WO2009092252A1 - 一种路由计算方法、系统及路径计算单元 - Google Patents

一种路由计算方法、系统及路径计算单元 Download PDF

Info

Publication number
WO2009092252A1
WO2009092252A1 PCT/CN2008/073654 CN2008073654W WO2009092252A1 WO 2009092252 A1 WO2009092252 A1 WO 2009092252A1 CN 2008073654 W CN2008073654 W CN 2008073654W WO 2009092252 A1 WO2009092252 A1 WO 2009092252A1
Authority
WO
WIPO (PCT)
Prior art keywords
path
pce
node
lower layer
path segment
Prior art date
Application number
PCT/CN2008/073654
Other languages
English (en)
French (fr)
Inventor
Jixiong Dong
Fatai Zhang
Original Assignee
Huawei Technologies Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co., Ltd. filed Critical Huawei Technologies Co., Ltd.
Priority to EP08871309A priority Critical patent/EP2237501B1/en
Priority to JP2010540006A priority patent/JP5189171B2/ja
Publication of WO2009092252A1 publication Critical patent/WO2009092252A1/zh
Priority to US12/824,690 priority patent/US8737394B2/en

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/42—Centralised routing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/02—Topology update or discovery
    • H04L45/04—Interdomain routing, e.g. hierarchical routing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing

Definitions

  • the present invention relates to the field of communications, and more particularly to a route calculation method, system, and path computation unit. Background technique
  • a communication network is a collection of geographically distributed nodes and communication links between these nodes for transmitting data. Nodes that are not adjacent in the communication network can communicate with each other through intermediate nodes, which can make the communication network resources more economical and more efficient.
  • the nodes of the communication network typically communicate over the network by exchanging data frames or packets, which are specified by a particular protocol, such as the TCP/IP protocol.
  • the protocol referred to in this article refers to a collection of rules that define how nodes interact with each other, similar to the TCP/IP protocol.
  • the communication network becomes so large, it is difficult to manage and maintain. Therefore, it is generally divided into a plurality of routing domains or an autonomous system (AS) for management.
  • Networks in autonomous systems are typically coupled together by traditional "intra-domain" routers that implement intra-domain routing protocols and are managed by common powers.
  • the autonomous system is generally divided into multiple areas.
  • a domain refers to any set of network elements within a common address management scope or path computation responsibility. Therefore, examples of domains may be zones, autonomous systems, and multiple autonomous systems.
  • the unified domain of the routing domain, the autonomous system and the zone is called a domain, and its specific meaning can be determined according to the context.
  • the inter-domain routers that implement the inter-domain routing protocol are used to interconnect nodes of different domains. This inter-domain router is also called a border router.
  • An example of an inter-domain routing protocol is Border Gateway Protocol (BGP), which performs inter-domain routing by exchanging routing and reachability information between adjacent inter-domain routers in the system.
  • BGP typically uses a reliable transport protocol, such as TCP, to establish connections and sessions.
  • Intra-domain routing protocol it can also be said that an example of the Internal Gateway Protocol (IGP) is the Open Shortest Path First (OSPF) routing protocol.
  • the OSPF protocol is based on link state technology and is therefore a link state routing protocol.
  • the link state routing protocol defines the routing information and the network topology information exchange and processing. For example, the information in OSPF is exchanged through the Link State Advertisement (LSA).
  • LSA Link State Advertisement
  • MPLS Multi-Protocol Label Switching
  • LSR Label Switched Routers
  • LSP Label Switch Paths
  • Multi-protocol label switching technology has also been introduced into the field of optical transport networks to develop Automatically Switched Optical Network (ASON).
  • ASON Automatically Switched Optical Network
  • the network connection service is provided by manual configuration or semi-automatic configuration of the network management.
  • the automatic switched optical network provides network connection services through the automatic establishment of the control plane.
  • the automatic switched optical network can be divided into a transport plane that truly carries network services, a management plane that performs management functions, and a control plane that runs control protocols.
  • the technology used in the control plane of the automatic switched optical network is called Generalized Multi-Protocol Label Switching. It is an extension of the multi-protocol label switching technology, which includes the Link Management Protocol. Routing protocol and signaling protocol.
  • the link management protocol obtains information such as the type of connection and the number of resources supported by the link through packet exchange based on the discovery of the adjacency relationship.
  • the information is called Traffic Engineering ( ⁇ ) information, and these include traffic engineering information.
  • a link is called a traffic engineering link.
  • local link traffic engineering information is advertised to other nodes in the domain through a routing protocol, such as the Traffic Engineering Extended Open Shortest Path First (OSPF-TE) protocol.
  • OSPF-TE Traffic Engineering Extended Open Shortest Path First
  • the connected network ingress node can perform path computation, obtain the link sequence that the connection needs to pass, and then pass the signaling protocol.
  • the Traffic Engineering Extended Resource Reservation Protocol (RSVP-TE) requests the nodes on the path to allocate resources and establish cross-connections to establish an end-to-end connection.
  • RSVP-TE Traffic Engineering Extended Resource Reservation Protocol
  • each node has only the traffic engineering information of the local domain and the reachability information of other domains, which reduces the impact of network topology changes on the new service and recovery congestion.
  • the scalability of the network After the domain is divided, the nodes in each domain only have the traffic engineering information of the domain and the reachability information of other domains, and there is no complete traffic engineering information of other domains. Therefore, how to calculate an end-to-end bandwidth in the case of multiple domains Paths such as switching capabilities, routing separation, protection constraints, and user policies have become a problem to be solved.
  • the domain-domain routing protocol (DDRP) technology adopts a hierarchical network model, and one domain of the lower layer is used in the upper layer.
  • the proxy node indicates that the proxy node can advertise the abstract topology representing the domain, the inter-domain link, the reachable address, etc., so as to form a layer-by-layer network.
  • each path calculation unit saves all traffic engineering in the domain it is responsible for.
  • Information (for the sake of the package, the network topology information is also included in the traffic engineering information).
  • the node that needs to calculate the path is generally called the Path Computation Client (PCC), and sends a request containing the routing calculation parameter information to the path calculation unit, which is based on its own stored traffic engineering database (Traffic Engineering Database). , TED ) for routing Calculate and feed back the result of the calculation to the requesting node.
  • the path calculation unit can simultaneously store traffic engineering information of one or more domains. When calculating a route across multiple domains, if it exceeds the scope of the path calculation unit, it will use the path calculation unit communication protocol.
  • the solution of the prior art 1 is that its route calculation is a serial process.
  • the first node calculates the route, only the entry and exit information of the part of the domain through which the path passes, because there is no actual traffic engineering information in the relevant domain, only the letter If the domain boundary calculation is triggered when the corresponding domain boundary is reached, it can be judged whether the path from the domain entrance to the exit can be successfully calculated. It is easy to find that the signaling goes to the middle and finds that there is no way to go or does not satisfy the routing constraint, resulting in signaling construction. When the rollback is repeated multiple times, the previously established cross-connection is deleted and reconstructed; it is also difficult to calculate the end-to-end separate route (the different paths of the same-same and the same) when applying the technical solution.
  • the prior art 2 is a flat single-layer model, that is to say, all path calculation units have the same status, so it becomes difficult to manage when the network is complicated or large; and the actual network is not implemented in a layered manner. Abstraction, when calculating cross-domain routing, it completely relies on the exchange of traffic engineering information of each domain between path computing units. When there are many domains in the service, the communication between these related path computing units becomes very frequent. And the amount of information exchanged is large, which reduces the efficiency and reliability of route calculation.
  • the embodiment of the invention provides a route calculation method, a system and a path calculation unit, which can efficiently perform cross-domain end-to-end path calculation.
  • An embodiment of the present invention provides a route calculation method, where the method includes:
  • the first underlying path calculation unit PCE receives the path calculation command, performs route calculation on the managed routing domain, obtains a path segment set with the first node as an endpoint, and sends the obtained path segment set to the upper layer PCE;
  • the upper layer PCE that receives the path segment set sent by the lower layer PCE sends a path calculation command to another lower layer PCE. After receiving the path segment set sent by all the lower layer PCEs, all the received path segment sets are combined and sent. Giving the upper layer PCE of the upper layer PCE; until the top layer PCE receives all the path segment sets sent by the lower layer PCE;
  • the top-level PCE combines all received sets of path segments to generate a set of paths between the first node and the second node.
  • the embodiment of the invention further provides a path calculation unit, including:
  • a path calculation command receiving module configured to receive a path calculation command
  • the path segment information receiving module is configured to send a path calculation command to the lower layer path calculation unit one by one according to the routing domain order to be established after the path calculation command receiving module receives the path calculation command, specifically: After the path segment set returned by the lower layer path calculation unit, the path calculation command is sent to another lower layer path calculation unit until all the path segment sets returned by the lower layer path calculation unit are received;
  • a path merging module configured to combine the plurality of path segment sets received by the path segment information receiving module, and send the merged path segment set;
  • the path segment set sending module is configured to send the received combined path segment set to the upper layer path computing unit.
  • the embodiment of the invention further provides a route calculation method, the method comprising: receiving a path calculation command;
  • the path calculation command is sent to the lower layer PCE one by one. After receiving the path segment set returned by the lower layer PCE, the path calculation command is sent to another lower layer PCE until all the packets are received. The set of path segments returned by the lower layer PCE;
  • the received set of the combined path segments is sent to the upper layer PCE.
  • the embodiment of the invention further provides a route calculation system, the system comprising:
  • a top-level PCE configured to: after receiving the path calculation request, calculate information about the first node and the second node that are carried by the request according to the path, and calculate between the first node and the second node The routing domain to be followed by the path to be established, and the path calculation command is issued to the first bottom PCE directly or layer by layer;
  • the first underlying PCE is configured to perform routing calculation on the routing domain managed by the path after receiving the path calculation command, obtain a path segment set with the first node as an endpoint, and send the obtained path segment set to the upper layer PCE. ;
  • An upper layer PCE that receives the path segment set sent by the lower layer PCE, configured to send a path calculation command to another lower layer PCE according to a routing domain that the path to be established between the first node and the second node is to be established, After all the path segment sets sent by the lower layer PCE are combined, the received path segment sets are combined and sent to the upper layer PCE of the upper layer PCE; until the top layer PCE receives all the path segment sets sent by the lower layer PCE;
  • the top-level PCE combines all received path segment sets to generate a set of paths between the first node and the second node.
  • FIG. 1 is a schematic diagram of a network structure of a cross-domain computing path in the prior art
  • FIG. 2 is a schematic diagram of a network structure of a route calculation method according to an embodiment of the present invention
  • FIG. 3 is a structural diagram of a path calculation unit according to an embodiment of the present invention. detailed description
  • the embodiment of the route calculation method of the present invention can be applied to the network structure shown in FIG. 2, as shown in FIG. 2:
  • the network is divided into multiple layers.
  • each underlying routing domain is composed of one or more PCEs. Directly responsible for the intra-domain routing topology of the domain and its route calculation; then, the underlying multiple routing domains are directly responsible for the higher-level PCE, which has these underlying inter-domain routing topologies, directly managed routes. Intradomain routing of the domain, and Policy-based intra-domain routing topology for some domains. According to this idea, based on the entire network
  • PCE performs hierarchical abstract management. There is no limit to the number of hierarchical layers of the network, which can be divided based on network size or strategy. All routing topologies (collections of routing topologies that the underlying PCE is responsible for) that each PCE is responsible for may be referred to as its computing domain.
  • each underlying routing domain is responsible for the routing topology of the domain and its route calculation by one or more PCEs, and then, the PCE of the plurality of routing domains at the bottom layer.
  • the higher-level PCE has these underlying inter-domain routing topologies, as well as intra-domain routing topologies for some domains based on policies.
  • hierarchical management of the entire network based on PCE is carried out. There is no limit to the number of hierarchical layers of PCE, which can be divided based on network size or strategy.
  • the lowest level 0 levelO consists of multiple routing domains: Domain 1, Domain 2, Domain 3, Domain 4. Of course, there are several routing domains besides Domain 1, Domain 2, Domain 3, and Domain 4. The convenience of introduction is no longer described.
  • Each routing domain in the underlying layer has a direct corresponding PCE responsible for routing topology and routing calculations for the domain. These PCEs at levelO are: PCE01, PCE02, PCE03, PCE04.
  • the topology of the underlying multiple PCEs is abstracted to form the top level of the levell, which is responsible for the PCE of the levell layer. As shown in the figure, the corresponding domains Domain 1 and Domain 2 responsible for PCE01 and PCE02 are abstracted so that they are in the same computing domain.
  • PCE11 The abstract topology of these domains that PCE11 is responsible for contains inter-domain link information for these domains.
  • the abstract method of PCE12 is similar to that of PCE11 and will not be described in detail here.
  • the topography of the Levell layer can be abstracted to a higher level.
  • PCE11, PCE12, ...PCE1X of the Leverl layer, and the computational domain of the routing topology in charge are further abstracted and layered, and these routing topologies are responsible for the PCE21 of the Level 2 layer.
  • the routing topology of PCE21 includes the interconnection information between PCE11 and PCE12. In other cases, based on the policy, PCE21 can also contain internal routing information of PCE11 and PCE12.
  • PCE11, PCE12, and PCE21 are not the underlying PCE, but may also have a directly managed routing domain.
  • the routing domains directly managed by PCE11, PCE12, and PCE21 are located at levell, levell, and levevl2, respectively, by PCE11, PCE12, PCE21 directly calculates the intra-domain route and the route between the routing domain and the computing domain of other peer PCEs. Since the calculation body of the route in the processing domain is not the underlying PCE, the route calculation method in this case is basically the same as the embodiment of the present invention, and therefore will not be separately described below.
  • This layer and the different layers of PCEs spread the corresponding routing topology information through routing protocols (for example, based on IGP or BGP protocols). Information about the capabilities and location of the PCE through the PCE discovery protocol between this layer and the different layers of PCE.
  • the PCE that receives the request judges according to the source and destination address information carried in the request message, if the requested destination node is not in the same source node as the source node.
  • the domain PCE finds a common upper-layer PCE (recorded as a top-level PCE) according to the capability information of each level of the PCE.
  • the computing domain of the upper-layer PCE includes both the source node and the destination node, and thus the destination is completed. The path calculation of the node.
