WO2005022823A1 - Procede pour realiser une voie avec entite de protection dans un reseau optique intelligent - Google Patents

Procede pour realiser une voie avec entite de protection dans un reseau optique intelligent Download PDF

Info

Publication number
WO2005022823A1
WO2005022823A1 PCT/CN2004/001009 CN2004001009W WO2005022823A1 WO 2005022823 A1 WO2005022823 A1 WO 2005022823A1 CN 2004001009 W CN2004001009 W CN 2004001009W WO 2005022823 A1 WO2005022823 A1 WO 2005022823A1
Authority
WO
WIPO (PCT)
Prior art keywords
link
path
node
protection
links
Prior art date
Application number
PCT/CN2004/001009
Other languages
English (en)
French (fr)
Inventor
Yu Wang
Junjie Feng
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.
Publication of WO2005022823A1 publication Critical patent/WO2005022823A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0287Protection in WDM systems
    • H04J14/0293Optical channel protection
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0227Operation, administration, maintenance or provisioning [OAMP] of WDM networks, e.g. media access, routing or wavelength allocation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0278WDM optical network architectures
    • H04J14/0284WDM mesh architectures

Definitions

  • the present invention relates to an intelligent optical network, and in particular, to a method for obtaining a path with a protected entity in an intelligent optical network.
  • the current transmission network is mainly based on synchronous digital hierarchy (SDH) / synchronous optical network (SONET) technology.
  • SDH synchronous digital hierarchy
  • SONET synchronous optical network
  • end-to-end service configuration is mainly manually configured according to customer needs.
  • the concept of an intelligent optical network has been proposed internationally. Its main function is to support end-to-end automatic configuration of services.
  • the intelligent optical network equipment In order to provide automatic service configuration capabilities on the transmission network, the intelligent optical network equipment must have end-to-end automatic service path optimization calculation capabilities.
  • the original transmission network itself has perfect protection capabilities. For example, the transmission network provides the protection capability of the multiplex segment protection ring based on the automatic protection switching protocol.
  • the intelligent optical network equipment needs to be compatible with the protection capability of the original transmission network.
  • CSPF Constrained Shortest Path First Calculation
  • the basic principle of the CSPF algorithm is as follows:
  • the CSPF algorithm uses two databases named as the shortest path tree (PATHS) and the potential next hop set (TENT), where the shortest path tree information is stored in PATHS, and TENT contains Before finding the shortest path Try node information.
  • PATHS shortest path tree
  • TENT potential next hop set
  • Step 1 Place the source node in TENT, and set the path cost of the source node to 0.
  • Step 2 Determine whether the current TENT is empty. If so, the shortest path to the destination node cannot be found, and the route calculation ends; otherwise, the node with the lowest path cost relative to the source node is selected in TENT and placed in PATHS.
  • the path cost is the sum of the link costs of all the links through which the path passes, and the link cost is calculated based on attributes such as the available bandwidth of the link and the length of the link.
  • the path cost refers to the path cost from the source node to a certain node.
  • a link is composed of two adjacent nodes
  • a path is composed of one or more links
  • a path composed of one link is a single-hop path
  • a path composed of more than one link is a multi-hop path.
  • Step 3 Determine whether the node just added to the PATHS is the destination node. If it is, then find the shortest path, output the successful calculation result, and the route calculation ends; otherwise, go to step 4.
  • Step 4 When a node is put into PATHS, check the link between the node and each neighbor node, match the attributes of the link with the constraint conditions, and match the constraint conditions of the neighbor nodes.
  • the constraint conditions for matching the attributes of the link are: whether the link is excluded, whether the available bandwidth of the link meets the specified bandwidth requirement, and whether the protection type of the link meets the specified protection requirement.
  • the constraints on matching neighboring nodes are: Whether the node is excluded. Whether a link or a node is excluded from the constraint of the excluded resource type means that the user can specify that the service path does not pass through certain nodes or certain links.
  • the link is ignored. If the constraint match is unsuccessful, the link is ignored. If the constraint matching is successful, calculate the path cost to the neighbor node through the link, and then look at the successful matching neighbor node In this state, if the neighbor node is neither in Paths nor TENT, then the neighbor node is placed in TENT; if the neighbor node already exists in Paths, it indicates that the new path is a longer path and is ignored New path
  • the neighbor node exists in TENT, it means that there is a new path to the neighbor node. Then, the path cost of the new path and the old path in TENT is compared. If the path cost of the new path is less than the path cost of the old path, then use The new path replaces the old path. If the path cost of the new path is greater than or equal to the path cost of the old path, the new path is ignored and the old path is retained; then, return to step 2.
  • step 4 the method for calculating the path cost to the neighbor node through the link is: the path cost to the neighbor node is equal to the path cost to the previous node plus the cost of the link.
  • a path passes through a protection entity, such as a multiplex segment protection ring
  • the links in the protection entity need to maintain the consistency of time slots, that is, two paths on the path that are continuous and belong to the same protection ring entity.
  • One or more links must have timeslots with the same channel number for service use, so that services can be guaranteed in the protection entity.
  • the existing path calculation methods do not consider the special requirements of the protection entity on the network for selecting a traffic engineering (TE) link. Summary of the invention
  • the main object of the present invention is to provide a method for obtaining a path with a protected entity in an intelligent optical network, so that it can obtain an optimal service path that meets user requirements in a network with a protected entity.
  • a method for obtaining a path with a protected entity in an intelligent optical network includes the following steps: a. Establish different traffic engineering (TE) links for data links of different protection types, and establish data links with the same protection attributes and belonging to the same protection entity as the same TE link;
  • TE traffic engineering
  • the establishment of the TE link in step a is: establishing different TE links according to different attributes of the time slot protection type in the intelligent optical network.
  • establishing different TE links according to different attributes of the timeslot protection type in the intelligent optical network specifically includes: using a TE link for working services and having a corresponding protection link to protect it as a protected TE link; a TE link used to protect the protected TE link as a protected TE link; a TE link used for working services and not protected by a corresponding protection link as unprotected TE link.
  • flooding the protected entity information as described in step b, and spreading the protected entity information in the entire routing area specifically includes:
  • LSA link state announcement
  • the protection ring information to which the TE link belongs includes: the same-ring inbound interface index of the local link when each protection ring entity passes the TE link.
  • the step c specifically includes:
  • c 1 The source node that obtained the network path is put into a set of potential next hops; c2. Select the node with the lowest path cost from the set of potential next hops and place it in the shortest path tree to determine whether the node is the destination node. If so, output the path calculation result and the route calculation ends; otherwise, according to the constraints Match the TE link with the slot consistency, and place the neighbor of the node with the smallest path cost into the set of potential next hops according to the state of the neighbor of the node with the smallest path cost;
  • step c3 Repeat step c2 above until the node with the smallest path cost is the destination node.
