WO2013159451A1

WO2013159451A1 - Method and apparatus for acquiring cross-domain end-to-end route, and child path computation entity

Info

Publication number: WO2013159451A1
Application number: PCT/CN2012/077972
Authority: WO
Inventors: 杨发明; 何桓; 张锋; 石浩鸿
Original assignee: 中兴通讯股份有限公司
Priority date: 2012-04-24
Filing date: 2012-06-29
Publication date: 2013-10-31
Also published as: CN103379032B; CN103379032A

Abstract

Disclosed are a method and an apparatus for acquiring a cross-domain end-to-end path, and a child path computation entity (PCE). The method comprises: a parent PCE receiving a first request message; under triggering of the first request message, the parent PCE sending a second request message to a child PCE thereof involved in a cross-domain path computation; the parent PCE receiving cross-domain topology information determined and reported by the child PCE according to the second request message; and the parent PCE computing an optimal cross-domain end-to-end path according to the received cross-domain topology information. By means of the above technical solution provided in the present invention, problems in the art such as poor accuracy of the optimal cross-domain end-to-end path computed through a cross-domain path technology are solved, so that network topology information finally obtained by the parent PCE for cross-domain path computation can better reflect actual current conditions of a current network topology, thereby improving the accuracy of the path computation.

Description

The present invention relates to the field of communications, and in particular to a method and device for acquiring cross-domain end-to-end routes and a sub-routing computing entity. BACKGROUND OF THE INVENTION As the size of a network increases, the size of network topology data used for end-to-end route calculation increases to such an extent that the device cannot withstand it. In order to solve this problem, a domain routing technology is required. The main features of the domain-based routing technology are: The network is divided into several domains, and the routing calculation unit of each domain only sees its own network topology, but does not see the network topology of other domains, so that the network topology load of the large-scale network is shared. Go to the routing calculation unit of each domain. In the terminology described in this document: "path" and "route" have the same meaning; "shortest short circuit" has the same meaning as "optimal route", and when there are routing constraints and constraint strategies, the shortest path is here to satisfy the routing constraints. The shortest path to the constraint strategy is the same; the "network element" and the "node" have the same meaning. In order to calculate cross-domain end-to-end routing, there are currently three technical methods: (1) per-domain routing calculation; (2) reverse recursive routing calculation; (3) hierarchical routing computing entity (PCE) Route calculation. The main method features of each domain route calculation are as follows: The domain ingress node is responsible for calculating the intra-domain routes that reach the destination through the egress of the domain, and performs calculations on a per-domain basis to obtain end-to-end routes. The reverse recursive route calculation method mainly consists of the following method steps: (1) Starting from the PCE of the domain where the destination node is located. The virtual shortest path tree (Virtual Shortest Path Tree, VSPT) is obtained by calculating the shortest distance from all the ingress nodes of the domain to the destination node; and the calculated VPST is transmitted to the PCEs of the adjacent domain;