  • the path request message of the source node received by the top-level PCE may carry the domain sequence information to be passed. If the source node's path calculation request carries the domain sequence information to be passed, the top-level PCE completes the corresponding path calculation according to the domain sequence. Otherwise, the top-level PCE determines the domain sequence of the entire cross-domain path based on the abstract inter-domain topology information it has and based on a certain strategy (eg, the shortest sequence).
  • a certain strategy eg, the shortest sequence
  • the top-level PCE initiates a path computation request to the lower PCE (recorded as the right side of the top-level PCE) near the destination routing domain. This process iterates until the path computation request is sent to the lowest-level PCE in the domain of the destination node.
  • the lowest-level PCE calculates the corresponding optimal path segment set in this domain and returns it to the upper-layer PCE.
  • the corresponding upper layer PCE After receiving the path calculation request, the corresponding upper layer PCE sends a path calculation request to other PCEs under the layer PCE according to the returned path segment set information, and the process iterates until the path calculation request is sent to the domain exit of the path segment.
  • the upper PCE returns the calculated set of path segments to its upper PCE until it returns to the top PCE.
  • the top-level PCE initiates a path computation request based on the collected top-level PCE right-direction path segment set and then to the lower-level PCE on the left side of the top-level PCE. This process is similar to processes 3, 4, 5, and 6.
  • the top-level PCE receives the set of path segments returned by the lower-layer PCE. According to the set of path segments, the end-to-end cross-domain optimal path can be calculated, and then the optimal path is returned to the underlying PCE where the first node is located.
  • the PCE at the bottom node of the first node After receiving the path calculation request response from the top PCE, the PCE at the bottom node of the first node returns the corresponding path information to the first node of the path calculation request.
  • the first node may initiate signaling to establish a corresponding path.
  • the source node is N10
  • the requested destination node is N40.
  • the method of this embodiment is divided into two sub-flows for explanation: 1) determining the top-level PCE; 2) calculating the path.
  • the specific process is as follows:
  • Sl l and N10 send a path calculation request message to PCE01, and the message carries the address information of the active node N10 and the destination node N40 address information.
  • the PCE01 that directly manages the N10 receives the path calculation request sent by the N10, and the PCE01 stores the topology information of the calculation domain that it is responsible for.
  • the PCE01 judges the source and destination addresses in the path request message, and finds that N10 and N40 are not in the computing domain in which they are responsible, so the upper path calculates the request message.
  • the direct upper layer PCE corresponding to the PCE01 is the PCE11.
  • the PCE11 After receiving the reported path calculation request message, the PCE11 performs the same judgment as the PCE01, and the judgment result is that the N40 is not in the calculation domain in which it is responsible, so the reporting path is continued. Calculate the request message.
  • S14 and PCE21 receive the path calculation request message of >3 ⁇ 4 on PCE11, and perform the same judgment as PCE01 and PCE11, and the judgment result is that the source and destination nodes belong to the calculation domain of PCE21. Therefore PCE21 is the top PCE.
  • the top layer PCE is determined in a layer-by-layer manner.
  • PCE01 can also directly store computing power information of other PCEs.
  • PCE01 can directly determine that PCE21 is the top-level PCE and directly send the path technology request message to PCE21.
  • the PCE determines an optimal domain sequence.
  • the determining of the optimal domain sequence may be determined according to the network status of each domain or a specific policy of the configuration, and the domain through which the path should pass is determined, and is not described in detail herein.
  • the domain sequence is Domainl-Domain2-Domain3-Domain4.
  • Domain4 where N40 is located belongs to the computing domain of PCE12, so PCE21 issues a path calculation command to PCE12.
  • the path calculation command carries the optimal domain sequence, and the intra-domain route calculation of each domain is performed in the order of the sequence.
  • Domain4 belongs to the computing domain of PCE04, so PCE12 sends a path calculation command to PCE04.
  • the PCE04 After receiving the path calculation command, the PCE04 performs the intra-domain path calculation, and returns the optimal path segment set in the domain to the PCE12.
  • the information of the set of ingress nodes corresponding to the set of path segments in the domain calculated by the PCE04 is also sent to the PCE12, and the ingress nodes are also edge nodes of the domain 04, and the information of the set of the ingress nodes may be sent separately or from the path segment of the PCE12. Obtained in the collection of information.
  • the PCE12 receives the information returned by the PCE04, determines the egress node set of the Domain03 according to the ingress node set of each path segment in the return information, and sends the path node calculation command to the PCE03 as the destination node of the path calculation.
  • the PCE03 calculates the optimal path segment set in the Domain3 that arrives at the specified egress node set, and returns the calculation result to the PCE12.
  • the optimal path segment set of Domain03 and Domain04 is spliced, and the path segment set of the cross-domain Domain03 and Domain04 is calculated, and the calculation result is returned to the top PCE21.
  • splicing is performed according to the egress node information.
  • PCE21 sends a path calculation command to PCE11, and performs processing similar to steps S22-S27:
  • PCE11 determines the set of the exit nodes of Domain02 and sends it to PCE02 according to the set of ingress nodes of the path segment of the cross-domain Domain03 and Domain04 that PCE12 feeds back to PCE21.
  • PCE02 calculates the path segment set in the domain and feeds it back to PCE11.
  • PCE01 then calculates the path segment set in the domain and feeds it back to PCE11.
  • PCE11 splices the path segment set across Domain1 and Domain2.
  • the PCE21 calculates an optimal cross-domain path set from N10 to N40 according to the path segment set of the cross-domain Domain03 and Domain04 and the path segment set of the cross-domain DomainO1 and Domain02, and returns the path information to PCE01, and finally PCE01 receives The path calculation result is returned to N10.
  • PCE21 has a directly managed routing domain Domain5
  • PCE21 calculates the domain path segment set of Domain5, and then sets Domain5's path segment set, Domain03 and Domain04 path segment set, and cross-domain DomainOl and Domain02 paths.
  • the three sets of segments are aggregated to calculate the optimal cross-domain path from N10 to N40.
  • the non-LevelO PCE has a directly managed domain
  • the difference from the embodiment is that the intra-domain path segment is not the underlying PCE, and the rest is basically the same as the embodiment.
  • the PCE 12 in this embodiment further includes other directly managed domains
  • the PCE 12 directly calculates the intra-domain path segment set of the domain, and integrates the three after receiving the path segment sets of Domain03 and Domain04.
  • This embodiment is exemplified only by a three-layer structure, and those skilled in the art will appreciate that the present invention is equally applicable to the case of one layer, two layers or more.
  • the PCE 21 transmits the path calculation command to the PCE04 by layer-by-layer transmission, and may also directly transmit it to the PCE04 in other embodiments.
  • the ingress node set and the egress node set may not be calculated or transmitted, and in step S26, the PCE03 only performs route calculation based on the topology in the domain.
  • the calculation starts from Domain4, and finally calculates Domain1, but those skilled in the art should know that in other embodiments, the reverse order of this embodiment can also be adopted, and Domian1 is first calculated to calculate Domian4.
  • an embodiment of the present invention further provides a path calculation unit, including: a path calculation command receiving module 310, a path segment information receiving module 320, a path combining module 330, and a path segment set sending module 340.
  • the path calculation command receiving module 310 is configured to receive a path calculation command, and the path calculation command is directly sent by the top PCE, or the top PCE is transmitted layer by layer.
  • the path segment information receiving module 320 is configured to: after the path calculation command receiving module 310 receives the path calculation command, according to the routing domain to be sequentially followed by the path to be established, if the path calculation unit has multiple lower layer path calculation units, The first lower layer path calculation unit issues a path calculation request, and after receiving the path segment set returned by the lower layer path calculation unit, issues a path calculation request to another lower layer path calculation unit until all lower layer path calculation units are sent back the path segment set. Of course, if there is only one lower path calculation unit, it can be directly sent to the lower path calculation unit.
  • the path calculation request sent to another lower layer path calculation unit carries the egress node set information fed back by the previous lower layer path calculation unit, and the bottom path calculation unit performs the route calculation.
  • the set of ingress nodes is obtained according to the set information of the egress nodes, and the route sets are used as the end points of the path segments in the domain to perform route calculation.
  • the path merging module 330 is configured to combine the plurality of path segment sets received by the path segment information receiving module to generate a longer time when there are multiple lower layer path computing units in the routing domain that is to be sequentially followed according to the path to be established.
  • the set of path segments is sent to the path segment set sending module.
  • the path segment set sending module 340 is configured to send the received path segment set to the upper layer path computing unit.
  • the path calculation unit of this embodiment may further include a path calculation module, configured to calculate a path segment set of the managed routing domain when the path calculation unit is an underlying path calculation unit, or directly manage the routing path unit if it is not the underlying path calculation unit. Routing domain The path segment set of the directly managed routing domain is calculated, and the calculation result is sent to the path segment set sending module 340, and the path segment set sending module 340 sends the path segment set to the upper layer path calculating unit.
  • a path calculation module configured to calculate a path segment set of the managed routing domain when the path calculation unit is an underlying path calculation unit, or directly manage the routing path unit if it is not the underlying path calculation unit. Routing domain The path segment set of the directly managed routing domain is calculated, and the calculation result is sent to the path segment set sending module 340, and the path segment set sending module 340 sends the path segment set to the upper layer path calculating unit.
  • the path calculation module of the unit also calculates the information of the set of exit nodes and sends it to the upper path calculation unit. Of course, this information can also be obtained from the set of path segments by the upper path calculation unit without separately calculating.
  • the upper path calculation unit generates an ingress node set of the next routing domain according to the egress node information and sends it to the path calculation unit where the next routing domain is located, so that the path calculation unit calculates the path segment set by using the node aggregated by the ingress node set as the ingress node of the path. .
  • the set of egress nodes is obtained from the set of ingress nodes, as opposed to the above.
  • the embodiment of the invention further provides a route calculation system, the system comprising:
  • the top-level PCE is configured to determine a routing domain to be followed by the path to be established after receiving the path calculation request, and issue a path calculation command, where the path calculation request carries information of the first node and the second node, and is used to request to establish The direct path between the two.
  • the first bottom layer PCE is configured to perform routing calculation on the routing domain managed by the path calculation command, obtain a path segment set with the first node as an endpoint, and send the obtained path segment set to the upper layer PCE, where
  • the path calculation command can be sent directly by the top-level PCE, or can be sent to the first underlying PCE layer by layer by the top-level PCE.
  • the PCE sends a path calculation command, after receiving the path segment set sent by all the lower layer PCEs, combines all the received path segment sets and sends them to the upper layer PCE of the upper layer PCE; until the top PCE receives all the lower layer PCEs. a set of path segments sent;
  • the top-level PCE combines all received path segment sets to generate a set of paths between the first node and the second node.
  • Each PCE in the system includes: a path calculation command receiving module 310, a path segment letter The information receiving module 320, the path combining module 330, and the path segment set sending module 340.
  • the path calculation command receiving module 310 is configured to receive a path calculation command, where the path calculation command is directly sent by the top PCE, or the top layer PCE is transmitted layer by layer.
  • the path segment information receiving module 320 is configured to: after the path calculation command receiving module receives the path calculation command, according to the routing domain to be sequentially followed by the path to be established, if the PCE has multiple lower layer PCEs, first to the first lower layer PCE A path calculation request is issued, and after receiving the path segment set returned by the lower layer PCE, a path calculation request is sent to another lower layer PCE until all the path segment sets sent back by the lower layer PCE are received. Of course, if there is only one lower PCE, it will be sent directly to this lower PCE.
  • the path calculation request sent to another lower layer PCE carries the egress node set information fed back by the previous lower layer PCE, and the underlying PCE performs the route calculation according to the egress node set information. Get the set of ingress nodes, and then use these ingress sets as the endpoints of the path segments in the domain to perform route calculation.
  • the path merging module 330 is configured to combine the received plurality of path segment sets and send the merged path segment set when there are multiple lower layer PCEs in the routing domain that is to be sequentially followed according to the path to be established.
  • the path segment set sending module 340 is configured to send the received set of the merged path segments to the upper layer PCE.
  • the PCE of this embodiment may further include a path calculation module, configured to calculate a path segment set of the managed routing domain when the lower layer PCE is the underlying PCE, or directly manage the routing domain when the routing domain is directly managed but not the underlying PCE.
  • the path segment set of the routing domain is sent to the path segment set sending module 340, and the path segment set sending module 340 sends the path segment set to the upper layer PCE.
  • the path calculation of the PCE of other routing domains except the PCE that performs routing calculation first is first established from the sink node of the path to be established.
  • the module also calculates the information of the set of egress nodes and sends them to the upper PCE. Of course, this information can also be obtained from the set of path segments by the upper layer PCE without separately calculating.
  • the upper layer PCE generates an ingress node set of the next routing domain according to the egress node information and sends it to the PCE where the next routing domain is located,
  • the PCE calculates the path segment set by using the node aggregated by the set of ingress nodes as the ingress node of the path.
  • the set of egress nodes is obtained from the set of ingress nodes, as opposed to the above.
  • the technical solution of the present invention may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several The instructions are for causing a computer device (which may be a personal computer, server, or network device, etc.) to perform the methods described in various embodiments of the present invention.
  • a computer device which may be a personal computer, server, or network device, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Description