  • matching the TE link according to the constraint condition and the time slot consistency includes: judging whether the TE link that the node with the lowest path cost reaches its neighbor node matches the constraint widget, and if it does not match, ignore the TE chain If it matches, check whether the TE link meets the slot consistency with the previous link. If it does, then select the TE link and calculate the node that has the highest cost to reach the path through the TE link. The path cost of the neighbor node, otherwise the TE link is ignored.
  • the method for checking whether the TE link satisfies the slot consistency with the previous Nai link is: judging whether the TE link reaching the neighbor node and the previous link on the path belong to the same protection ring entity, if If they do not belong to the same protection ring entity, the time slot consistency is satisfied; if they belong to the same protection ring entity, then all TE links on the path that belong to the same protection ring entity as the TE link are found, and then all the TEs found are determined. Whether the time slots with the same channel number are used for the link. If the time slots with the same channel number are used, the time slot consistency is satisfied; otherwise, the time slot consistency is not satisfied.
  • the calculating the path cost to the neighbor node includes:
  • the placing the neighbor node into a potential next hop set specifically includes: If the neighbor node is neither in the potential next hop set nor in the shortest path tree, placing the neighbor node in the potential next hop set;
  • the neighbor node is in the shortest path tree, ignore the neighbor node; if the neighbor node is in the set of potential next hops, reach the new and old path sizes of the neighbor node and the new and old paths The node status of the link determines the path to the neighbor node.
  • determining the path to the neighbor node based on the new and old paths to the neighbor node and the new and old links is as follows:
  • the path cost between the new path and the old path in the set of potential next hops is compared. If the path cost of the new path is less than the path cost of the old path, the new path is used instead of the old path. If the path cost of the path is greater than or equal to the path cost of the old path, the new path is ignored.
  • the method for selecting a link through a previous link on a path is specifically: if the new link and the previous link on the path belong to the same protection ring, then a new link is selected and the old path is replaced by the new path Path; if the old link and the previous link on the path belong to the same protection ring, the old link is selected and the new path is ignored; otherwise, if it is not possible to determine which link to choose, the neighbor node reached by the link A link undetermined flag is set, and the retrieval information and corresponding path costs of the multiple TE links are recorded on the neighbor node.
  • step c further includes:
  • the step c4 specifically includes:
  • the shortest path that satisfies the constraint conditions can be obtained based on the inherent protection type of the transmission network, and the method of obtaining the protection topology in advance can effectively reduce the number of repeated calculations and improve network efficiency.
  • FIG. 1 is a flowchart of a method for obtaining a path with a protection entity in an intelligent optical network according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of link channel protection under the protection of a two-fiber bidirectional multiplex segment
  • FIG. 3 is a schematic diagram of a link of a multiplex segment ring using a same-ring inbound interface index of the link as protection ring information according to an embodiment of the present invention
  • Figure 4 is a schematic diagram of an intelligent optical network topology
  • Figure 5, Figure 6, Figure 7, Figure 8, Figure 9, and Figure 10 are schematic diagrams of different processes of the path in the process of building the shortest path tree of the intelligent optical network topology shown in Figure 4;
  • FIG. 11 is a schematic diagram of a process of determining a link in a protection entity through a backtracking mechanism in the process of constructing the shortest path tree of the topology of the intelligent optical network shown in FIG. 4.
  • the main idea of the invention is: establishing different flows for data links of different protection types Engineering (TE) link, and data links with the same protection attributes and belonging to the same protection entity are established into the same TE link, and then the established TE link is flooded to obtain the protection entity in the network route of.
  • TE protection types Engineering
  • the TE link is formed by bundling data links with the same TE attributes.
  • the data link can be understood as the minimum bandwidth granularity, which can be VC-4 or VC-12.
  • a TE link can be a single optical fiber, multiple optical fibers, or only a part of a time slot in an optical fiber. It is a logical concept.
  • the protection entity refers to a special network topology entity, such as a linear multiplex section and a multiplex section protection ring, which is established in a transmission network in order to implement protection functions for services.
  • FIG. 1 shows a flowchart of a method for obtaining a path with a protected entity in an intelligent optical network according to an embodiment of the present invention.
  • the method of the present invention includes the following steps:
  • Step 101 Establish different TE links for data links of different protection types. Then, data links having the same protection attribute and belonging to the same protection entity are established as the same TE link. Specifically, the TE link can be divided and established into the following categories:
  • a TE link for working services has a corresponding protection link to protect it;
  • the TE link used to protect the protected TE link can transmit additional services, but can be occupied when protection is implemented;
  • Unprotected TE link A TE link for working services, but there is no protection link to protect it, and it cannot be preempted by any other service.
  • time slots in an optical fiber have the same attributes, and a TE link is established for flooding.
  • the so-called flooding refers to spreading a certain link state in the routing area to distribute and synchronize the link state database between routers.
  • the following situations require special treatment:
  • all time slots in an optical fiber link belong to the protection entity and have the same protection type, but belong to different protection entities.
  • a part of time slots in an optical fiber belongs to a linear multiplex section, a part of time slots belongs to the MSP_1 ring, and a part of time slots belongs to the MSP-2 ring.
  • the protection type of all time slots is 1: 1, a TE link cannot be established.
  • Different TE links need to be established for the time slots in different protection entities, that is, three TE link information must be established for the above-mentioned one optical fiber.
  • the following is a further example of the division and establishment of the TE link.
  • Figure 2 is a schematic diagram of link channel protection under two-fiber bidirectional multiplex segment protection.
  • the channels filled with diagonal lines are protected channels
  • the channels filled with dots are protected channels
  • the channels that are not filled are unprotected. aisle.
  • channels 1-3, 5, and 8 of the 16 channels are protected channels
  • channels 9-11, 13, and 16 are protection channels, and correspondingly protects the protected channels in the front in order.
  • These protection channels It can carry additional services, and the services carried on the protection channel will be preempted when the two-way shared multiplex section fails; other channels, that is, channels 4, 6, 7, 12, 14, and 15 are unprotected channels.
  • These unprotected channels can carry unprotected services.
  • channels ⁇ 1, 2, 3, 5, 8 ⁇ are bundled into a TE link, and the protection attribute is 1: 1; channel ⁇ 9, 10, 11, 13, 16 ⁇ is bundled Into a TE link, the protection attribute is a protection link, which can carry additional services; other channels ⁇ 4, 6, 7, 12, 14, 15 ⁇ are bundled into a TE link, the protection attribute is unprotected, and services cannot be seize.
  • Step 102 Flood the protection entity information and diffuse the protection in the entire routing area Entity information.