(2) The PCE in the latter domain performs the following operations in sequence. According to the neighbor's VPSTs, the virtual shortest path tree (VPST) is obtained and diffused to the neighbors by calculating the shortest distance from all the ingress nodes of the domain to the destination node. The calculation needs to be satisfied: The path must pass through the leaves of the neighbor VPSTs; (3) The PCE of the domain where the source node is located performs the following operations to obtain an end-to-end cross-domain short-circuit. Calculate the shortest path from the source node to the destination node based on the neighbor's VPSTs. The calculation needs to be satisfied: The path must pass through the leaves of the neighbor VPSTs. The hierarchical PCE performs cross-domain end-to-end route calculation through the cooperation of the parent PCE and the child PCE. The child PCE is responsible for calculating the intra-domain path, and the parent PCE is responsible for calculating the cross-domain path. Hierarchical PCE only gives a method architecture, and there are different ways to implement it. One of the implementation methods of the hierarchical PCE is as follows: The sub-PCE calculates the intra-domain path according to the end-to-end route calculation request and maps the intra-domain path to a logical link. These logical links form a virtual topology of a domain, and the parent PCE forms all sub-PCEs. The virtual topology and the other inter-domain links form a network virtual topology, whereby the network virtual topology calculates cross-domain end-to-end routing. The computer system of each domain path is simple and suitable for path calculation with simple inter-domain connection. However, it is difficult to find the optimal route for complex path calculation between inter-domain connections. The advantage of reverse recursive routing calculation is that cross-domain end-to-end optimal routing can be obtained. The disadvantages are: need to run the protocol, repeated recursive calculation through distributed operations, the method is more complicated, and the calculation efficiency is low. The advantage of the hierarchical PCE route calculation method is that the cross-domain end-to-end optimal route can be obtained, and the relative reverse recursive route calculation method does not require recursive calculation, and the method is simple. However, the cross-domain topology information used by the parent PCE to calculate the cross-domain topology information is the pre-obtained inter-domain link information, and is not the real-time obtained current network topology. Therefore, the final cross-domain end-to-end optimal route is calculated. The accuracy is poor. In view of the above problems in the related art, an effective solution has not yet been proposed. SUMMARY OF THE INVENTION The present invention provides a cross-domain end-to-end route acquisition method, device, and sub-route, which are related to the problem that the cross-domain end-to-end optimal route is poorly calculated by the cross-domain routing technology. Calculate the entity to at least solve the above problem. According to an aspect of the present invention, a method for obtaining an inter-domain end-to-end route is provided, including: a parent route calculation entity (PCE) receives a first request message; and a parent PCE sends a message to the first request message The sub-PCE of the parent PCE participating in the cross-domain routing calculation sends the second request message; the parent PCE receives the cross-domain topology information determined and reported by the sub-PCE according to the second request message; the parent PCE according to the received cross-domain The topology information is calculated to obtain an end-to-end optimal route across domains. The second request message carries the end-to-end routing request information and the specified domain identifier, where the end-to-end routing request information carries the selection policy used by the sub-PCE to select a route. The first request message carries the selection policy; the selection policy carried in the second request message is obtained by: obtaining, by the parent PCE, the selection policy from the first request message, and Carry in the second request message. The cross-domain topology information includes at least one of the following: an intra-domain route, an inter-domain link, and an inter-domain intra-domain link, where two endpoints of the intra-domain intra-domain link are respectively two boundary network elements of the domain. on. The determining, by the sub-PCE, the cross-domain topology information according to the received second request message, the method includes: selecting, by the sub-PCE, an available inter-domain chain from an inter-domain link between a specified domain and a neighboring domain according to the selection policy. Determining, as well as determining, by the sub-PCE, the network element where the endpoints of the selected inter-domain links are located as the cross-domain topology network element; and determining, by the sub-PCE, the determined cross-domain topology network element Intra-domain routing and the intra-domain intra-domain link. Determining, by the foregoing sub-PCE, the intra-domain route and the inter-domain intra-domain link according to the determined cross-domain topology network element, including at least one of the following: processing any two of the cross-domain topology networks in the specified domain If the route of the two cross-domain topological network elements indicated by the local policy of the specified domain can only use or must use the link between the two ends of the cross-domain topology network element, Searching for, in the links of the two cross-domain topological network elements, the links that meet the selection policy, as the inter-domain intra-domain links, for any two of the specified domains. The cross-domain topology network element, if the local policy requires the cross-domain routing of the two cross-domain topology network elements, the link between the two ends of the cross-domain topology network element must be preferably used, but not There is a link that meets the selection policy, or the local policy does not require that the inter-domain route through the two cross-domain topology NEs can only use or preferably use the two ends of the link in the two cross-domain topology network elements respectively. Link An intra-domain route that meets the selection policy of the cross-domain topology network element; in the intra-domain route, if the intra-domain route passes