controller, and wide area subnet controller; A global resource scheduler is configured to determine at least one end-to-end transmission path corresponding to a target service according to a network topology structure; traverse the at least one end-to-end transmission path in sequence, and for a currently traversed end-to-end transmission path, send time information and edge shaping parameters related to the currently traversed end-to-end transmission path to at least one local subnet controller and at least one wide area subnet controller related to the currently traversed end-to-end transmission path, wherein the edge shaping parameters are used to characterize bandwidth resources reserved by an edge node corresponding to the wide area subnet for the target service within a unit time; For each local area subnet controller, the local area subnet controller is configured to determine a first service curve of the local area subnet corresponding to the local area subnet controller based on the time information and the edge shaping parameter; For each wide area subnet controller, the wide area subnet controller is configured to determine a second service curve of the wide area subnet corresponding to the wide area subnet controller based on the time information and the edge shaping parameter; The global resource scheduler is further configured to determine the delay upper bound corresponding to the currently traversed end-to-end transmission path according to all the first service curves and all the second service curves; when the delay upper bound corresponding to the currently traversed end-to-end transmission path is greater than a preset delay upper bound threshold, determine the next end-to-end transmission path of the currently traversed end-to-end transmission path as the new end-to-end transmission path currently traversed, until the new end-to-end transmission path is reached. The delay upper bound corresponding to the currently traversed end-to-end transmission path is less than or equal to the preset delay upper bound threshold; When the delay upper bound corresponding to the currently traversed end-to-end transmission path is less than or equal to a preset delay upper bound threshold, the currently traversed end-to-end transmission path is determined as the target end-to-end transmission path.

9. A device for determining a service transmission path, the device comprising: A transmission path and bandwidth determination module, configured to determine at least one end-to-end transmission path corresponding to the target service according to the network topology; The target transmission path determination module is configured to sequentially traverse the at least one end-to-end transmission path, and determine the delay upper bound corresponding to the end-to-end transmission path currently traversed according to the time information involved in the end-to-end transmission path currently traversed and the edge shaping parameter of the end-to-end transmission path currently traversed; when the delay upper bound corresponding to the end-to-end transmission path currently traversed is greater than a preset delay upper bound threshold, determine the next end-to-end transmission path of the end-to-end transmission path currently traversed as the new end-to-end transmission path currently traversed, until the delay upper bound corresponding to the new end-to-end transmission path currently traversed is less than or equal to the preset delay upper bound threshold; When the delay upper bound corresponding to the currently traversed end-to-end transmission path is less than or equal to a preset delay upper bound threshold, the currently traversed end-to-end transmission path is determined as the target end-to-end transmission path.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for determining a service transmission path as claimed in any one of claims 1 to 7 when executing the program.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for determining a service transmission path according to any one of claims 1 to 7 is implemented.

Images