  • the information diffusion of the protection entity is implemented through the flooding process of routing protocols, such as OSPF, IS-IS and other protocols.
  • the information of the protection entity is mainly included in the link state announcement (LSA) of the TE link type (TE-LINK). ). Therefore, in this embodiment, the information of the protection entity is flooded through the TE link type LSA.
  • LSA link state announcement
  • TE link type LSA The following describes the TE link type LSA in detail:
  • the TE link-type LSA is an opaque (opaque) LSA and is used to flood TE link information.
  • the LSA includes the interface index of the TE link, bandwidth information, protection type, time slot occupation, and information about the protection ring to which it belongs.
  • the information related to the protection entity is: the protection type of the TE link, the time slot occupation situation and the information of the protection ring to which it belongs.
  • the protection types include 1 + 1 protection, 1: 1 protection, or no protection. For example, both the linear multiplex section and the multiplex section ring are 1: 1 protection types.
  • the timeslot occupation of the TE link can be represented by 8 bytes, 8 bytes represent 64 timeslots, each bit of each byte represents 1 timeslot, and 0/1 represents the timeslot Whether it is available. A maximum of 8 bytes can represent a TE link with a bandwidth of 10G.
  • the information of the protection ring to which the TE link belongs is used to identify which protection ring the TE link belongs to.
  • a device in the network may belong to multiple different protection rings, that is, the links of multiple protection rings pass through the device.
  • the protection ring information to which the TE link belongs includes a TE link attribute, which is defined as a local interface index of a TE link that belongs to the same protection ring as the TE link and has the same protection type but in the opposite direction. In the following, this TE link attribute is referred to as the same-loop inbound interface index of the TE link.
  • the protection ring shown in FIG. 3 is a multiplex segment ring composed of four network elements NE1, NE2, NE3, and NE4, and the attributes of each link are shown in FIG. 3 and Table 1.
  • the link may be expressed in the form of "network element, local interface index of the link-remote interface index of the link", for example, NEl, 0x6cl-0x6c3, then the in-loop interface index of the link is 0x6c2. If the link is not an on-ring link, the in-ring interface index of the link is 0.
  • the protection ring information of the TE link can also be achieved by assigning a unique ring ID to a certain multiplex segment ring.
  • the protection entity information in the entire area can be known on each intelligent device in the entire routing area, and the progress is obtained. Data foundation for route calculation.
  • the following steps 103 to 115 are the route calculation process of the present invention to obtain the path with the protection entity in the network according to the information of the protection entity after flooding.
  • Step 103 Obtain a source node and a destination node of the network path, place the source node in a potential next hop set, and set the path cost of the source node to 0.
  • Steps 104-106 determine whether the current potential next hop set is empty, if so, the shortest path with the protected entity cannot be found, an error message is returned, and the route calculation ends; otherwise, select from the potential next hop set The node M with the smallest path cost relative to the source node is put into the PATS.
  • Step 107 Determine whether the node M put in the PATHS is the destination node, and if the node If M is not the destination node, step 108 is performed, otherwise step 114 is performed.
  • Steps 108-111 Find the TE link where node M added to the path reaches its neighbor node, determine whether the found TE link matches the constraint, and if it does not match, ignore the link; if it matches, then further judge the node Whether the TE link that M reaches the neighbor node and the previous link on the path meet the timeslot consistency, if the timeslot consistency is not satisfied, the link is ignored, otherwise step 112 is performed.
  • Time slot consistency means that when a certain link or all links in the path are links of the same ring, crossovers must be established using the same time slot to ensure that services get multiplexed segments in the event of a node failure in the path protection. Therefore, the time slot consistency of the link needs to be checked.
  • the following methods can be used to check whether the TE link reaching the neighboring node meets the time slot consistency with the previous link on the path, which specifically includes: judging whether the TE link reaching the neighboring node and the previous link on the path belong to the same protection ring An entity, if the two TE links do not belong to the same protection ring entity, the slot consistency of the two TE links is not considered, and it is considered that the slot consistency is satisfied; If the links belong to the same protection ring entity, find all TE links on the path that belong to the same protection ring entity as the TE link, including the two TE links, and then determine whether all TE links have the same The time slot with the channel number is used. If the time slot with the same channel number is used, the time slot consistency is satisfied; otherwise, the time slot consistency is not satisfied.
  • Steps 112 to 113 Calculate the path cost of reaching the neighbor node of node M through the TE link, and then, according to the state of the neighbor node of this node M, place the neighbor node of this node M into the set of potential next hops Then, return to step 104.
  • a method of increasing a link cost of a non-same link is used to minimize a path cost of the same link node, so as to reach a potential
  • it is preferred choose the purpose of the link link node.
  • K is used to increase the link cost of non-same-loop links, which is a sufficiently large constant, such as 500000.
  • the link cost of the same-loop link can also be reduced, for example, fixed to 1 to minimize the path cost of the same-loop link node.
  • the calculation method can be specifically:
  • the first nanochain may sometimes be selected
  • the path with the lowest cost is not the path with the lowest cost. Therefore, in practical applications, such a path cost calculation method that reduces the link cost of the same-loop link needs to be selected according to the actual network structure.
  • the neighbor node In the process of placing the neighbor node of the node M into the set of potential next hops, the following principles need to be followed: If the neighbor node is neither in the PATHS nor the TENT, then the neighbor node is placed in the TENT; if the neighbor is If the node already exists in the path, it means that the new path is a longer path, and the new path is ignored, that is, the neighbor node is not put into the set of potential next hops;
  • the neighbor node exists in the TENT, it indicates that there is a new path to the neighbor node, and it also indicates that there are at least two TE links to the neighbor node. Whether the two old and new links mentioned above belong to the same node, that is, the nodes constituting the link are the same, and if they belong to the same node, the link is selected from the two old and new links through the previous link on the path; If the new and old links belong to different nodes, compare the new path with
  • the path cost of the old path in TENT If the path cost of the new path is less than the path cost of the old path, the new path is used to replace the old path. If the path cost of the new path is greater than or equal to the path cost of the old path, the new path is ignored. path.
  • the method for selecting a link from the new and old two-nanai link through the previous link on the path may specifically be: if the new link and the previous link on the path belong to the same protection ring, then selecting the new chain And replace the old path with a new path; if the old link and the previous link on the path belong to the same protection ring, select the old link and ignore the new path; otherwise, if you cannot determine which link to choose, select
  • the neighbor node where the link arrives sets a link undetermined flag, and records the retrieval information of the Donai TE link and the corresponding path cost on the neighbor node.
  • Step 114 The minimum path is successfully constructed, and the destination node of the path is traced back to determine the consistency of the time slot of the previous TE link through the information of the next TE link.