only one hop link, the link is the inter-domain intra-domain link. The intra-domain route of the specified domain is not reported. Otherwise, the calculated intra-domain route is reported to the parent PCE. The intra-domain route reported to the parent PCE includes the following information: The network element, link, and intra-domain route cost that the intra-domain route passes through. The intra-domain route reported to the parent PCE also includes the following information: Bandwidth information. The bandwidth information is determined by the bandwidth of all links through which the intra-domain route passes, wherein the idle available bandwidth value of the bandwidth information takes the minimum value of the available bandwidth of all links. The above method further includes: the parent PCE mapping the intra-domain route to a logical link. The bandwidth of the foregoing logical link is determined by: when the bandwidth information is included in the intra-domain route reported to the parent PCE, the bandwidth of the logical link is determined by the bandwidth information; otherwise, it is carried by the second request message. The bandwidth attribute value of the end-to-end routing request information is determined. Before receiving the first request message, the parent PCE further includes: configuring, for the parent PCE, domain information of each domain required to calculate the cross-domain end-to-end optimal route and sub-PCE information associated with each domain. The parent PCE calculates the cross-domain end-to-end optimal route according to the received cross-domain topology information, including: the parent PCE adds the received cross-domain topology information to the entire network topology data; The cross-domain end-to-end optimal route is calculated according to the network-wide topology data after the cross-domain topology information is added. According to another aspect of the present invention, a cross-domain end-to-end routing acquisition device is provided, which is located in a parent routing computing entity (PCE), and includes: a first receiving module, configured to receive a first request message; and a sending module, And being configured to send, by the first request message, a second request message to the sub-PCE of the parent PCE participating in the cross-domain routing calculation; the second receiving module is configured to receive the sub-PCE according to the second request The cross-domain topology information is determined and reported by the message; and the calculating module is configured to calculate the cross-domain end-to-end optimal route according to the received cross-domain topology information. The second request message carries the end-to-end routing request information and the specified domain identifier, where the end-to-end routing request information carries the selection policy of the sub-PCE routing. The cross-domain topology information includes at least one of the following: an intra-domain route, an inter-domain link, and an inter-domain intra-domain link, where two endpoints of the intra-domain intra-domain link are respectively two boundary network elements of the domain. on. The calculation module includes: an adding unit, configured to add the cross-domain topology information received by the receiving module to the entire network topology data; and the calculating unit is configured to set the entire network topology data according to the added cross-domain topology information The cross-domain end-to-end optimal route is calculated. According to still another aspect of the present invention, a seed routing computing entity is provided, comprising: a receiving module, configured to receive a second request message from a parent routing computing entity PCE, wherein the second request message is in a parent PCE After receiving the first request message, the first request message is triggered to be sent by the parent PCE, and the reporting module is configured to: determine, according to the received second request message, report the cross-domain topology information to the parent PCE. . The foregoing reporting module determines the cross-domain topology information by: selecting an available inter-domain link from the inter-domain link of the specified domain and the adjacent domain according to the selection policy, and placing the selected inter-domain link and located The network element where the endpoints of the specified domain are located is determined to be a cross-domain topology network element, where the selection policy is determined by the second request message. The end-to-end routing request information carried in the path is carried and used to select a route; and the intra-domain route and the intra-domain intra-domain link are determined according to the determined cross-domain topology network element. The reporting module determines the intra-domain route and the inter-domain intra-domain link according to the determined cross-domain topology network element by using one of the following processes: for any two of the cross-domain topology network elements in the specified domain, If the routes of the two cross-domain topological network elements indicated by the local policy of the specified domain can only use or must use the links of the two ends of the cross-domain topology network element respectively, the slave chain A link at the two ends of the cross-domain topology network element is searched for the link that meets the selection policy, as the cross-domain intra-domain link, and any two of the cross-domains in the specified domain. The topology network element, if the local policy requires the cross-domain routing of the two cross-domain topology network elements, the link between the two ends of the cross-domain topology network element is preferably used, but there is no match. The link of the selected policy, or the local policy, does not require that the cross-domain routing of the two cross-domain topologies may only use or preferably use the two ends of the link in the chain of the two cross-domain topologies. Road, then calculate The intra-domain route of the cross-domain topology network element that meets the selection policy; in the calculated intra-domain route, if the intra-domain route only passes through a one-hop link, the link is the inter-domain intra-domain link. The intra-domain route of the specified domain is not reported. Otherwise, the calculated intra-domain route is reported to the parent PCE. With the present invention, the slave PCE obtains the slaves under the trigger of the received first request message.