一种路由计算方法、 系统及路径计算单元 本申请要求于 2007 年 12 月 29 日提交中国专利局、 申请号为 200710125724.9, 发明名称均为 "一种路由计算方法、 单元及系统" 的中国专利申请的优先权， 其全部内容通过引用结合在本申请中。 技术领域
本发明涉及通信领域， 尤其相关于一种路由计算方法、 系统及 路径计算单元。 背景技术
通信网络是由地理上分布的多个节点，以及这些节点间用于传输 数据的通信链路组成的集合。通信网络中不相邻的节点之间可以通过 中间节点来通信， 这样可以让通信网络资源更节约， 利用率更高。 现 有的通信网络有多种类型， 例如 SDH (同步数字体系） /SONET (同 步光网络） 网络、 IP网络等。 通信网络的节点间一般通过交换数据帧 或者数据包在网络上通信， 这些帧或者包是由特定的协议规定的， 例 如 TCP/IP协议。 本文中所说的协议就是指类似 TCP/IP协议的， 定义了 节点间如何交互的规则的集合。
当通信网络变得 4艮庞大时， 就显得难以管理和维护， 因此一般划 分为多个路由域( routing domain )或者自治系统( Autonomous System, AS )进行管理。 自治系统中的网络一般由传统的执行域内路由协议 的"域内"路由器耦合在一起， 由共同的权力管理。 当要改进路由的可 伸缩性时， 一般会把自治系统划分为多个区 （area ) 。 一般而言， 域 ( domain )是指在共同的地址管理范围或者路径计算职责范围内的任 意网络单元集合， 因此， 域的例子可以是区， 自治系统和多个自治系 统。 为筒单起见， 本文中把路由域、 自治系统和区都统一筒称为域， 其具体含义可以依据上下文进行判定。 当要增加能交互数据的节点数 目时，会采用执行域间路由协议的域间路由器把不同域的节点互联起 来， 这种域间路由器也称为边界路由器（border router ) 。
域间路由协议的例子是边界网关协议 BGP ( Border Gateway Protocol version 4 ) , BGP通过在系统中相邻的域间路由器之间交换 路由和可达信息来执行域间的路由功能。 BGP—般用可靠的传输协 议， 例如 TCP, 来建立连接和会话。 域内路由协议， 也可以说内部网 关协议（Interior Gateway Protocol, IGP )的例子是开放最短路径优先 ( Open Shortest Path First, OSPF )路由协议。 OSPF协议基于链路状 态技术， 因此也是链路状态路由协议。 链路状态路由协议定义了一个 域内路由信息和网络拓朴信息交换和处理的方式， 例如 OSPF中这些 信息通过链路状态通告 （ Link State Advertisement, LSA )进行交换。
多协议标签交换 ( Multi-Protocol Label Switching, MPLS )技术的 出现和发展满足了数据网络发展的新需求， 例如保证可用带宽和快速 恢复。 MPLS技术允许在具有标签交换路由器（ Label Switched Router, LSR ) 的 IP/MPLS网络中建立端到端的隧道， 这些隧道一般被称为标 签交换路径 ( Label Switch Path, LSP ) 。 建立标签交换路径涉及到计 算网络中标签交换路由器的路径， 这种计算一般称为路由计算（route computation ) 。
多协议标签交换技术也引入到了光传送网领域， 发展出自动交换 光网给 ( Automatically switched optical network , ASON ) 。 和传统的 光传送网络通过手工配置或者网管半自动地配置来提供网络连接服务 不同， 自动交换光网络通过控制平面自动建立的方式来提供网络连接 服务。 自动交换光网络可以划分为真正承载网络服务的传送平面、 执 行管理功能的管理平面和运行控制协议的控制平面。
自动交换光网络的控制平面所使用的技术称为通用多协议标签 交换技术 ( Generalized Multi-Protocol Label Switching ) , 是对多协议 标签交换技术的扩展，它包括链路管理协议（ Link Manage Protocol ) 、 路由协议和信令协议。链路管理协议在邻接关系发现的基础上通过报 文交换得到该链路支持的连接类型和资源数目等信息，这些信息称为 流量工程（ Traffic Engineering , ΤΕ )信息， 这些包含流量工程信息的 链路称为流量工程链路。 在一个域内， 本地链路流量工程信息通过路 由协议， 例如流量工程扩展的开放最短路径优先（OSPF-TE )协议发 布到域内的其他节点。 以这些信息为基础， 当网络管理系统或者用户 要求网络建立一条网络连接服务时，连接的网络入口节点就可以进行 路径计算（path computation ) , 得到连接需要经过的链路序列， 然后 通过信令协议， 如流量工程扩展的资源预留协议（RSVP-TE ) , 向路 径上节点请求分配资源并建立交叉连接， 实现端到端连接的建立。
不管是 IP/MPLS网络， 还是光传送网络， 都存在前面提到的域划 分的问题。 特别具有流量工程管理能力的网络划分为多个域后， 每个 节点只有本域的流量工程信息和其它域的可达信息，减少了网络拓朴 变化对业务新建、 恢复阻塞的影响， 增强了网络的可扩展性。 但是分 域后， 每个域的节点只有本域的流量工程信息和其它域的可达信息， 没有其它域的完整流量工程信息，因此在多域的情况下如何计算出一 条满足端到端的带宽、 交换能力、 路由分离、 保护约束、 用户策略等 约束条件的路径成为了一个需要解决的问题。
为解决多域情况下的路由计算或者路径计算（为筒单起见， 统一 称为路由计算）问题， 域 -域路由协议（DDRP )技术采用层次网络的 模型， 将下层的一个域在上层用一个代理节点表示， 代理节点可以发 布代表域的抽象拓朴， 域间链路， 可达地址等， 这样逐层向上， 形成 一个层次的网络。 在计算一条跨多域的端到端路径时， 先计算出请求 节点所在域的严格路由以及边界节点后续的松散路由， 当信令走到中 间域的边界后再通过域边界计算等方式来计算该中间域的严格路由 直到到达宿节点所在的域。
如图 1所示，在解决多域情况下路由计算问题的另一技术——路 径计算单元（ Path Computation Element, PCE )技术中， 每个路径计 算单元保存了它所负责的域内的所有流量工程信息（为筒单起见， 把 网络拓朴信息也包含在流量工程信息中）。 需要算路的节点一般称为 路径计算客户 （Path Computation Client, PCC ), 发一个包含路由计 算参数信息的请求给路径计算单元，由该路径计算单元根据它自己存 储的流量工程数据库（Traffic Engineering Database, TED )进行路由 计算， 并将计算的结果反馈给请求节点。路径计算单元可以同时存储 一个或多个域的流量工程信息。 当计算一条跨多域的路由时， 如果超 出本路径计算单元所负责的范围， 它会利用路径计算单元通信协议
( Path Calculate Element Communication Protocol ,路径计算单元 CP ) 来与其它相关的多个路径计算单元协调计算出最终的路由。
在实现本发明的过程中， 发明人发现现有技术至少存在以下问 题：
现有技术一的解决方案是其路由计算是一个串行的过程，在首节 点计算路由时， 只有路径经过的部分域的入口和出口信息， 因为没有 相关域内的实际流量工程信息，只有当信令走到相应的域边界时触发 域边界计算才能判断从域入口到出口是否能算路成功，会很容易出现 信令走到中途发现无路可走或不满足路由约束，导致信令建路时多次 回滚， 之前建立好的交叉连接被删除重建； 还有应用该技术方案时计 算端到端的分离路由 （同源同宿的不同路径）很困难。
现有技术二是一种扁平化的单层模型，也就是说所有的路径计算 单元具有相同地位， 所以当网络复杂或者庞大时同样变得难以管理； 同时由于没有采用分层方式对实际网络进行抽象，在跨域路由计算的 时候完全依靠路径计算单元之间交换各域的流量工程信息来完成， 当 业务所经域较多的时候使得这些相关的路径计算单元之间的通信变 得非常频繁，且交互的信息量很大，降低了路由计算的效率和可靠性。
发明内容
本发明实施例提供一种路由计算方法、 系统及路径计算单元， 可 以高效地进行跨域的端到端路径计算。
本发明实施例提供一种路由计算方法， 该方法包括：
第一底层路径计算单元 PCE接收路径计算命令， 对其管理的路 由域进行路由计算， 获得以第一节点为端点的路径段集合， 并将获得 的路径段集合发送给其上层 PCE;
按照所述第一节点与第二节点间待建立路径所要依次经过的路 由域，收到下层 PCE发送的路径段集合的上层 PCE向另一个下层 PCE 发送路径计算命令， 在收到所有下层 PCE发送的路径段集合后， 将 收到的所有路径段集合进行组合后发送给所述上层 PCE的上层 PCE; 直至顶层 PCE收到所有下层 PCE发送的路径段集合；
所述顶层 PCE对收到的所有路径段集合进行组合， 生成所述第 一节点与所述第二节点之间路径的集合。
本发明实施例还提供一种路径计算单元， 包括：
路径计算命令接收模块， 用于接收路径计算命令；
路径段信息接收模块，用于在所述路径计算命令接收模块接收到 路径计算命令后， 按照待建立路径所要经过的路由域顺序， 逐个向下 层路径计算单元发送路径计算命令， 具体为： 在收到一个下层路径计 算单元返回的路径段集合后，向另一个下层路径计算单元发送路径计 算命令， 直到收到所有下层路径计算单元返回的路径段集合；
路径合并模块，用于将所述路径段信息接收模块收到的多个路径 段集合进行合并， 发送合并后的路径段集合；
所述路径段集合发送模块，用于将收到的所述合并后的路径段集 合发送给上层路径计算单元。
本发明实施例还提供了一种路由计算方法， 该方法包括： 接收路径计算命令；
按照待建立路径所要经过的路由域顺序， 逐个向下层 PCE发送 路径计算命令， 具体为： 在收到一个下层 PCE返回的路径段集合后， 向另一个下层 PCE发送路径计算命令， 直到收到所有下层 PCE返回 的路径段集合；