  • the specific backtracking method may be as follows: starting from the destination node of the path, each node on the path is checked forward, and if a link undetermined flag is not set on the node, the link to the node has been determined; if set on the node If the link is not determined, then a TE link that belongs to the same protection ring as the next link on the path is selected from the multiple TE links recorded on the node; if the TE link that does not belong to the next link on the path cannot be selected For TE links on the same protection ring, the TE link corresponding to the path with the lowest path cost is selected. For example, if you trace back to node N and find that the link indeterminate flag is set on node N, determine the link and time slot from node N-1 to node N through the link from node N to node N + 1.
  • Step 115 Output the path calculation result, and the route calculation ends.
  • Figure 4 depicts a schematic diagram of an intelligent optical network topology, where 1 to 9 represent nine nodes, of which ⁇ 2, 3, 4, 5, 6 ⁇ form a multiplex segment ring, and the link cost is 160; ⁇ 3, 4, 8, 7 ⁇ constitutes another multiplex segment ring with a link cost of 320; node 1 to node 2 and node 4 to node 9 are linear multiplex segments with a link cost of 40.
  • the figure includes two parts: the shortest path tree and the set of potential next hops.
  • the shortest path tree on the left uses a binary tree representation, where P is the parent pointer, C is the child pointer, and S is the sibling pointer.
  • the set of potential next hops on the right contains the attempted nodes to find the shortest path. Since node 1 is the source node, the path cost of node 1 is zero.
  • node 1 with the lowest path cost is placed in the shortest path tree, and neighbor node 2 is selected into the set of potential next hops. Because the link from node 1 to node 2 is a multiplex segment, it meets the protection constraint of 1: 1. The path cost from node 1 to node 2 is 40.
  • the node 2 with the lowest path cost is selected from the set of potential next hops and added to the shortest path tree, while the neighbors of node 2 are selected and put into the set of potential next hops. Because the path to node 2 is not the path in the protection ring, there is no requirement for time slot consistency when selecting the link. Then, node 3 and node 6 are put into the set of potential next hops. It should be noted that the link reaching node 3 and node 6 is the link on the protection ring. Because the link from node 2 to node 3 or node 6 and the link from node 1 to node 2 are not links on the same protection ring, it is necessary to increase the link cost of node 2 to node 3 or node 6 and set different rings. The incremental cost of the link is 500000, so the path cost of node 1 to node 3 or node 6 is 500200.
  • the node 6 with the smallest path cost is selected from the set of potential next hops and put into the shortest path tree, while the neighbors of node 6 are selected and put into the set of potential next hops. Because In order to reach the node 6 is a path in the protection ring, it is necessary to have time slot consistency requirements when selecting the link, because the link to the node 5 and the link to the node 6 belong to the same protection ring. So the same time slot is selected, and node 5 is put into the set of potential next hops. The path cost from node 1 to node 5 is 500360.
  • the node 3 with the lowest path cost is selected from the set of potential next hops and added to the shortest path tree, while the neighbors of node 3 are selected and put into the set of potential next hops. Because the path to node 3 is the path in the protection ring, it is necessary to have the time slot consistency requirement when selecting the link. Since node 3 has two links to reach neighbor node 4, it is preferred to select the link that belongs to the same protection ring as reaching node 3, and select the same time slot in the link, and put node 4 and node 7 into the potential One hop collection.
  • the path cost from node 1 to node 4 is 500360.
  • the link cost of node 3 to node 7 needs to be increased, and the path cost of node 1 to node 7 It is 1000520.
  • the node 5 with the lowest path cost is selected from the set of potential next hops and put into the shortest path tree, while the neighbors of node 5 are selected and put into the set of potential next hops. Because node 4 is already in the set of potential next hops, and the new path through node 5 to node 4 is longer than the old path, the new path is ignored.
  • the node 4 with the smallest path cost is selected from the set of potential next hops and put into the minimum path tree, while the neighbors of node 4 are selected and put into the set of potential next hops. Because node 5 is already in the set of potential next hops, and the old path is shorter than the path reached from node 4, the old path is kept unchanged. Because the link to the local node 8 or node 9 and the link to the node 4 are not in the same protection ring, there is no requirement for time slot consistency, and the node 8 and node 9 are placed in the set of potential next hops.
  • the path cost from node 1 to node 8 is 1000680.
  • the path cost from node 1 to node 9 is 1000400.
  • the node 9 with the smallest path cost is selected from the set of potential next hops and placed In the shortest path tree, at the same time, the neighbor node of node 9 is selected and put into the potential next hop set. Since node 4 has been placed in the shortest path tree, neighbor node 4 of node 9 cannot be added to the set of potential next hops.
  • the node 7 with the smallest path cost is selected from the set of potential next hops and put into the shortest path tree, while the neighbors of node 7 are selected and put into the set of potential next hops. Since the neighbor node 8 of node 7 is already in the set of potential next hops, and the new path to reach node 8 via node 7 is longer than the old path, the old path is retained.
  • node 8 with the lowest path cost from the set of potential next hops is selected and added to the shortest path tree. Since node 8 is the destination node, the calculation ends when it reaches the destination node.
  • FIG. 11 is a schematic diagram of a process of determining a link in a protection entity through a backtracking mechanism in the process of constructing the shortest path tree of the topology of the intelligent optical network shown in FIG. 4.
  • what is to be calculated is: The shortest path from node 3 to node 1 for protection.
  • node 3 selects node 3 and put it in the shortest path tree.
  • node 2 and node 7 are easier to determine the link, because there is only one node 2 and 7 link.
  • the link on the same ring cannot be selected through the previous link, so it cannot be determined which link is selected. Since the link from node 3 to node 4 cannot be determined temporarily, a link undetermined mark is made on this node 4 and then the shortest path tree is continued to be constructed.