The inter-domain topology information of the PCE is calculated based on the cross-domain topology information obtained at this time, and the cross-domain end-to-end optimality calculated by the cross-domain routing technology is solved. The accuracy of the route is poor. The network topology information obtained by the parent PCE for cross-domain routing calculation can better reflect the current network topology and improve the accuracy of route calculation. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are set to illustrate,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, In the drawings: FIG. 1 is a flowchart of a method for obtaining cross-domain end-to-end routing according to an embodiment of the present invention; FIG. 2 is a structural block diagram of an apparatus for acquiring cross-domain end-to-end routing according to an embodiment of the present invention; FIG. 4 is a structural block diagram of a sub-route calculation entity according to an embodiment of the present invention; FIG. 5 is a cross-sectional view of a sub-route calculation entity according to an embodiment of the present invention; A flowchart of an acquisition method of an end-to-end route; 6 is a schematic diagram of a network including multiple domains and a PCE hierarchy according to an embodiment of the present invention; FIG. 7 is a schematic diagram of a process for determining a cross-domain topology by a sub-PCE according to an embodiment of the present invention; FIG. 8 is a cross-domain topology formed by a parent PCE. Schematic diagram of the entire network topology of the end-to-end optimal route of the domain. BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, the present invention will be described in detail with reference to the accompanying drawings. It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict. FIG. 1 is a flowchart of a method for obtaining cross-domain end-to-end routing according to an embodiment of the present invention. As shown in FIG. 1, the method includes: Step S102: A parent route calculation entity (PCE) receives a first request message. Step S104: A parent PCE participates in a cross-domain participation of the parent PCE by triggering the first request message. The sub-PCE of the route calculation sends the second request message; Step S106, the parent PCE receives the cross-domain topology information determined and reported by the sub-PCE according to the second request message; Step S108, the parent PCE is configured according to the cross-domain topology received The information is calculated to obtain an end-to-end optimal route across domains. Through the above processing steps, since the parent PCE obtains the cross-domain topology information from the child PCE instead of the parent PCE, the cross-domain topology information is obtained in advance, which enables the last obtained cross-domain route calculation. The network topology can reflect the actual status of the current network topology in real time and improve the accuracy of route calculation. The second request message carries the end-to-end routing request information and the specified domain identifier, where the end-to-end routing request information carries the selection policy used by the sub-PCE to select a route. In this way, the sub-PCE can select the link or the route that meets the selection policy to be reported to the parent PCE according to the foregoing selection policy, and the parent PCE sends the second request message to the sub-PCE corresponding to the specified domain identifier according to the specified domain identifier. The foregoing first request message carries the selection policy; based on this, the selection policy carried in the second request message may be obtained by: obtaining, by the parent PCE, the selection policy from the first request message, and It is carried in the second request message. The process described in the foregoing embodiment may be in the following form. After receiving the cross-domain end-to-end path calculation request (equivalent to the first request message), the parent PCE requests all the sub-PCEs that need to participate in the cross-domain path calculation to the parent PCE. A cross-domain topology (equivalent to a second request message) is provided, where the cross-domain topology request includes cross-domain end-to-end path request information and a specified domain. The end-to-end path request information includes attribute information of the route (which may be considered as a factor of the above selection policy): bandwidth, source node address, destination node address, constraint routing information, and the like. If there is routing constraint information, the parent PCE will send the "cross-domain topology request" to the end-to-end routing request based on the constraint routing information to allow the sub-PCE of the domain to pass. The cross-domain topology information may include at least one of the following: an intra-domain route, an inter-domain link, and an inter-domain intra-domain link, where the two endpoints of the intra-domain intra-domain link are respectively in the two border networks of the domain. Yuan. Determining, according to the received second request message, the cross-domain topology information may be implemented by using the following process: the sub-PCE selects an available domain from the inter-domain link between the specified domain and the adjacent domain according to the selection policy. An inter-link, and a network element where the endpoints of the selected inter-domain link and located in each of the specified domains are located as a cross-domain topology network element; and the sub-PCE determines the cross-domain topology network element according to the determined Intra-domain routing and the intra-domain intra-domain link. Determining, by the foregoing sub-PCE, the intra-domain route and the inter-domain intra-domain link according to the determined cross-domain topology network element, which may be implemented by using at least one of the following processes: (1) for any two of the cross-domains within the specified domain The domain topology network element, if the route of the two cross-domain topology network elements indicated by the local policy of the specified domain can only be used or must be preferably used, the two ends of the link are respectively in the chain of the two cross-domain topology network elements. And the link between the two ends of the cross-domain topology network element is searched for the link that meets the selection policy, as the inter-domain intra-domain link;