将接收到的多个路径段集合进行合并， 发送合并后的路径段集 合；
将收到的所述合并后的路径段集合发送给上层 PCE。
本发明实施例还提供一种路由计算系统， 该系统包括：
顶层 PCE, 用于在收到路径计算请求后， 根据路径计算请求携带 的第一节点、 第二节点的信息， 计算所述第一节点与所述第二节点间 待建立路径所要依次经过的路由域， 并直接或逐层向第一底层 PCE 发出路径计算命令；
所述第一底层 PCE,用于在收到路径计算命令后对其管理的路由 域进行路由计算， 获得以第一节点为端点的路径段集合， 并将获得的 路径段集合发送给其上层 PCE;
收到下层 PCE发送的路径段集合的上层 PCE, 用于按照所述第 一节点与所述第二节点间待建立路径所要依次经过的路由域，向另一 个下层 PCE发送路径计算命令， 在收到所有下层 PCE发送的路径段 集合后， 将收到的所有路径段集合进行组合后发送给所述上层 PCE 的上层 PCE;直至所述顶层 PCE收到所有下层 PCE发送的路径段集 合；
其中， 所述顶层 PCE对收到的所有路径段集合进行组合， 生成 所述第一节点与所述第二节点之间路径的集合。
由上可知， 应用本发明的技术方案， 采用分层及依次计算路径段 集合的方式， 可以高效地进行跨域的端到端路径计算。 附图说明
图 1为现有技术中跨域计算路径的网络结构示意图；
图 2为应用本发明实施例路由计算方法的网络结构示意图； 图 3为本发明实施例路径计算单元结构图。 具体实施方式
下面通过附图和实施例， 对本发明路由计算方法、 系统及路径计 算单元进行说明。
本发明路由计算方法实施例可以应用于图 2所示的网络结构中， 如图 2所示： 将网络分成多层， 在进行多域互连时， 每个底层路由域 由一个或多个 PCE直接负责该域的域内路由拓朴及其路由计算； 然 后，对底层的多个路由域由更高层的 PCE直接负责，该更高层的 PCE 具有这些底层的域间路由拓朴， 直接管理的路由域的域内路由， 以及 基于策略的一些域的域内路由拓朴。 按照这种思想， 对整个网络基于
PCE进行分层抽象管理。 网络的分层的层次数量没有限制， 可以基于 网络规模或者策略进行划分。各 PCE负责的所有路由拓朴（下层 PCE 负责的路由拓朴的集合）可以称为其计算域。
本发明实施例提出的大规模网络进行多域互连时，每个底层路由 域由一个或多个 PCE 负责该域的路由拓朴及其路由计算， 然后， 对 底层的多个路由域的 PCE抽象成更高层的 PCE, 该更高层的 PCE具 有这些底层的域间路由拓朴，以及基于策略的一些各域的域内路由拓 朴。 按照这种思想， 对整个网络基于 PCE进行分层抽象管理。 PCE 的分层的层次数量没有限制， 可以基于网络规模或者策略进行划分。
下面以具体实施例对本发明进行具体说明。
如图 2 所示： 最底层的第 0 层 levelO 由多个路由域构成： Domain 1、 Domain2、 Domain3、 Domain4 , 当然, 除了 Domain 1、 Domain2、 Domain3、 Domain4 外还可有若干个路由域， 为方便介绍 不再描述。 每个在底层的路由域具有直接对应的 PCE 负责该域的路 由拓朴和路由计算， 这些位于 levelO的 PCE是： PCE01、 PCE02、 PCE03、 PCE04。
对底层的多个 PCE所负责的拓朴进行抽象， 形成第一层 levell 的拓朴， 由 levell层的 PCE负责。 如图所示， 对 PCE01和 PCE02所 负责的相应域 Domain 1、 Domain2进行抽象，使它们处于同一计算域，
PCE11负责的这些域的抽象拓朴包含这些域的域间链路信息。 PCE12 的抽象方法与 PCE11类似， 在此不再详述。
对 Levell层的拓朴可以抽象到更高层次。 例如， 对 Leverl层的 PCE11、 PCE12、 ...PCE1X、 所负责的路由拓朴的计算域进一步抽象 分层， 并把这些路由拓朴由 Level2层的 PCE21负责。 PCE21的路由 拓朴包含 PCE11、 和 PCE12之间的互连信息， 在其它情况下基于策 略， PCE21也可以包含 PCE11、 和 PCE12的内部路由信息。
同时， 虽然图中没有注明， 但本领域技术人员通过上述描述应该 可以推测到： PCE11、 PCE12、 PCE21不是底层 PCE,但也可能具有直 接管理的路由域， 这种情况下 PCE11、 PCE12、 PCE21直接管理的路 由域分别位于 levell、 levell和 levevl2, 由 PCE11、 PCE12、 PCE21 直接计算其域内路由， 及该路由域与其它同层 PCE的计算域之间的 路由。 由于处理域内路由的计算主体不是底层 PCE外， 这种情况下 的路由计算方法与本发明实施例基本相同， 因此下文不再做单独描 述。
本层以及不同层 PCE之间通过路由协议（例如可以基于 IGP或 BGP协议）扩散相应的路由拓朴信息。 本层以及不同层 PCE之间通 过 PCE发现协议扩散 PCE的能力、 位置等信息。
本实施例中， 当源节点向其所在域的 PCE发起路径计算请求时， 收到请求的 PCE根据请求消息中携带的源、 目的地址信息进行判断， 如果请求的目的节点与源节点不在同一个路由域， 则该域 PCE根据 各层次 PCE的能力信息， 找出共同的上层 PCE (记为顶层 PCE ) , 该上层 PCE的计算域中同时包括了源节点与目的节点， 因此由其完 成到达目的节点的路径计算。
根据不同实施例， 顶层 PCE收到的源节点的路径请求消息中可 能携带有所要经过的域序列信息。如果源节点的路径计算请求中携带 了所要经过的域序列信息， 则该顶层 PCE根据该域序列完成相应的 路径计算。 否则， 该顶层 PCE根据其具有的抽象域间拓朴信息， 并 根据某种策略（如， 最短序列）确定整个跨域路径的域序列。
该顶层 PCE向靠近目的路由域的下层 PCE (记为顶层 PCE的右 侧方向）发起路径计算请求。 这个过程迭代， 直到路径计算请求发送 到目的节点所在域的最底层 PCE。
最底层的 PCE在本域内计算相应的最优路径段集合， 并返回给 上层 PCE。
相应的上层 PCE接收到路径计算请求后， 根据返回的路径段集 合信息， 向该层 PCE所管辖的其它 PCE发送路径计算请求， 这个过 程迭代，直到路径计算请求发送到该路径段的域出口所在节点的最底 层的 PCE。
上层的 PCE把计算的路径段集合返回给其上层的 PCE, 直至返 回给顶层 PCE。
顶层 PCE根据收集到的顶层 PCE右侧方向路径段集合， 然后向 顶层 PCE的左侧的下层 PCE发起路径计算请求。该过程与过程 3、 4、 5、 6类似。
最终顶层 PCE接收到下层 PCE返回的路径段集合， 根据这些路 径段集合可以计算出端到端跨域最优的路径，然后把这条最优的路径 返回给首节点所在的底层 PCE。
首节点所在底层 PCE接收到顶层 PCE的路径计算请求应答后， 把相应的路径信息返回给路径计算请求的首节点。
首节点接收到计算好的路径信息后，可以发起信令建立相应的路 径。
以上是计算端到端跨域路径， 实际上端到端跨域分离路径的计算 方式与以上过程类似， 因此也不再详述。
如图 2 所示， 本实施例中源节点为 N10, 其请求的目的节点为 N40。 为方便理解， 将本实施例方法分为两个子流程进行说明： 1 ) 确定顶层 PCE; 2 )计算路径。 具体流程如下：
1 )确定顶层 PCE
Sl l、 N10向 PCE01发出路径计算请求消息， 消息中携带有源节 点 N10的地址信息以及目的节点 N40地址信息。
512、 直接管理 N10的 PCE01接收到 N10发出的路径计算请求， PCE01储存有其负责的计算域的拓朴信息。 PCE01对路径请求消息中 的源、 目的地址进行判断， 通过判断结果发现 N10与 N40不在其所 负责的计算域中， 因此上 路径计算请求消息。
513、由图 2可知， PCE01对应的直接上层 PCE是 PCE11 , PCE11 收到上报的路径计算请求消息后进行与 PCE01 同样的判断， 判断结 果是 N40 不在其所负责的计算域中， 因此继续上报路径计算请求消 息。 S14、 PCE21 收到 PCE11 上 >¾的路径计算请求消息， 进行与 PCE01、 PCE11 同样的判断， 判断结果是源、 目的节点属于 PCE21 的计算域。 因此 PCE21是顶层 PCE。
需要说明的是， 上述步骤 S11-S14中， 是逐层上报的方式确定顶 层 PCE。 在其它实施例中， PCE01也可以直接储存其它 PCE的计算 能力信息， 在这种情况下， PCE01 可以直接判断出 PCE21 是顶层 PCE, 直接将路径技术请求消息发送给 PCE21。
2 )计算路径
521、 PCE确定最优的域序列， 最优域序列的确定可以是根据各 域网络状况或者配置的具体策略确定， 确定路径应该经过的域， 在此 不 做 详 细 说 明 。 支 设 本 实 施 例 中 ， 域 序 列 是 Domainl-Domain2-Domain3-Domain4。
522、 PCE21 的直接下层 PCE 中， N40所在的 Domain4属于 PCE12的计算域， 因此 PCE21向 PCE12发出路径计算命令。 本实施 例中， 路径计算命令中携带有最优域序列， 后续将按照这个序列的顺 序进行各域的域内路由计算。
523、 PCE12的直接下层 PCE中， Domain4属于 PCE04的计算 域， 因此 PCE12向 PCE04发送路径计算命令。
524、 PCE04收到路径计算命令后， 进行域内路径计算， 将域内 的最优路径段集合返回给 PCE12。 本实施例中， 还发送 PCE04计算 的域内路径段集合对应的入口节点集合的信息给 PCE12,这些入口节 点也就是 Domain04的边缘节点， 这些入口节点集合的信息可以单独 发送也可以由 PCE12从路径段集合的信息中获得。