  • the TE link corresponding to the path with the lower path cost should be selected, that is, the path with the lowest link cost from node 3 to node 4 should be selected, because the multiplex segment ring ⁇ 2, 3, 4, 5, 6 ⁇
  • the cost of each link on the link is smaller than the cost of each link of the multiplex segment ring ⁇ 3, 4, 8, 7 ⁇ , so node 3 on the multiplex segment ring ⁇ 2, 3, 4, 5, 6 ⁇ should be selected

Landscapes

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

Description

一种在智能光网络中获得具有保护实体路径的方法 技术领域 本发明涉及智能光网络, 具体涉及一种在智能光网络中获得具有保 护实体路径的方法。 发明背景
目前的传输网络主要是基于同步数字体系 (SDH ) /同步光网络 ( SONET )技术的, 在实际的设备运营中, 端到端的业务配置主要是根 据客户需要手工配置的。 为了提高传输网络的可运营性, 国际上提出了 智能光网络的概念, 其主要功能就是支持业务的端到端自动配置。 为了 在传输网络上提供业务自动配置能力, 智能光网络设备必须具有端到端 的优化业务路径自动计算能力。 原有的传输网络本身具有完善的保护能 力, 比如传输网络提供了基于自动保护倒换协议的复用段保护环的保护 能力, 智能光网络设备需要能够兼容原有传输网络保护的能力。
选路是智能光网络的核心, 因为自动光网络就是需要增加光网的智 能, 使点到点的光通道成为有弹性的可管理的光网。 在目前的路径计算 方法中主要使用的是流量工程技术中的约束最短路径优先计算( CSPF ) 算法, CSPF是一种改进的最短路径优先算法, 它是一种在计算通过网络 的最短路径时, 将特定的约束也考虑进去的算法。 此算法的基本思路是 在没有保护实体的网络中, 通过使用基于约束的最短路径算法计算出端 到端的最优路径, 其考虑的约束主要包括: 带宽、 管理组、 禁止节点等。
CSPF算法的基本原理是这样: CSPF算法使用名称分别为最短路径 树(PATHS )和潜在下一跳集合(TENT )的两个数据库, 其中, PATHS 中保存了最短路径树的信息, 而 TENT中包含了在找到最短路径之前的 尝试节点的信息。 在计算的初始状态, PATHS和 TENT都为空。 对于某 个源节点, 仅当找到了到达一个节点的最短路径时, 才将该节点放入 PATHS数据库中。
CSPF算法的具体实现步骤如下:
步骤 1 : 将源节点放入 TENT中, 并将源节点的路径代价设置为 0。 步骤 2: 判断当前 TENT是否为空, 如果是, 则找不到到达目的节 点的最短路径, 路由计算结束; 否则, 在 TENT中选取相对于源节点的 路径代价最小的节点放入 PATHS中。
其中, 路径代价就是路径经过的所有链路的链路代价之和, 而链路 代价是根据链路的可用带宽、 链路的长度等属性计算出来的。 下文中, 路径代价均是指从源节点到达某一节点的路径代价。 这里, 一条链路由 两个相邻节点构成, 路径由一条或一条以上的链路构成, 一条链路构成 的路径为单跳路径 , 一条以上链路构成的路径为多跳路径。
步骤 3: 判断刚加入 PATHS中的节点是否为目的节点, 如果是, 则 找到最短路径, 输出计算成功结果, 路由计算结束; 否则执行步骤 4。
步骤 4: 当某个节点被放入 PATHS时, 检查该节点到各个邻居节点 的链路, 对链路的属性与约束条件进行匹配, 并对邻居节点进行约束条 件的匹配。
其中, 对链路的属性进行匹配的约束条件内容为: 链路是否被排除, 链路的可用带宽是否满足指定的带宽要求, 链路的保护类型是否满足指 定的保护要求。 对邻居节点进行匹配的约束条件内容为: 节点是否被排 除。 链路或节点是否被排除属于排除资源类型的约束, 是指用户可以指 定业务路径不经过某些节点或某些链路。
如果约束匹配不成功, 则忽略该链路。 如果约束匹配成功, 则计算 通过所述链路的到达邻居节点的路径代价, 并再看匹配成功的邻居节点 所处的状态, 如果该邻居节点既不在 PATHS中也不在 TENT中, 则将该 邻居节点放入 TENT中; 如果该邻居节点已经存在于 PATHS中, 则表示 新路径是一条较长的路径, 忽略新路径;
如果该邻居节点存在于 TENT中, 则表示存在到达该邻居节点的新 路径, 然后, 比较新路径与 TENT中旧路径的路径代价大小, 如果新路 径的路径代价小于旧路径的路径代价, 则用新路径代替旧路径, 如果新 路径的路径代价大于或等于旧路径的路径代价, 则忽略新路径, 保留旧 路径; 然后返回步骤 2。
步骤 4中, 所述计算通过所述链路到达邻居节点的路径代价的方法 为: 到达邻居节点的路径代价等于到达前一节点的路径代价加上所述链 路的代价。
在传输网络中, 路径如果经过了保护实体, 比如复用段保护环, 在 保护实体中的链路需要保持时隙的一致性, 也就是说, 路径上连续的且 属于同一保护环实体的两条或多条链路必须具有相同通道号的时隙供业 务使用, 这样才能保证业务在保护实体中得到保护。 而现有的路径计算 方法中, 没有考虑网络中保护实体对流量工程(TE )链路选取的特殊要 求。 发明内容
有鉴于此, 本发明的主要目的在于提供一种在智能光网络中获得具 有保护实体路径的方法, 使其能在一个存在保护实体的网络中获得满足 用户要求的最优业务路径。
为达到上述目的, 本发明的技术方案是这样实现的:
一种在智能光网络中获得具有保护实体路径的方法, 其特征在于, 该方法包括以下步骤: a. 为不同保护类型的数据链路建立不同的流量工程(TE )链路, 将 具有相同保护属性且属于相同保护实体中的数据链路建立为同一 TE链 路;
b. 对所建立的 TE链路的保护实体信息进行洪泛, 扩散整个路由区 域内的保护实体信息;
c 根据洪泛后的保护实体信息, 获得所述网络中具有保护实体的路 径。
可选地, 步骤 a中所述 TE链路的建立为: 按所述智能光网络中时 隙保护类型的不同属性建立不同的 TE链路。