(2) For any two of the cross-domain topological network elements in the specified domain, if the local policy requires cross-domain routing through the two cross-domain topological network elements, the two ends of the link must be preferably used in two locations. A link of a cross-domain topology network element, but there is no link that meets the selection policy, or the local policy does not require cross-domain routing through the two cross-domain topology network elements to use or preferably use the link two. If the endpoints are in the two links of the cross-domain topology network element, the intra-domain routes that meet the selection policy are calculated through the cross-domain topology network element. If the one-hop link is used, the link is the intra-domain link, and the intra-domain route of the specified domain is not reported. Otherwise, the calculated intra-domain route is reported to the parent PCE. In the above process, the intra-domain route reported to the parent PCE may include the following information: the endpoint, the link, and the intra-domain routing cost of the intra-domain route. The intra-domain routing cost may be determined by characteristics such as the hop count and link delay of the intra-domain route. In a preferred implementation of the present invention, the intra-domain route reported to the parent PCE further includes the following information: bandwidth information. The bandwidth information is determined by the bandwidth of all links through which the path in the domain passes, and the idle available bandwidth value of the bandwidth information takes the minimum value of the available bandwidth of all links. When the intra-domain routing information is included in the cross-domain topology information reported to the parent PCE, the parent PCE needs to map the intra-domain routing to a logical link. If the reported information does not include bandwidth information, the bandwidth of the logical link is determined by the bandwidth attribute value of the end-to-end routing request information. When the intra-domain route reported to the parent PCE further includes the bandwidth information, at this time, if the parent PCE maps the intra-domain route to a logical link, the bandwidth of the mapped logical link is determined by the bandwidth information. Before receiving the first request message, the parent PCE may further include the following process: configuring, for the parent PCE, domain information of each domain required for calculating the cross-domain end-to-end optimal route and sub-PCE information associated with each domain . In a specific application, the configured domain information may include a domain identifier, and the configured sub-PCE information includes a sub-PCE identifier and a sub-PCE communication address. The parent PCE calculates the cross-domain end-to-end optimal route according to the received cross-domain topology information, and may be implemented by the foregoing process: the parent PCE adds the received cross-domain topology information to the entire network topology data. The parent PCE calculates the cross-domain end-to-end optimal route according to the network-wide topology data after the cross-domain topology information is added. In this embodiment, an apparatus for obtaining an inter-domain end-to-end route is provided. The device is located in the parent route calculation entity PCE, and is used to implement the foregoing embodiments and preferred embodiments. The modules involved in the device will be described below. The term "module" as used hereinafter may implement a combination of software and/or hardware for a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware, is also possible and conceivable. 2 is a structural block diagram of an apparatus for acquiring cross-domain end-to-end routing according to an embodiment of the present invention. As shown in FIG. 2, the device includes: a first receiving module 20, configured to receive a first request message; a sending module 22, connected to the first receiving module 20, configured to be triggered by the first request message The sub-PCE of the parent PCE participating in the cross-domain routing calculation sends a second request message; the second receiving module 24 is communicatively connected to the sub-PCE 28, and is configured to receive the cross-determined and reported cross-sub-PCE according to the second request message. Domain topology information; The calculation module 26 is connected to the second receiving module 24, and is configured to calculate the cross-domain end-to-end optimal route according to the received cross-domain topology information. The second request message carries the end-to-end routing request information. And specifying a domain identifier, where the end-to-end routing request information carries a selection policy of the sub-PCE routing. Preferably, the cross-domain topology information includes at least one of the following: an intra-domain route, an inter-domain link, and an inter-domain intra-domain link, where two endpoints of the intra-domain intra-domain link are respectively in the domain. On the boundary network element. In a preferred implementation of the present invention, as shown in FIG. 3, the foregoing calculating module 26 includes: an adding unit 260, configured to add the cross-domain topology information received by the receiving module to the entire network topology data; The unit 262 is connected to the adding unit 260, and is configured to calculate the cross-domain end-to-end optimal route according to the network-wide topology data after adding the cross-domain topology information. The embodiment also provides a seed routing computing entity. As shown in FIG. 4, the sub-routing computing entity includes: a receiving module 40, configured to receive a second request message from the parent routing computing entity PCE, where the The second request message is that the parent PCE sends the first request message to trigger the sending of the parent PCE after receiving the first request message; the reporting module 42 is configured to determine, according to the received second request message, The parent PCE reports the cross-domain topology information. Preferably, the reporting module 42 determines the cross-domain topology information by: selecting an available inter-domain link from the inter-domain link of the specified domain and the adjacent domain according to the selection policy, and selecting the selected domain. The network element where the endpoints of the respective specified domains are located is determined to be a cross-domain topology network element, where the selection policy is carried by the end-to-end routing request information carried in the second request message, and is used for Selecting a route; and determining the intra-domain route and the cross-domain intra-domain link according to the determined cross-domain topology network element. In a preferred implementation of the present invention, the reporting module 42 determines the intra-domain route and the inter-domain intra-domain link according to the determined cross-domain topology network element by using one of the following processes: Any two of the cross-domain topological network elements, if the routing of the two cross-domain topological network elements indicated by the local policy of the specified domain can only be used or must be preferably used at both ends of the link respectively in the two said cross The link of the domain topology network element searches for the link that meets the selection policy in the link between the two cross-domain topology network elements as the intra-domain intra-domain link; Any two of the cross-domain topology network elements in the specified domain, if the local policy requires cross-domain routing through the two cross-domain topology network elements, the two ends of the link must be preferably used in the two cross-domains respectively. The link of the topology network element, but there is no link that meets the selection policy, or the local policy does not require the cross-domain routing of the two cross-domain topology network elements to be used only or preferably Two places Calculating the intra-domain route of the cross-domain topology network element that meets the selection policy, and calculating the link of the cross-domain topology network element; In the intra-domain route, if the intra-domain route only passes through the one-hop link, the link is the inter-domain intra-domain link, and the intra-domain route of the specified domain is not reported. Otherwise, the calculated intra-domain route needs to be reported. In order to better understand the above embodiments, the parent PCE will be specifically described below in conjunction with specific embodiments and related drawings. The main design idea of the following embodiments is to solve the problem that the existing related cross-domain routing technology cannot effectively calculate the optimal route or can not support the specific requirement scenario, and propose a route calculation method: acquiring the intra-domain path according to the cross-domain end-to-end routing request, The inter-domain link and the intra-domain cross-border link (also called the inter-domain intra-domain link) form a network topology based on the obtained information, and then calculate the cross-domain end-to-end optimal route on the network topology. Embodiment 1 In order to calculate an optimal route in accordance with a routing request condition and a domain local policy, the topology of the entire network is composed of the following types of links: inter-domain links; virtual links corresponding to intra-domain optimal paths; intra-domain optimal paths Corresponding physical link; cross-domain intra-domain link planned by the domain local policy (the two endpoints of the link are on the same domain) In this embodiment, two types of routing computing entities, the parent PCE and the child PCE, are involved. The parent PCE is responsible for calculating cross-domain end-to-end routes based on the network topology of the entire network. The child PCE is responsible for calculating intra-domain routes based on the network topology of the domain it serves. Allow one child PCE to serve multiple domains, that is, calculate intra-domain routes for multiple domains. Considering that a large-scale network is divided into domains, the number of domains and the inter-domain links are much smaller than the intra-domain links. Therefore, in this embodiment, the parent PCE needs to obtain the cross-domain end-to-end routing request after the parent PCE obtains the sub-PCE. The cross-domain topology information of the inter-domain link is obtained instead of the pre-inter-domain link information of the parent PCE. This enables the network topology obtained for cross-domain routing calculation to reflect the actual status of the current network topology in real time and improve as much as possible. The accuracy of the route calculation. As shown in FIG. 5, the specific implementation of this embodiment is as follows: Step S502: Configure, to the parent PCE, all domain information that may be used for cross-domain routing calculation and sub-PCE information associated with each domain. The configured domain information includes a domain identifier, and the configured sub-PCE information includes a sub-PCE identifier and a sub-PCE communication address. Step S504: After receiving the cross-domain end-to-end path calculation request, the parent PCE requests all the cross-domains to be involved. The sub-PCE of the path calculation provides a cross-domain topology to the parent PCE, wherein the cross-domain topology request includes cross-domain end-to-end path request information and a specified domain. The end-to-end path request information includes attribute information of the route: bandwidth, source node address, destination node address, constraint routing information, and the like. If there is routing constraint information, the parent PCE will send the "cross-domain topology request" to the end-to-end routing request based on the constraint routing information to allow the sub-PCE of the domain to pass. Step S506: After receiving the cross-domain topology request, the sub-PCE determines the cross-domain topology information of the specified domain and reports the information to the parent PCE. Cross-domain topology information includes: inter-domain links, inter-domain intra-domain routes, and inter-domain intra-domain links. The link information described herein includes, in addition to the link identification information, attribute information used for calculating the connection-oriented optimal path: link cost information, bandwidth information, and the like. The specific method steps for the sub-PCE to determine the cross-domain topology information are as follows:

(1) Determine inter-domain links and cross-domain topology NEs. Select the inter-domain link that meets the cross-domain end-to-end routing request as the inter-domain inter-domain link that can be used for inter-domain end-to-end routing from the inter-domain link between the specified domain and the adjacent domain. The NE where the endpoint of the link is located in the specified domain is the cross-domain topology NE. If the source NE or the destination NE of the inter-domain end-to-end route is in the specified domain, the source NE or the destination NE is also a cross-domain topology NE.

(2) Determine cross-domain intra-domain links and cross-domain intra-domain routes. The cross-domain intra-domain link and the cross-domain intra-domain path are further determined according to the cross-domain topology network element determined in the previous step. You can use the following two steps to determine the inter-domain intra-domain link and the inter-domain intra-domain path: If the local policy requires cross-domain routing through the two top-level NEs, you can only use or must use the link. The network element where the endpoint is located is the link of the two topological network elements. The network element where the two ends of the link are located finds the link that meets the end-to-end cross-domain routing requirement for the links of the two topological network elements. The link is the inter-domain link to be reported to the parent PCE by the sub-PCE. If the cross-domain intra-domain link usage policy requires cross-domain routing through the two top-level NEs, the network element where the two ends of the link are located must be the two. The link of the topology NE, but the link that meets the cross-domain routing request does not exist, or the local policy does not require the cross-domain routing of the two top-level NEs. The links of the two topological network elements calculate the optimal intra-domain routes that pass through the two topological network elements and satisfy the end-to-end cross-domain routing request. If the intra-domain path only passes through the one-hop link, the link is the intra-domain link to be reported to the parent PCE, and the intra-domain routing information is no longer required to be reported. Otherwise, the routing information in the domain needs to be reported to the parent PCE. The intra-domain routing information reported to the parent PCE includes: the node, the link, and the intra-domain routing cost of the intra-domain route. Further, the intra-domain path information reported by the Bayu PCE may further include bandwidth information determined by the bandwidth of all links through which the path within the domain passes, wherein the "idle available bandwidth value" may take the minimum value of all links "free available bandwidth". Step S508: After receiving the cross-domain topology response of the sub-PCE, the parent PCE sorts and adds the cross-domain topology in the response to the entire network topology data. If there is a cross-domain intra-domain route, it needs to be mapped to a logical chain. road. Reflect If the intra-area path information is a logical link, if the intra-domain routing information reported by the sub-PCE provides bandwidth information, the bandwidth information of the logical link is determined according to the bandwidth information, otherwise the bandwidth value is determined by the cross-domain end-to-end routing request bandwidth attribute value. It is determined that the idle available bandwidth of the link is equal to the committed bandwidth value of the cross-domain end-to-end routing request. Step S510: After the parent PCE processes the cross-domain topology response of all the sub-PCEs, calculate the cross-domain end-to-end optimal route that satisfies the cross-domain end-to-end routing request (ie, the selection policy) according to the obtained network-wide topology. In the related art, the hierarchical PCE route calculation method requires computing the intra-domain path and mapping the intra-domain path to a logical link. Therefore, the method does not include support for the following important demand scenarios: 1) The calculated intra-domain path only passes one hop. Link, which is a physical link or a link mapped by an existing service layer path, so there is no need to logically link the intra-domain path. 2) According to the network plan, for a domain, the intra-domain link that crosses the boundary of the domain (the network element where the two endpoints of the link are located is the domain boundary node) is planned and configured in advance, and the link is a physical link or A link mapped by an existing service layer path. The domain local policy requires that the end-to-end path can only use or preferentially use these cross-domain links. The routes calculated by the related technologies cannot meet the local policy requirements of these domains. The above technical problem can be obviously solved by adopting the above-mentioned intra-domain routing and intra-domain path determination manner in this embodiment. Embodiment 2 This embodiment relates to a connection-oriented network, including a connection-oriented packet switching network, Multi-Protocol Label Switching (MPLS), such as an MPLS/MPLS_TP (Transport Profile) network, and a synchronous digital system. (Synchronous Digital Hierarchy, abbreviated as SDH) Network, Optical Transport Network (OTN), etc., is a technology for calculating cross-domain end-to-end optimal routing in a connection-oriented multi-domain network. In this embodiment, as shown in FIG. 6, the parent PCE and the child PCE are running on different servers, wherein the parent PCE has configured {Domainl, child PCE1 }, {Domain2, child PCE2}, {Domain3, child PCE3}. The domain information and the sub-PCE information corresponding to {Domain4, sub-PCE4}, {Domain5, and sub-PCE5} respectively, the domain information includes the domain identifier, and the sub-PCE information includes the sub-PCE identifier and the corresponding communication address. In this embodiment, as shown in FIG. 6, the MPLS network is divided into five domains, and the configuration of the inter-domain links is as shown in the figure. The inter-domain cross-domain link Link3_2 is configured in the domain. The inter-domain link is the link of the existing MPLS tunnel mapping where the endpoint is located on the border network element, and the sub-domain of the management domain is configured with the cross-domain intra-domain link. Local policy: Cross-domain paths through border NEs D and F are only allowed to use Link3_2. In this embodiment, the key information "free available bandwidth" of the inter-domain link and the inter-domain intra-domain link is as shown in Table 1: Table 1

The link bandwidth information in Table 1 is known by the sub-PCE of the relevant domain. The idle bandwidth can be different in both directions of the link. For this embodiment, the idle bandwidth values in the two directions of the link are assumed to be the same. Assume that the parent PCE receives a unidirectional LSP route calculation request. The key information of the route request is: source NE A, destination NE, and committed bandwidth is 10M. Similar to the embodiment shown in Figure 5, you can complete the cross as follows. Domain routing calculation: Step 1. After receiving the cross-domain path calculation request, the parent PCE issues the following five requests according to the domain and related sub-PCE configuration information: Request sub-PCE1 provides end-to-end cross-domain routing. Request sub-PCE2 provides end-to-end cross-domain routing of the cross-domain topology of the domain; request sub-PCE3 provides end-to-end routing cross-domain routing on the cross-domain topology of the domain; request sub-PCE4 provides end-to-end cross-domain routing on Dom _a in4 Cross-domain topology; request sub-PCE5 provides end-to-end cross-domain routing of cross-domain topology for domains; this request contains information contained in end-to-end cross-domain routing requests. Step 2: After receiving the cross-domain topology request for the domain Domain1, the sub-PCE1 commits the bandwidth according to the LSP.

10M cross-domain end-to-end routing attributes and other requirements, the following process steps are processed (see Figure 7): (1) Determine the inter-domain links that can be used for inter-domain routes as Link1, Link2, and Link3. The cross-domain topology NEs that may be used are A, B, and C.

(2) Calculate the intra-domain optimal route or cross-domain intra-domain link that satisfies the cross-domain routing requirement between any two network elements for the cross-domain topology network element "A, B, C" determined in the above steps. Assume that the calculated intra-domain routes are Pathl_l, Pathl_2, and Pathl_3, and assuming that the number of hops Pathl_3 passes is 1, it is determined that the corresponding "cross-domain intra-domain link" is Link1_3.

(3) The sub-PCE1 reports the inter-domain topology information of the inter-domain link: Link1, Link2, and Link3, and the inter-domain intra-domain route: Pathl_l, Pathl_2, and the intra-domain link: Linkl_3 to the parent PCE. After receiving the cross-domain topology request for the domain domain, the sub-PCE2 performs the following process steps according to the requirements of the inter-domain end-to-end route attribute whose LSP promises bandwidth is 10M (see Figure 7):

(1) Determining the inter-domain links that can be used for inter-domain routes are Link2, Link3, Link5, Link6, and Link7. The cross-domain topology NEs that may be used are G, H, I, and N.