525、 PCE12收到 PCE04返回的信息， 根据返回信息中各路径段 的入口节点集合， 确定 Domain03的出口节点集合， 并将该出口节点 集合作为路径计算的目的节点发送路径计算命令给 PCE03。
526、 PCE03计算本 Domain3内到达指定出口节点集合的最优路 径段集合， 并将计算结果返回给 PCE12。
527、 PCE12收到 PCE03返回的信息后，根据各路径的边缘节点， 拼接 Domain03 和 Domain04 的最优路径段集合， 计算出跨域 Domain03 和 Domain04 的路径段集合， 并将计算结果返回给顶层 PCE21。 在本实施例中， 按照出口节点信息进行拼接。
528、 PCE21向 PCE11发送路径计算命令， 进行与步骤 S22-S27 相似的处理： PCE11根据 PCE12反馈给 PCE21的跨域 Domain03和 Domain04的路径段的入口节点集合， 确定 Domain02的出口节点集 合并发给 PCE02, PCE02计算域内路径段集合并反馈给 PCE11 , 接 着 PCE01计算域内路径段集合并反馈给 PCE11 , 最后由 PCE11拼接 跨 Domainl和 Domain2的路径段集合。
529、 PCE21根据跨域 Domain03和 Domain04的路径段集合以及 跨域 DomainOl和 Domain02的路径段集合，计算出从 N10到 N40的 最优跨域路径集合， 并将路径信息返回给 PCE01 , 最后 PCE01将接 收到的路径计算结果返回给 N10。 在其它实施例中， 假设 PCE21还 有直接管理的路由域 Domain5 , 则 PCE21计算 Domain5的域内路径 段集合， 再将 Domain5的路径段集合、 Domain03和 Domain04的路 径段集合、跨域 DomainOl和 Domain02的路径段集合三种进行集合， 计算出从 N10到 N40的最优跨域路径。
需要说明的是， 在其它实施例中， 如果非 LevelO的 PCE有直接 管理的域， 与本实施例的区别仅在于计算其域内路径段的不是底层 PCE,其它与本实施例基本相同。例如假设本实施例中 PCE12还包括 有其它直接管理的域， 则 PCE12直接计算该域的域内路径段集合， 并在收到 Domain03和 Domain04的路径段集合后将三者进行整合即 可。 本实施例仅以三层结构进行举例， 本领域人员应该知道， 本发明 同样可以应用于一层、 两层或者更多层的情况。
以上步骤 S22、 S23中， PCE21通过逐层发送， 将路径计算命令 发送给 PCE04,在其它实施例中也可以直接发送给 PCE04。 在其它实 施例中， 在步骤 S24、 S25, 也可以不计算或发送入口节点集合与出 口节点集合， 而在步骤 S26中 PCE03仅依靠域内的拓朴情况进行路 由计算。 前述实施方式中， 先从 Domain4开始计算， 最后计算 Domainl , 但本领域人员应该知道，在其它实施例中也可以采用于本实施例相反 的顺序， 先计算 Domianl最后计算 Domian4。
请参考图 3 , 本发明实施例还提供一种路径计算单元， 包括： 路 径计算命令接收模块 310、 路径段信息接收模块 320、 路径合并模块 330以及路径段集合发送模块 340。
路径计算命令接收模块 310, 用于接收路径计算命令， 该路径计 算命令由顶层 PCE直接发送， 或者顶层 PCE逐层传送。
路径段信息接收模块 320, 用于在路径计算命令接收模块 310接 收到路径计算命令后， 按照待建立路径所要依次经过的路由域， 如果 该路径计算单元存在多个下层路径计算单元，则先向第一个下层路径 计算单元发出路径计算请求，并在收到下层路径计算单元返回的路径 段集合后向另一个下层路径计算单元发出路径计算请求，直到收到所 有下层路径计算单元发回路径段集合。 当然， 如果仅存在一个下层路 径计算单元， 则直接发给这个下层路径计算单元即可。 本实施例中， 为了方便后续的路径段合并，向另一个下层路径计算单元发出的路径 计算请求携带有前一个下层路径计算单元反馈的出口节点集合信息， 底层路径计算单元在进行路由计算时，根据这些出口节点集合信息得 到入口节点集合，再以这些入口集合为域内路径段的端点进行路由计 算。
路径合并模块 330, 用于当按照待建立路径所要依次经过的路由 域， 有多个下层路径计算单元时， 将所述路径段信息接收模块收到的 多个路径段集合进行合并，生成更长的路径段集合并发送给路径段集 合发送模块。
所述路径段集合发送模块 340, 用于将收到的路径段集合发送给 上层路径计算单元。
本实施例的路径计算单元， 还可以包括路径计算模块， 用于当这 个路径计算单元为底层路径计算单元时计算管理的路由域的路径段 集合， 或者虽然不是底层路径计算单元但有直接管理的路由域时， 计 算直接管理的路由域的路径段集合，并将计算结果发送给所述路径段 集合发送模块 340, 由路径段集合发送模块 340将路径段集合发送给 上层路径计算单元。
如果首先从待建立路径的宿节点开始建立路径， 本实施例中， 当 被计算的路由域有同层的其它路由域时，除了首先进行路由计算的路 径计算单元外，其它路由域的路径计算单元的路径计算模块还计算出 口节点集合的信息， 并发送给上层路径计算单元。 当然， 这些信息也 可以不用单独计算， 而由上层路径计算单元从路径段集合中获得。 上 层路径计算单元根据出口节点信息生成下一个路由域的入口节点集 合并发送给下一个路由域所在的路径计算单元，以便路径计算单元将 入口节点集合汇总的节点作为路径的入口节点计算路径段集合。在其 它实施例中， 如果首先从待建立路径的源节点开始建立路径， 则是根 据入口节点集合获得出口节点集合， 与上述情况相反。
本发明实施例还提供一种路由计算系统， 该系统包括：
顶层 PCE,用于在收到路径计算请求后确定待建立路径所要依次 经过的路由域， 并发出路径计算命令， 该路径计算请求中携带有第一 节点与第二节点的信息， 用以请求建立二者直接的路径。
第一底层 PCE,用于在收到路径计算命令后对其管理的路由域进 行路由计算， 获得以第一节点为端点的路径段集合， 并将获得的路径 段集合发送给其上层 PCE, 其中路径计算命令可以由顶层 PCE直接 发送， 也可以由顶层 PCE逐层发送给第一底层 PCE。
收到下层 PCE发送的路径段集合的上层 PCE, 用于按照第一节 点与第二节点间待建立路径所要依次经过的路由域， 向另一个下层
PCE发送路径计算命令，在收到所有下层 PCE发送的路径段集合后， 将收到的所有路径段集合进行组合后发送给其上层 PCE的上层 PCE; 直至所述顶层 PCE收到其所有下层 PCE发送的路径段集合；
其中， 所述顶层 PCE对收到的所有路径段集合进行组合， 生成 第一节点与第二节点之间路径的集合。
该系统中的各 PCE包括： 路径计算命令接收模块 310、路径段信 息接收模块 320、 路径合并模块 330以及路径段集合发送模块 340。 路径计算命令接收模块 310, 用于接收路径计算命令， 该路径计 算命令由顶层 PCE直接发送， 或者顶层 PCE逐层传送。
路径段信息接收模块 320, 用于在路径计算命令接收模块接收到 路径计算命令后， 按照待建立路径所要依次经过的路由域， 如果该 PCE存在多个下层 PCE,则先向第一个下层 PCE发出路径计算请求， 并在收到下层 PCE返回的路径段集合后向另一个下层 PCE发出路径 计算请求， 直到收到所有下层 PCE发回的路径段集合。 当然， 如果 仅存在一个下层 PCE, 则直接发给这个下层 PCE即可。 本实施例中， 为了方便后续的路径段合并， 向另一个下层 PCE发出的路径计算请 求携带有前一个下层 PCE反馈的出口节点集合信息， 底层 PCE在进 行路由计算时， 根据这些出口节点集合信息得到入口节点集合， 再以 这些入口集合为域内路径段的端点进行路由计算。
路径合并模块 330, 用于当按照待建立路径所要依次经过的路由 域， 有多个下层 PCE时， 将收到的多个路径段集合进行合并， 发送 合并后的路径段集合。
所述路径段集合发送模块 340, 用于将收到的该合并后的路径段 集合发送给上层 PCE。
本实施例的 PCE, 还可以包括路径计算模块， 用于当下层 PCE 为底层 PCE时计算管理的路由域的路径段集合， 或者虽然不是底层 PCE但有直接管理的路由域时， 计算直接管理的路由域的路径段集 合， 并将计算结果发送给所述路径段集合发送模块 340, 由路径段集 合发送模块 340将路径段集合发送给上层 PCE。
如果首先从待建立路径的宿节点开始建立路径， 本实施例中， 当 被计算的路由域有同层的其它路由域时， 除了首先进行路由计算的 PCE外， 其它路由域的 PCE的路径计算模块还计算出口节点集合的 信息， 并发送给上层 PCE。 当然， 这些信息也可以不用单独计算， 而 由上层 PCE从路径段集合中获得。 上层 PCE根据出口节点信息生成 下一个路由域的入口节点集合并发送给下一个路由域所在的 PCE,以 便 PCE将入口节点集合汇总的节点作为路径的入口节点计算路径段 集合。 在其它实施例中， 如果首先从待建立路径的源节点开始建立路 径， 则是根据入口节点集合获得出口节点集合， 与上述情况相反。
以上实施例提供的技术方案中全部或部分步骤，本领域的技术人 员可以清楚地了解到其可以通过硬件实现，也可以借助软件加必要的 通用硬件平台的方式来实现。基于这样的理解， 本发明的技术方案可 以以软件产品的形式体现出来，该软件产品可以存储在一个非易失性 存储介质 （可以是 CD-ROM, U盘， 移动硬盘等） 中， 包括若干指 令用以使得一台计算机设备（可以是个人计算机， 服务器， 或者网络 设备等）执行本发明各个实施例所述的方法。
最后应当说明的是:以上实施例仅用以说明本发明的技术方案而 非对其限制； 尽管参照较佳实施例对本发明进行了详细的说明， 所属 领域的普通技术人员应当理解：依然可以对本发明的具体实施方式进 行修改或者对部分技术特征进行等同替换；而不脱离本发明技术方案 的原则， 其均应涵盖在本发明请求保护的技术方案范围当中。