优选地, 按所述智能光网络中时隙保护类型的不同属性建立不同的 TE链路具体包括:将用于工作业务且有对应的保护链路对其进行保护的 TE链路作为被保护的 TE链路;将用于保护所述被保护的 TE链路的 TE 链路作为保护的 TE链路; 将用于工作业务且没有对应的保护链路对其 进行保护的 TE链路作为无保护的 TE链路。
上述方案中, 步骤 b所述对保护实体信息进行洪泛, 扩散整个路由 区域内的保护实体信息具体包括:
bl. 进行一种或一种以上不同 TE链路类型的链路状态发布(LSA ), 所述 LSA中带有所述 TE链路的保护类型和时隙占用信息以及所述 TE 链路所属保护环信息;
bl. 根据 LSA中携带的 TE链路的保护类型和时隙占用信息以及 TE 链路所属保护环信息获得整个路由区域内的保护实体信息。
具体地, 所述 TE链路所属保护环信息包括: 每个保护环实体经过 所述 TE链路时的本链路的同环入接口索引。
优选地, 所述步骤 c具体包括:
c 1. 获取所述网络路径的源节点放入潜在下一跳集合中; c2. 从所述潜在下一跳集合中选取路径代价最小的节点放入最短路 径树中, 判断该节点是否为目的节点, 如果是, 则输出路径计算结果, 路由计算结束; 否则, 根据约束条件和时隙一致性匹配 TE链路, 并根 据该路径代价最小的节点的邻居节点所处的状态, 将所述路径代价最小 节点的邻居节点放入潜在下一跳集合中;
c3. 重复上述步骤 c2, 直到所述路径代价最小节点为目的节点。 步骤 c2中, 所述根据约束条件和时隙一致性匹配 TE链路包括: 判断所述路径代价最小节点到达其邻居节点的 TE链路是否匹配约 束奈件, 如果不匹配, 则忽略该 TE链路; 如果匹配, 则检验该 TE链路 是否满足与前一条链路的时隙一致性, 如果满足, 则选取该 TE链路, 计算通过该 TE链路到达所述路径代价最 'j、节点的邻居节点的路径代价, 否则忽略该 TE链路。
其中, 所述检验该 TE链路是否满足与前一奈链路的时隙一致性的 方法为: 判断到达该邻居节点的 TE链路与路径上前一条链路是否属于 同一保护环实体, 如果不属于同一保护环实体, 则满足时隙一致性; 如果属于同一保护环实体, 则查找出路径上与所述 TE链路属于同 一保护环实体的所有 TE链路,然后判断查到的所有 TE链路上是否具有 相同通道号的时隙供使用, 如果具有相同通道号的时隙供使用, 则满足 时隙一致性; 否则不满足时隙一致性。
优选地, 所述计算到达邻居节点的路径代价包括:
判断所述 TE链路与路径上的前一条 TE链路是否属于同一保护环实 体, 如果属于同一保护环实体, 则增大所述 TE链路的代价, 如果不属 于同一保护环实体, 则保持所述 TE链路的代价不变; 然后计算到达邻 居节点的路径代价。
步骤 c2中, 所述将所述邻居节点放入潜在下一跳集合具体包括: 如果所述邻居节点既不在所述潜在下一跳集合中, 也不在所述最短 路径树中, 则将所述邻居节点放入所述潜在下一跳集合中;
如果所述邻居节点在所述最短路径树中, 则忽略该邻居节点; 如果所述邻居节点在所述潜在下一跳集合中, 则 居到达所述邻居 节点新、 旧路径大小以及新、 旧链路的节点状态确定到达所述邻居节点 的路径。
其中, 所述根据到达所述邻居节点新、 旧路径以及新、 旧链路确定 到达所述邻居节点的路径的具体为:
判断所述的新旧两条链路是否属于相同的节点, 如果属于相同的节 点, 则通过路径上的前一条链路从所述的新旧两条链路中选取链路; 如 果所述的新旧两条链路属于不同的节点, 则比较新路径与潜在下一跳集 合中的旧路径的路径代价大小, 如果新路径的路径代价小于旧路径的路 径代价, 则用新路径代替旧路径; 如果新路径的路径代价大于或等于旧 路径的路径代价, 则忽略该新路径。
上述方案中,所述通过路径上的前一条链路选取链路的方法具体为: 如果新链路与路径上的前一条链路属于同一保护环, 则选取新链路, 并 用新路径代替旧路径; 如果旧链路与路径上的前一条链路属于同一保护 环, 则选取旧链路, 忽略新路径; 否则, 如果不能确定选取哪条链路, 则对所述链路到达的邻居节点设置链路未确定标志, 并在该邻居节点上 记录所述的多条 TE链路的检索信息和对应的路径代价。
相应地, 步驟 c进一步包括:
c4. 所述最短路径树构建成功后, 从所述最短路径树的目的节点回 溯, 通过后一条链路信息确定前一条链路。
所述步骤 c4具体包括:
从所述最短路径树的目的节点向前检查路径上的每一个节点, 如果 节点上没有设置链路未确定标志, 则到达所述节点的链路已经确定; 如 果节点上设置了链路未确定标志, 则从记录在所述节点上的多奈 TE链 路中选取与路径上后一条链路属于同一保护环的 TE链路; 如果不能选 出与路径上后一条链路属于同一保护环的 TE链路, 则选择路径代价最 小的路径对应的 TE链路。
利用本发明能够基于传输网络的固有保护类型, 获得满足约束条件 的最短路径, 并且通过预先获得保护拓朴的方法, 有效地减少重复计算 的次数, 提高了网络效率。 附图简要说明
图 1是本发明实施例在智能光网络中获得具有保护实体路径的方法 的流程图;
图 2是一个二纤双向复用段保护下的链路通道保护示意图; 图 3是本发明实施例用本链路的同环入接口索引为保护环信息的复 用段环的链路示意图;
图 4是一个智能光网络拓朴示意图;
图 5、 图 6、 图 7、 图 8、 图 9、 图 10分别是图 4所示智能光网络拓 朴结构构建最短路径树过程中的路径的不同进程示意图;
图 11是图 4所示的智能光网络拓朴结构构建最短路径树过程中通过 回溯机制确定选取保护实体中链路的进程示意图。 实施本发明的方式
为使本发明的目的、 技术方案和优点表达得更加清楚明白, 下面结 合附图及具体实施例对本发明再作进一步详细的说明。
本发明的主要思想是: 对不同保护类型的数据链路建立不同的流量 工程(TE )链路, 且将具有相同保护属性、 属于相同保护实体的数据链 路建立成同一 TE链路, 然后对所建立的 TE链路进行洪泛, 进而获取所 述网络中具有保护实体的路径。
其中, 所述 TE链路是由有相同 TE属性的数据链路捆绑而成的, 在 智能光网络中, 数据链路可以理解为最小带宽颗粒度, 可以是 VC-4、 VC-12。 一条 TE链路可以是一根光纤、 多根光纤、 或者只是一根光纤中 的部分时隙, 是一个逻辑上的概念。 所述保护实体是指: 在传输网络中, 为了实现对业务的保护功能所建立的、 特殊的网络拓朴实体, 如线性复 用段、 复用段保护环等。