(2) For the cross-domain topology network element "G, H, I, N" determined in the above steps, calculate the intra-domain optimal route or the inter-domain intra-domain link that satisfies the cross-domain routing requirement between any two network elements. It is assumed that the calculated intra-domain routes are Path2_l, Path2_2, and Path2_3, and it is assumed that the hop count of Path2_3 is 1, thereby determining the "cross-domain intra-domain link" Link2_3.

(3) Sub-PCE2 reports the inter-domain topology information of the inter-domain link: Link2, Link3 Link5, Link6, Link7, Inter-domain intra-domain routing: Path2_l, Path2_2, and Inter-domain intra-domain link: Link2_3 The parent PCE; after receiving the cross-domain topology request for the domain domain, the sub-PCE3 performs the following process steps according to the requirements of the inter-domain end-to-end route attribute whose LSP promises bandwidth is 10M (see Figure 7):

(1) Determining the inter-domain links that can be used for inter-domain routes are Link1, Link4, and Link5. The cross-domain network elements that may be used are D, F, and E.

(2) For the cross-domain topology network element "D, F, E" determined in the above steps, calculate the intra-domain optimal route or the inter-domain intra-domain link that satisfies the cross-domain routing requirement between any two network elements. According to the intra-domain intra-domain link usage policy configured for the domain, the cross-domain path through the boundary network elements D and F is only allowed to use Link3_2 to determine the cross-domain intra-domain link Link3_2. Between "D, E" and "3⁄4" respectively. The intra-domain routes Path3_l and Path3_3 that meet the cross-domain routing requirements are calculated between F and F. Path3_l has a hop count of 1, thus determining the cross-domain intra-domain link Link3. (3) Sub-PCE3 reports the inter-domain topology information of the inter-domain links: Link1, Link4, Link5, Inter-domain intra-domain routing: Path3_3, and Inter-domain intra-domain links: Link3_l, Link3_2 to the parent PCE. The detailed process of the sub-PCE processing the cross-domain topology request is described in the above steps. Therefore, the processing after the sub-domain topology request is received by the sub-PCE4 and the sub-PCE5 is not described in detail. Only the sub-PCE4 and the sub-PCE5 are cross-domain. The processing result of the topology request (see Figure 7): The sub-PCE4 reports the inter-domain topology information of the inter-domain links: Link4, Link6, and Link7, and the inter-domain intra-domain routes: Path4_l, Path4_2, and Path4_3 to the parent PCE. The sub-PCE does not find the topology component that satisfies the cross-domain routing request. Therefore, the cross-domain topology information reported by the sub-PCE5 to the parent PCE is empty. Step 3: After receiving the sub-domain topology request response, the parent PCE adds the reported cross-domain topology information to the entire network topology, where the reported intra-domain route is mapped to a logical link. After the parent PCE processes all the sub-PCE cross-domain topology request responses, the network topology of the entire network for calculating the cross-domain routing shown in FIG. 8 is formed, and then the cross-domain is calculated according to the network topology and using the constraint path shortest priority algorithm CSPF. End-to-end optimal routing. "Linkl_l, Linkl_2 Link2_l Link2_2, Link3_3 Link4_l, Link4_2,

Link4_3 is the logical link that the parent PCE is mapped by the inter-domain intra-domain route. In the technical solution of the foregoing embodiment, the parent PCE uses the CSPF because the intra-domain path in the cross-domain topology information reported by the sub-PCE is also the optimal path. The cross-domain end-to-end route calculated by the algorithm based on the whole network topology formed by the cross-domain topology information of all sub-PCEs must also be the optimal route, and the effect is equivalent to the optimal cross-domain end-to-end calculated on one topology of the whole network without dividing the domain. Because the parent PCE does not need to maintain the entire network topology in real time, all the topology information required by the parent PCE is obtained in real-time for the cross-domain end-to-end route calculation request, and the parent PCE or the child PCE can be used in addition to the cross-domain topology information processing. The existing route that satisfies the condition is calculated by using the traditional Constrained Shortest Path First (CSPF). The route calculation method in the above embodiment is relatively simple, and the accuracy of the route calculation is improved as much as possible. In a large-scale network. In another embodiment, a software is also provided, which is used to execute the above The technical solutions described in the embodiments and the preferred embodiments. In another embodiment, a storage medium is provided, the software being stored, including but not limited to: an optical disk, a floppy disk, a hard disk, a rewritable memory, and the like. Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computing device, which can be concentrated on a single computing device or distributed over a network composed of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device, such that they may be stored in the storage device by the computing device and, in some cases, may be different from the order herein. The steps shown or described are performed, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps are fabricated as a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software. The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and scope of the present invention are intended to be included within the scope of the present invention.

Claims

Claim

A method for obtaining an inter-domain end-to-end route, including:

The parent route calculation entity PCE receives the first request message;

The parent PCE sends a second request message to the sub PCE of the parent PCE participating in the cross-domain routing calculation, triggered by the first request message;

The parent PCE receives cross-domain topology information that is determined and reported by the sub-PCE according to the second request message;

The parent PCE calculates the cross-domain end-to-end optimal route according to the received cross-domain topology information.

2. The method according to claim 1, wherein the second request message carries end-to-end routing request information and a specified domain identifier, where the end-to-end routing request information carries the sub-PCE Select a routing policy.

The method according to claim 2, wherein the first request message carries the selection policy; the selection policy carried in the second request message is obtained by:

The parent PCE obtains the selection policy from the first request message and carries it in the second request message.

The method according to claim 1, wherein the cross-domain topology information comprises at least one of the following: an intra-domain route, an inter-domain link, and an inter-domain intra-domain link, where the cross-domain intra-domain link The two endpoints are respectively on the two boundary network elements of the domain.