Claims

权利要求
1、 一种路由计算方法， 其特征在于， 该方法包括：
第一底层路径计算单元 PCE接收路径计算命令， 对其管理的路 由域进行路由计算， 获得以第一节点为端点的路径段集合， 并将获得 的路径段集合发送给其上层 PCE;
按照所述第一节点与第二节点间待建立路径所要依次经过的路 由域，收到下层 PCE发送的路径段集合的上层 PCE向另一个下层 PCE 发送路径计算命令， 在收到所有下层 PCE发送的路径段集合后， 将 收到的所有路径段集合进行组合后发送给所述上层 PCE的上层 PCE; 直至顶层 PCE收到所有下层 PCE发送的路径段集合；
所述顶层 PCE对收到的所有路径段集合进行组合， 生成所述第 一节点与所述第二节点之间路径的集合。
2、 如申请权利要求 1所述的方法， 其特征在于，
若所述上层 PCE仅有一个下层 PCE, 则直接将收到的路径段集 合发送给所述上层 PCE的上层 PCE。
3、 如申请权利要求 1所述的方法， 其特征在于， 所述上层 PCE 向所述另一个下层 PCE发送的路径计算命令， 携带有所述另一个下 层 PCE待计算路径段集合的边缘节点信息。
4、 如申请权利要求 3所述的方法， 其特征在于， 所述另一个下 层 PCE待计算路径段集合的边缘节点信息， 由所述上层 PCE根据前 一个下层 PCE反馈的边缘路径段集合信息计算得到。
5、 如申请权利要求 1所述的方法， 其特征在于， 所述第一底层 PCE接收路径计算命令包括：
所述顶层 PCE直接向所述第一底层 PCE发送路径计算命令； 或 者，
所述顶层 PCE向下层 PCE发送路径计算命令， 收到该路径计算 命令的 PCE继续向其下层 PCE发送路径计算命令， 直至所述第一底 层 PCE收到该路径计算命令。
6、 如申请权利要求 1所述的方法， 其特征在于， 所述第一节点 为待建立路径的源节点，所述第二节点为待建立路径的宿节点；或者， 所述第二节点为待建立路径的源节点，所述第一节点为待建立路 径的宿节点。
7、 如申请权利要求 1所述的方法， 其特征在于， 所述第一底层 PCE接收的路径计算命令，携带有第一节点与第二节点间待建立路径 所要依次经过的路由域的信息。
8、 一种路径计算单元， 其特征在于， 包括：
路径计算命令接收模块， 用于接收路径计算命令；
路径段信息接收模块，用于在所述路径计算命令接收模块接收到 路径计算命令后， 按照待建立路径所要经过的路由域顺序， 逐个向下 层路径计算单元发送路径计算命令， 具体为： 在收到一个下层路径计 算单元返回的路径段集合后，向另一个下层路径计算单元发送路径计 算命令， 直到收到所有下层路径计算单元返回的路径段集合；
路径合并模块，用于将所述路径段信息接收模块收到的多个路径 段集合进行合并， 发送合并后的路径段集合；
所述路径段集合发送模块，用于将收到的所述合并后的路径段集 合发送给上层路径计算单元。
9、 如申请权利要求 8所述的路径计算单元， 其特征在于， 还包 括：
路径计算模块， 用于根据收到的路径计算命令， 计算直接管理的 路由域的路径段集合， 并将计算结果发送给所述路径段集合发送模 块。
10、 如申请权利要求 8所述的路径计算单元， 其特征在于： 所述路径计算命令接收模块收到的路径计算命令中携带有边缘 节点信息，所述边缘节点信息由上层路径计算单元根据前一个下层路 径计算单元反馈的边缘路径段集合信息计算得到。
11、 一种路由计算方法， 其特征在于， 该方法包括：
接收路径计算命令； 按照待建立路径所要经过的路由域顺序， 逐个向下层 PCE发送 路径计算命令， 具体为： 在收到一个下层 PCE返回的路径段集合后， 向另一个下层 PCE发送路径计算命令， 直到收到所有下层 PCE返回 的路径段集合；
将接收到的多个路径段集合进行合并， 发送合并后的路径段集 合；
将收到的所述合并后的路径段集合发送给上层 PCE。
12、 一种路由计算系统， 其特征在于， 该系统包括：
顶层 PCE, 用于在收到路径计算请求后， 根据路径计算请求携带 的第一节点、 第二节点的信息， 计算所述第一节点与所述第二节点间 待建立路径所要依次经过的路由域， 并直接或逐层向第一底层 PCE 发出路径计算命令；
所述第一底层 PCE,用于在收到路径计算命令后对其管理的路由 域进行路由计算， 获得以第一节点为端点的路径段集合， 并将获得的 路径段集合发送给其上层 PCE;
收到下层 PCE发送的路径段集合的上层 PCE, 用于按照所述第 一节点与所述第二节点间待建立路径所要依次经过的路由域，向另一 个下层 PCE发送路径计算命令， 在收到所有下层 PCE发送的路径段 集合后， 将收到的所有路径段集合进行组合后发送给所述上层 PCE 的上层 PCE;直至所述顶层 PCE收到所有下层 PCE发送的路径段集 合；
其中， 所述顶层 PCE对收到的所有路径段集合进行组合， 生成 所述第一节点与所述第二节点之间路径的集合。
13、 一种计算机程序产品， 包括计算机程序代码， 所述计算机程 序代码通过网元设备实施， 其特征在于， 触发所述网元设备执行步骤 包括： 接收路径计算命令； 按照待建立路径所要经过的路由域顺序， 逐个向下层 PCE发送路径计算命令， 具体为： 在收到一个下层 PCE 返回的路径段集合后， 向另一个下层 PCE发送路径计算命令， 直到 收到所有下层 PCE返回的路径段集合； 将接收到的多个路径段集合 进行合并， 发送合并后的路径段集合； 将收到的所述合并后的路径段 集合发送给上层 PCE。
PCT/CN2008/073654 2007-12-29 2008-12-22 一种路由计算方法、系统及路径计算单元 WO2009092252A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP08871309A EP2237501B1 (en) 2007-12-29 2008-12-22 Routing computation method and system, and path computation element
JP2010540006A JP5189171B2 (ja) 2007-12-29 2008-12-22 経路計算方法及びシステム並びに経路計算エレメント
US12/824,690 US8737394B2 (en) 2007-12-29 2010-06-28 Route computation method and system, and path computation element

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200710125724.9 2007-12-29
CN2007101257249A CN101471853B (zh) 2007-12-29 2007-12-29 一种路由计算方法、单元及系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US12/824,690 Continuation US8737394B2 (en) 2007-12-29 2010-06-28 Route computation method and system, and path computation element

Publications (1)

Publication Number Publication Date
WO2009092252A1 true WO2009092252A1 (zh) 2009-07-30

Family

ID=40828990

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2008/073654 WO2009092252A1 (zh) 2007-12-29 2008-12-22 一种路由计算方法、系统及路径计算单元

Country Status (5)

Country Link
US (1) US8737394B2 (zh)
EP (1) EP2237501B1 (zh)
JP (1) JP5189171B2 (zh)
CN (1) CN101471853B (zh)
WO (1) WO2009092252A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015201891A (ja) * 2010-01-05 2015-11-12 日本電気株式会社 通信システム、制御装置、ノードの制御方法
JP2016021779A (ja) * 2010-11-22 2016-02-04 日本電気株式会社 通信システム、通信装置、制御装置、パケットフローの転送経路の制御方法及びプログラム

Families Citing this family (48)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8064360B2 (en) * 2009-01-23 2011-11-22 Empire Technology Development Llc Wireless home network routing protocol
CN101945033B (zh) * 2009-07-08 2014-08-13 中兴通讯股份有限公司 一种流量工程域合并/分割时的处理方法、装置及系统
CN102036126B (zh) * 2009-09-25 2014-07-02 中兴通讯股份有限公司 实现自动交换光网络和传统光网络互通的方法和系统
CN102065512B (zh) * 2009-11-12 2013-08-07 中兴通讯股份有限公司 多层网络中区域边界控制的方法、建立连接的方法和系统
CN102347893B (zh) * 2010-07-30 2016-05-11 中兴通讯股份有限公司 一种多层网络及lsp的建立方法
CN103098423A (zh) * 2011-08-16 2013-05-08 华为技术有限公司 不同域之间通信的方法和网络设备
CN102724111B (zh) * 2012-05-30 2018-08-10 南京中兴新软件有限责任公司 呼叫路径信息的获取方法及系统
KR101884713B1 (ko) 2012-07-20 2018-08-30 삼성전자주식회사 홈 네트워크 시스템 및 상기 시스템에서의 공유기의 네트워크 설정 방법
CN102811170B (zh) * 2012-07-25 2018-03-23 中兴通讯股份有限公司 一种实现域间路由洪泛的方法及代言节点
ES2710295T3 (es) * 2012-09-14 2019-04-24 Fluidmesh Networks S R L Método y sistema de gestión de movilidad en redes conmutadas por etiquetas
CN102917429B (zh) * 2012-10-22 2015-09-09 华为技术有限公司 报文传输方法及装置
US10404582B1 (en) 2012-12-27 2019-09-03 Sitting Man, Llc Routing methods, systems, and computer program products using an outside-scope indentifier
US10419334B1 (en) 2012-12-27 2019-09-17 Sitting Man, Llc Internet protocol routing methods, systems, and computer program products
US10397101B1 (en) 2012-12-27 2019-08-27 Sitting Man, Llc Routing methods, systems, and computer program products for mapping identifiers
US10212076B1 (en) 2012-12-27 2019-02-19 Sitting Man, Llc Routing methods, systems, and computer program products for mapping a node-scope specific identifier
US10397100B1 (en) 2012-12-27 2019-08-27 Sitting Man, Llc Routing methods, systems, and computer program products using a region scoped outside-scope identifier
US10404583B1 (en) 2012-12-27 2019-09-03 Sitting Man, Llc Routing methods, systems, and computer program products using multiple outside-scope identifiers
US10587505B1 (en) 2012-12-27 2020-03-10 Sitting Man, Llc Routing methods, systems, and computer program products
US10374938B1 (en) 2012-12-27 2019-08-06 Sitting Man, Llc Routing methods, systems, and computer program products
US10904144B2 (en) 2012-12-27 2021-01-26 Sitting Man, Llc Methods, systems, and computer program products for associating a name with a network path
US10476787B1 (en) 2012-12-27 2019-11-12 Sitting Man, Llc Routing methods, systems, and computer program products
US10447575B1 (en) 2012-12-27 2019-10-15 Sitting Man, Llc Routing methods, systems, and computer program products
US10419335B1 (en) 2012-12-27 2019-09-17 Sitting Man, Llc Region scope-specific outside-scope indentifier-equipped routing methods, systems, and computer program products
US10411997B1 (en) 2012-12-27 2019-09-10 Sitting Man, Llc Routing methods, systems, and computer program products for using a region scoped node identifier
US10411998B1 (en) 2012-12-27 2019-09-10 Sitting Man, Llc Node scope-specific outside-scope identifier-equipped routing methods, systems, and computer program products
JP2015023533A (ja) * 2013-07-23 2015-02-02 日本電気株式会社 通信システム
US9369387B2 (en) 2013-10-24 2016-06-14 Cisco Technology, Inc. Segment routing based wide area network orchestration in a network environment
EP3076611B1 (en) 2013-11-27 2018-11-21 Nec Corporation Communication system, communication method, network information combination apparatus, and network information combination program
CN104753778A (zh) * 2013-12-27 2015-07-01 中兴通讯股份有限公司 一种跨域路径处理方法和装置
WO2015127649A1 (zh) * 2014-02-28 2015-09-03 华为技术有限公司 全局主节点和域主节点、指示建立流及建立流的方法
EP3139552B1 (en) * 2014-05-21 2019-10-30 Huawei Technologies Co., Ltd. Virtual shortest path tree establishment and processing method, and path computation element
CN105227461B (zh) 2014-06-26 2020-03-10 中兴通讯股份有限公司 获取跨域分离路径的方法、路径计算单元
CN106688277B (zh) 2014-07-31 2021-03-23 康维达无线有限责任公司 在时隙化信道调频网络中的有效集中式资源和调度管理
CN105634714A (zh) * 2014-10-28 2016-06-01 中兴通讯股份有限公司 一种跨域时钟同步方法、装置和系统
CN105634943B (zh) * 2014-11-04 2019-09-24 南京中兴软件有限责任公司 路由计算方法及装置
CN105681190A (zh) * 2014-11-21 2016-06-15 中兴通讯股份有限公司 一种隧道约束信息的发送、接收方法及装置
JP6310409B2 (ja) * 2015-02-25 2018-04-11 日本電信電話株式会社 通信パス管理装置、通信パス管理方法、および通信パス管理プログラム
JP2018509842A (ja) * 2015-03-19 2018-04-05 ゼットティーイー コーポレイション マルチドメイン仮想トポロジー（ｍｄｖｔ）を確立および管理するための方法およびシステム
CN106850430A (zh) * 2015-12-03 2017-06-13 华为技术有限公司 一种域间路由方法、装置及网络侧设备
US10404571B2 (en) * 2015-12-29 2019-09-03 Futurewei Technologies, Inc. Communication among network controllers
CN107204928B (zh) * 2016-03-18 2021-06-08 华为技术有限公司 更新时钟同步拓扑的方法、确定时钟同步路径的方法及设备
US10728098B2 (en) 2016-07-06 2020-07-28 Futurewei Technologies, Inc. Connections and accesses for hierarchical path computation element (PCE)
CN108234312B (zh) * 2016-12-15 2021-03-05 中国电信股份有限公司 一种流量调度方法、pce及sdn网络系统
CN109426681B (zh) * 2017-08-25 2023-10-31 苏州织巢信息科技有限公司 一种线路自动化生成方法及装置
CN107959681B (zh) * 2017-12-06 2019-04-30 中盈优创资讯科技有限公司 Ip网络端到端的访问路径确定方法及装置
CN110661704B (zh) * 2018-06-30 2021-10-26 华为技术有限公司 转发路径的计算方法及sdn控制器
CN111385198B (zh) * 2018-12-29 2022-05-10 北京华为数字技术有限公司 路径确定方法、装置及通信系统
CN110099002B (zh) * 2019-04-12 2021-06-04 烽火通信科技股份有限公司 一种路径计算方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1460808A2 (en) * 2003-03-19 2004-09-22 Alcatel Inter-Domain constraint-based shortest path first technique for supporting hierarchical routing in interconnected multi-domain optical transport networks
CN1561048A (zh) * 2004-03-02 2005-01-05 中兴通讯股份有限公司 一种通过路由服务器实现域用户路由的方法
CN1588920A (zh) * 2004-09-17 2005-03-02 清华大学 层次光网络中的并行层次光标记交换通道的建立方法
CN101083548A (zh) * 2006-06-02 2007-12-05 华为技术有限公司 一种多域路由计算方法和系统