图 1给出了本发明一实施例在智能光网络中获得具有保护实体路径 的方法的流程图, 参照图 1所示, 本发明的方法包括以下步骤:
步骤 101 : 为不同保护类型的数据链路建立不同的 TE链路, 那么, 具有相同保护属性并且属于相同保护实体中的数据链路就建立成同一 TE链路。 具体可将 TE链路的划分和建立分为以下几类进行:
a )被保护的 TE链路: 用于工作业务的 TE链路, 有对应的保护链 路对其进行保护;
b )保护的 TE链路: 用于保护所述被保护 TE链路的 TE链路, 可 以传输额外业务, 但在实现保护时是可以被占用的;
c )无保护的 TE链路: 用于工作业务的 TE链路, 但没有保护链路 对其进行保护, 而且不能被任何其它业务抢占。
一般, 一根光纤中的时隙具有相同的属性, 就建立成一条 TE链路 进行洪泛。 所谓洪泛, 是指在路由区域内, 扩散某一链路状态, 以分布 和同步路由器之间的链路状态数据库。 但以下的情况需要特殊处理:
1 )如果一根光纤链路中的某些时隙是保护实体中的, 某些时隙不是 保护实体中的, 也就是说时隙的保护类型属性不同。 在这种情况下, 需 要把同一根光纤链路中保护类型不同的时隙建立成不同的 TE链路进行 洪泛, 保证每条 TE链路的属性一致。 例如: 二纤双向复用段保护环的 光纤中, 一半时隙用于工作, 一半时隙用于保护。 这就需要把一根光纤 链路建立成二条 TE链路进行洪泛。
2 )如果一根光纤链路中的所有时隙都是属于保护实体中的, 具有相 同的保护类型, 但是属于不同的保护实体中。 例如: 一根光纤中一部分 时隙属于某个线性复用段中、一部分时隙属于 MSP_1环中, 一部分时隙 属于 MSP— 2环中。 这种情况下, 虽然所有时隙的保护类型都是 1 : 1, 但是不能建立一条 TE链路。 需要为处于不同保护实体中的时隙分别建 立不同的 TE链路, 也就是说要为上述一根光纤建立三条 TE链路信息。
. 下面通过一个具体示例对上述 TE链路的划分与建立原则作进一步 说明:
图 2是一个二纤双向复用段保护下的链路通道保护示意图,图 2中, 用斜线填充的通道为被保护通道, 用点填充的通道为保护通道, 没有填 充的通道为无保护通道。 参照图 2所示, 16个通道的第 1-3、 5、 8 通道 是被保护的通道; 9-11、 13、 16通道是保护通道, 按顺序对应保护前面 的被保护通道, 这些保护通道上可以承载额外业务, 保护通道上所承载 的业务在该双向共享复用段失效时将被抢占; 其他通道, 即: 通道 4、 6、 7、 12、 14、 15 则是无保护通道, 在这些无保护通道上可以承载无保护 业务。 那么, 按照以上的 TE链路划分规则, 通道 {1, 2, 3 , 5 , 8}捆绑 成一条 TE链路, 保护属性是 1 : 1 ; 通道 {9, 10, 11 , 13 , 16}捆绑成一 条 TE链路,保护属性是保护链路, 可以承载额外业务; 其他通道 { 4、 6、 7、 12、 14、 15}捆绑成一条 TE链路, 保护属性为无保护, 且业务不能 被抢占。
步骤 102: 进行保护实体信息的洪泛, 扩散整个路由区域内的保护 实体信息。
这里, 保护实体的信息扩散是通过路由协议, 如 OSPF、 IS-IS等协 议的洪泛过程实现的, 保护实体的信息主要包含在 TE 链路类型 ( TE-LINK )的链路状态发布(LSA )。 因此, 本实施例中, 保护实体的 信息是通过 TE链路类型的 LSA进行洪泛的,下面对 TE链路类型的 LSA 进行详细说明:
TE链路类型的 LSA是一种不透明( Opaque ) LSA, 用于洪泛 TE链 路信息。 在该 LSA中包括 TE链路的接口索引、 带宽信息、 保护类型、 时隙的占用情况、 所属保护环信息等。 与保护实体相关的信息为: TE链 路的保护类型、 时隙占用情况和所属保护环信息。 其中, 保护类型有 1 + 1保护、 1 : 1保护, 或者无保护, 例如: 线性复用段、 复用段环都是 1 : 1保护类型。 TE链路的时隙占用情况可以使用 8个字节表示, 8个字 节代表 64个时隙, 每个字节的每一位表示 1个时隙, 并且, 使用 0/1表 示该时隙是否可用, 8个字节最大可以表示 10G带宽的 TE链路。
TE链路所属保护环信息用于标识所述的 TE链路属于哪个保护环。 网络中的某个设备可能属于多个不同的保护环, 也就是说多个保护环的 链路通过该设备。 通过 TE链路所属保护环信息, 就可以在给定某个保 护环上的一条 TE链路时, 查找出该保护环上的其他 TE链路。 本实施例 中, TE链路所属保护环信息包含一个 TE链路属性, 其定义为: 与所述 TE链路属于同一保护环且具有相同保护类型但方向相反的 TE链路的本 地接口索引。 下文中, 将该 TE链路属性简称为 TE链路的的同环入接口 索引。
例如, 图 3所示的保护环为由 4个网元 NE1、 NE2、 NE3和 NE4组 成的复用段环, 其中各条链路的属性如图 3和表 1所示。 可以以 "网元, 链路的本地接口索引 -链路的远端接口索引" 的形式表示链路, 例如, NEl,0x6cl - 0x6c3 , 则该链路的同环入接口索引为 0x6c2。 如果该链路 不是环上链路, 则该链路的同环入接口索引为 0。
Figure imgf000013_0002
Figure imgf000013_0001
另夕卜, TE链路所属保护环信息也可以通过给某一复用段环指定唯一 的环 ID来实现。
在进行了以上的 TE链路的保护类型、 时隙占用情况和所属保护环 信息的洪泛后, 在整个路由区域内各个智能设备上就能知道整个区域内 的保护实体信息, 即获得了进行路由计算的数据基础。
下面步骤 103 ~ 115就是本发明根据洪泛后的保护实体信息,获得所 述网络中具有保护实体的路径的路由计算过程。
步骤 103: 获取网络路径的源节点、 目的节点, 将源节点放入潜在 下一跳集合中, 并将源节点的路径代价设置为 0。
步骤 104 - 106: 判断当前潜在下一跳集合是否为空, 如果是, 则找 不到所述具有保护实体的最短路径, 返回错误信息, 路由计算结束; 否 则,从潜在下一跳集合中选取相对于源节点的路径代价最小的节点 M放 入 PATHS中。
步骤 107: 判断放入 PATHS中的节点 M是否为目的节点,如果节点 M不是目的节点, 则执行步骤 108, 否则执行步骤 114。
步骤 108 - 111 :查找加入 PATHS中的节点 M到达其邻居节点的 TE 链路, 判断查找到的 TE链路是否匹配约束条件, 如果不匹配, 则忽略 该链路; 如果匹配, 则进一步判断节点 M到达邻居节点的 TE链路与路 径上的前面链路是否满足时隙一致性, 如果不满足时隙一致性, 则忽略 该链路, 否则执行步骤 112。