The method according to claim 4, wherein the determining, by the sub-PCE, the cross-domain topology information according to the received second request message includes:

The sub-PCE selects an available inter-domain link from the inter-domain link of the specified domain and the adjacent domain according to the selection policy, and locates the endpoints of the selected inter-domain link and located in each of the designated domains. The network element is determined to be a cross-domain topology network element;

The sub-PCE determines the intra-domain route and the intra-domain intra-domain link according to the determined cross-domain topology network element.

The method according to claim 5, wherein the sub-PCE determines the intra-domain route and the inter-domain intra-domain link according to the determined cross-domain topology network element, and includes at least one of the following processes: For any two of the cross-domain topological network elements in the specified domain, if the routes of the two cross-domain topological network elements indicated by the local policy of the specified domain can only be used or must be used, respectively In the link between the two cross-domain network elements, the links at the two ends of the cross-domain network element are searched for the link that meets the selection policy, as the cross-domain Intra-domain link;

For any two of the cross-domain topology network elements in the specified domain, if the local policy requires cross-domain routing through the two cross-domain topology network elements, it is preferable to use the two ends of the link respectively in the two A link of a cross-domain topology network element, but there is no link that meets the selection policy, or the local policy does not require cross-domain routing through the two cross-domain topology network elements. In the two links of the cross-domain topology network element, the intra-domain route that meets the selection policy of the cross-domain topology network element is calculated; in the calculated intra-domain route, if the intra-domain route passes through only one If the link is hopped, the link is the intra-domain link, and the intra-domain route of the specified domain is not reported. Otherwise, the calculated intra-domain route is reported to the parent PCE.

The method according to claim 6, wherein the intra-domain route reported to the parent PCE includes the following information: a network element, a link, and an intra-domain routing cost that the intra-domain route passes.

8. The method of claim 7, wherein the intra-domain route reported to the parent PCE further comprises the following information: bandwidth information.

9. The method according to claim 8, wherein the bandwidth information is determined by a bandwidth of all links through which an intra-domain route passes, wherein an idle available bandwidth value of the bandwidth information takes a minimum value of available idle bandwidth of all links. .

10. The method of claim 4 or 8, further comprising: the parent PCE mapping the intra-domain route to a logical link.

11. The method according to claim 10, wherein the bandwidth of the logical link is determined by: when bandwidth information is included in an intra-domain route reported to a parent PCE, the bandwidth of the logical link is by the bandwidth The information is determined; otherwise, determined by the bandwidth attribute value of the end-to-end routing request information carried in the second request message.

The method according to any one of claims 1 to 9, wherein before the parent PCE receives the first request message, the method further includes:

The domain information of each domain required for calculating the cross-domain end-to-end optimal route and the sub-PCE information associated with each domain are configured for the parent PCE.

The method according to any one of claims 1 to 9, wherein the parent PCE calculates the cross-domain end-to-end optimal route according to the received cross-domain topology information, including:

The parent PCE adds the received cross-domain topology information to the entire network topology data; the parent PCE calculates the cross-domain end-to-end optimality according to the network-wide topology data after adding the cross-domain topology information. routing.

An apparatus for obtaining an inter-domain end-to-end route, which is located in a parent routing computing entity PCE, and includes: a first receiving module, configured to receive a first request message;

a sending module, configured to send, by the triggering of the first request message, a second request message to a sub-PCE of the parent PCE participating in cross-domain routing calculation;

a second receiving module, configured to receive cross-domain topology information that is determined and reported by the sub-PCE according to the second request message;

The calculation module is configured to calculate an inter-domain end-to-end optimal route according to the received cross-domain topology information.

The device according to claim 14, wherein the second request message carries end-to-end routing request information and a specified domain identifier, wherein the end-to-end routing request information carries the sub-PCE selection Routing selection strategy.

The device according to claim 14, wherein the cross-domain topology information comprises at least one of the following: an intra-domain route, an inter-domain link, and an inter-domain intra-domain link, where the cross-domain intra-domain link The two endpoints are respectively on the two boundary network elements of the domain.

The device according to any one of claims 14 to 16, wherein the calculation module comprises:

An adding unit, configured to add the cross-domain topology information received by the receiving module to the entire network topology data;

The calculating unit is configured to calculate the cross-domain end-to-end optimal route according to the network-wide topology data after adding the cross-domain topology information.

18. A sub-routing computing entity, comprising:

a receiving module, configured to receive a second request message from the parent routing computing entity PCE, wherein the second request message is triggered by the first request message after the parent PCE receives the first request message Send by the parent PCE; The reporting module is configured to determine, according to the received second request message, the cross-domain topology information to be reported to the parent PCE.

19. The sub-routing computing entity of claim 18, wherein the reporting module determines the cross-domain topology information by:

Selecting an available inter-domain link from the inter-domain link of the specified domain and the adjacent domain according to the selection policy, and determining the network element where the endpoints of the selected inter-domain link and each of the specified domains are located as the cross-domain a topology network element, where the selection policy is carried by the end-to-end routing request information carried in the second request message, and is used to select a route;

Determining the intra-domain route and the inter-domain intra-domain link according to the determined cross-domain topology network element.

20. The sub-routing computing entity according to claim 19, wherein the reporting module determines the intra-domain routing and the inter-domain intra-domain link according to the determined cross-domain topology network element by using one of the following processing procedures: For any two of the cross-domain topological network elements in the specified domain, if the routes of the two cross-domain topological network elements indicated by the local policy of the specified domain can only be used or must be used, respectively In the link between the two cross-domain network elements, the links at the two ends of the cross-domain network element are searched for the link that meets the selection policy, as the cross-domain Intra-domain link;