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6389010B1 (en) * 1995-10-05 2002-05-14 Intermec Ip Corp. Hierarchical data collection network supporting packetized voice communications among wireless terminals and telephones
GB9608543D0 (en) * 1996-04-25 1996-07-03 Philips Electronics Nv Determining routes in a network comprising nodes and links
US7599349B2 (en) * 2004-01-29 2009-10-06 Cisco Technology, Inc. Computing inter-autonomous system MPLS traffic engineering LSP paths
JP2005252368A (ja) * 2004-03-01 2005-09-15 Nippon Telegr & Teleph Corp <Ntt> 経路計算システム、経路計算方法、及び通信ノード
US8996722B2 (en) * 2004-11-01 2015-03-31 Alcatel Lucent Softrouter feature server
US7646719B2 (en) * 2004-12-02 2010-01-12 Cisco Technology, Inc. Inter-domain TE-LSP selection
US8320255B2 (en) * 2005-02-02 2012-11-27 Cisco Technology, Inc. Inter-domain path computation technique
US9059867B2 (en) * 2005-03-02 2015-06-16 Cisco Technology, Inc. Technique for selecting a path computation element based on response time delay
CN100417117C (zh) * 2005-06-15 2008-09-03 华为技术有限公司 自动交换光网络中节点可达性的识别方法
US7599302B2 (en) * 2005-07-19 2009-10-06 Cisco Technology, Inc. Dynamic enforcement of MPLS-TE inter-domain policy and QoS
US7464348B1 (en) * 2005-09-30 2008-12-09 Cadence Design Systems, Inc. Method and system for mapping source elements to destination elements as interconnect routing assignments
WO2007041860A1 (en) * 2005-10-14 2007-04-19 Nortel Networks Limited Gmpls control of ethernet
US8369213B2 (en) * 2006-12-22 2013-02-05 Cisco Technology, Inc. Optimization of distributed tunnel rerouting in a computer network with path computation at an intermediate node
US7701940B2 (en) * 2007-03-09 2010-04-20 Cisco Technology, Inc. Inter-domain point-to-multipoint path computation in a computer network
US8077713B2 (en) * 2007-09-11 2011-12-13 Cisco Technology, Inc. Dynamic update of a multicast tree
US8064447B2 (en) * 2008-03-27 2011-11-22 Futurewei Technologies, Inc. Computing point-to-multipoint paths
US9231851B2 (en) * 2011-01-31 2016-01-05 Futurewei Technologies, Inc. System and method for computing point-to-point label switched path crossing multiple domains

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1460808A2 (en) * 2003-03-19 2004-09-22 Alcatel Inter-Domain constraint-based shortest path first technique for supporting hierarchical routing in interconnected multi-domain optical transport networks
CN1561048A (zh) * 2004-03-02 2005-01-05 中兴通讯股份有限公司 一种通过路由服务器实现域用户路由的方法
CN1588920A (zh) * 2004-09-17 2005-03-02 清华大学 层次光网络中的并行层次光标记交换通道的建立方法
CN101083548A (zh) * 2006-06-02 2007-12-05 华为技术有限公司 一种多域路由计算方法和系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2237501A4 *

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015201891A (ja) * 2010-01-05 2015-11-12 日本電気株式会社 通信システム、制御装置、ノードの制御方法
JP2017063506A (ja) * 2010-01-05 2017-03-30 日本電気株式会社 通信システム、制御装置、ノードの制御方法
US10200307B2 (en) 2010-01-05 2019-02-05 Nec Corporation Communication system, control device, processing rule setting method, packet transmission method, and program
JP2016021779A (ja) * 2010-11-22 2016-02-04 日本電気株式会社 通信システム、通信装置、制御装置、パケットフローの転送経路の制御方法及びプログラム
US9497118B2 (en) 2010-11-22 2016-11-15 Nec Corporation Communication system, communication device, controller, and method and program for controlling forwarding path of packet flow
JP2017163597A (ja) * 2010-11-22 2017-09-14 日本電気株式会社 通信システム、通信装置、制御装置、パケットフローの転送経路の制御方法及びプログラム
US10541920B2 (en) 2010-11-22 2020-01-21 Nec Corporation Communication system, communication device, controller, and method and program for controlling forwarding path of packet flow
US11134012B2 (en) 2010-11-22 2021-09-28 Nec Corporation Communication system, communication device, controller, and method and program for controlling forwarding path of packet flow

Also Published As

Publication number Publication date
EP2237501A1 (en) 2010-10-06
EP2237501A4 (en) 2010-12-15
EP2237501B1 (en) 2013-03-27
JP2011509014A (ja) 2011-03-17
JP5189171B2 (ja) 2013-04-24
US8737394B2 (en) 2014-05-27
CN101471853B (zh) 2012-02-29
CN101471853A (zh) 2009-07-01
US20100265943A1 (en) 2010-10-21

Similar Documents

Publication Publication Date Title
WO2009092252A1 (zh) 一种路由计算方法、系统及路径计算单元
US11451471B2 (en) Using PCE as SDN controller
US7593340B2 (en) Method and system for multi-domain route computation
EP3295622B1 (en) Path computation element central controllers (pceccs) for network services
CN101228739B (zh) 用于动态实施mpls-te域间策略和qos的方法、设备和系统
US8199642B2 (en) Dynamically and efficiently forming hierarchical tunnels
WO2011009380A1 (zh) 一种多层多域分布式光网络并行跨域故障恢复方法和装置
JP2006333469A (ja) 自律システムにおけるトラフィックエンジニアリングに関するトポロジの追跡
WO2009089778A1 (fr) Procédé, élément de calcul de trajectoire et système d&#39;obtention de trajectoire
CN104429030A (zh) 用信号告知双向分组te隧道的合作路由和非合作路由lsp
WO2015024440A1 (zh) 一种获取ip链路的链路开销值的方法及系统
WO2008011771A1 (fr) Procédé, équipement de réseau et système permettant d&#39;établir une voie de commutation dans un réseau internet optique
WO2012079421A1 (zh) 域间链路状态信息的处理方法及路径计算单元
Farrel et al. Problem Statement and Architecture for Information Exchange between Interconnected Traffic-Engineered Networks
Casellas et al. IDEALIST control plane architecture for multi-domain flexi-grid optical networks
Xia Research on signaling mechanism based on convergence network
Brandt Auto-bandwidth control in dynamically reconfigured hybrid-SDN MPLS networks
Martínez et al. Requirements and enhancements for the evolution of the GMPLS control plane of the ADRENALINE testbed to support multi-layer (MPLS-TP/WSON) capabilities
Takeda et al. Layer 1 VPN architecture and its evaluation
Chaieb et al. Generic architecture for MPLS-TE routing
Verchere et al. The Advances in Control and Management for Transport Networks
Chaieb et al. MPLS-TE Routing: Adopting a Generic Architecture and Evaluating Various Implementation Approaches

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 08871309

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2010540006

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2008871309

Country of ref document: EP