其中, 约束条件包括保护类型、 带宽等。 时隙一致性是指在路径中 的某段链路或全部链路是同一环的链路时,必须使用相同时隙建立交叉, 才能保证在路径中节点失效的情况下业务得到复用段的保护。 因此, 需 要检查链路的时隙一致性。
可以通过以下方法检查到达邻居节点的 TE链路与路径上的前面链 路是否满足时隙一致性, 具体包括: 判断到达该邻居节点的 TE链路与 路径上前一条链路是否属于同一保护环实体, 如果所述的两条 TE链路 不属于同一保护环实体, 则不考虑所述的两条 TE链路的时隙一致性, 并认为满足时隙一致性; 如果所述的两条 TE链路属于同一保护环实体, 则查找出路径上与所述 TE链路属于同一保护环实体的所有 TE链路, 包 括所述的两条 TE链路,然后判断所有 TE链路上是否具有相同通道号的 时隙供使用, 如果具有相同通道号的时隙供使用, 则满足时隙一致性; 否则, 不满足时隙一致性。
步骤 112 ~ 113: 计算通过该 TE链路到达节点 M的邻居节点的路径 代价, 然后, 根据该节点 M的邻居节点所处的状态, 将该节点 M的邻 居节点放入潜在下一跳集合中, 然后返回步骤 104。
优选地, 本发明计算由源节点到达邻居节点的路径代价时, 采用增 大非同环链路的链路代价的方法来使同环链路节点的路径代价最小, 从 而达到在下次从潜在下一跳集合中选取路径代价最小的节点时, 优先选 择同环链路节点的目的。
增大非同环链路的链路代价的路径代价计算方法具体可以为: 如果到达邻居节点的链路与到达前一节点的链路位于同一个复用段 环, 则: 到达邻居节点的路径代价 =到达前一节点的路径代价 +到达 邻居节点的链路的链路代价;否则到达邻居节点的路径代价 =到达前一 节点的路径费用 +到达邻居节点的链路的链路代价 + K。其中, K用于 增大非同环链路的链路代价, 是一个足够大的常数, 比如 500000。
也可以减小同环链路的链路代价, 比如固定为 1 , 来使同环链路节 点的路径代价最小, 其计算方法具体可以为:
如果到达邻居节点的链路与到达前一节点的链路位于同一个复用段 环, 则: 到达邻居节点的路径代价 =到达前一节点的路径代价 + 1 ; 否 则到达邻居节点的路径代价 =到达前一节点的路径代价 +到达邻居节 点的链路的链路代价。
上述减小同环链路的链路代价的路径代价计算方法, 如果网络中存 在多条满足条件的路径, 且路径上的链路都是同环链路时, 有时可能会 选择第一奈链路的代价最小的路径, 而不是选路径代价最小的路径。 因 此, 在实际应用中, 需要根据实际网络结构来选择这种减小同环链路的 链路代价的路径代价计算方法。
在将该节点 M的邻居节点放入潜在下一跳集合的过程中, 需要遵从 以下原则: 如果该邻居节点既不在 PATHS中也不在 TENT中, 则将该邻 居节点放入 TENT中; 如果该邻居节点已经存在于 PATHS中, 则表示新 路径是一条较长的路径, 忽略该新路径, 也就是不将该邻居节点放入潜 在下一跳集合中;
如果该邻居节点存在于 TENT中, 则表示存在到达该邻居节点的新 路径, 也表示存在至少两条到达该邻居节点的 TE链路, 然后, 判断所 述的新旧两条链路是否属于相同的节点, 即组成链路的节点相同, 如果 属于相同的节点, 则通过路径上的前一条链路从所述的新旧两条链路中 选取链路; 如果所述的新旧两条链路属于不同的节点, 则比较新路径与
TENT 中的旧路径的路径代价大小, 如果新路径的路径代价小于旧路径 的路径代价, 则用新路径代替旧路径; 如果新路径的路径代价大于或等 于旧路径的路径代价, 则忽略该新路径。
其中, 通过路径上的前一条链路从所述的新旧两奈链路中选取链路 的方法具体可以为:如果新链路与路径上的前一条链路属于同一保护环, 则选取新链路, 并用新路径代替旧路径; 如果旧链路与路径上的前一条 链路属于同一保护环, 则选取旧链路, 忽略新路径; 否则, 如果不能确 定选取哪条链路, 则对所述链路到达的邻居节点设置链路未确定标志, 并在该邻居节点上记录所述的多奈 TE链路的检索信息和对应的路径代 价。
步骤 114: 最小路径构建成功, 从路径的目的节点回溯, 通过后一 条 TE链路信息确定前一条 TE链路的时隙一致性。
具体的回溯方法可以是: 从路径的目的节点开始向前检查路径上的 每一个节点, 如果节点上没有设置链路未确定标志, 则到达所述节点的 链路已经确定; 如果节点上设置了链路未确定标志, 则从记录在所述节 点上的多条 TE链路中选取与路径上后一条链路属于同一保护环的 TE链 路; 如果不能选择出与路径上后一条链路属于同一保护环的 TE链路, 则选择路径代价最小的路径对应的 TE链路。 比如, 如果回溯到节点 N 时, 发现节点 N上设置了链路未确定标志, 则通过 N节点到 N+1节点 的链路确定 N-1节点到 N节点的链路及时隙。
步骤 115: 输出路径计算结果, 路由计算结束。
为使本领域技术人员更好地理解本发明, 下面参照图 4举例说明本 发明方法中最短路径树的构建过程。
图 4描绘了一个智能光网络拓朴示意图, 图中 1 ~ 9表示九个节点, 其中, {2, 3 , 4, 5 , 6}构成一个复用段环, 链路代价均为 160; {3 , 4, 8, 7}构成另一个复用段环, 链路代价均为 320; 节点 1到节点 2、 节点 4到节点 9是线性复用段, 链路代价均为 40。
如果要建立从节点 1到节点 8之间保护类型为 1 : 1的路由, 则从节 点 1到节点 8的最短路径树的构建过程可通过图 5至图 10详细说明。图 中包括两个部分: 最短路径树和潜在下一跳集合, 位于左侧的最短路径 树使用二叉树的表示方法, 其中, P表示父指针, C表示子指针, S表示 兄弟指针。 位于右侧的潜在下一跳集合包含用于寻找最短路径的尝试节 点。 由于节点 1的为源节点, 因此节点 1的路径代价为 0。
如图 5所示, 将路径代价最小的节点 1放入最短路径树中, 同时选 取邻居节点 2放入潜在下一跳集合中。 因为从节点 1到节点 2的链路是 复用段, 符合 1 : 1的保护约束奈件。 节点 1到节点 2的路径代价为 40。
如图 6所示, 从潜在下一跳集合中选取路径代价最小的节点 2加入 最短路径树中, 同时选取节点 2的邻居节点放入潜在下一跳集合中。 因 为到达节点 2的路径不是保护环中的路径, 所以在选取链路时就没有时 隙一致性的要求, 那么, 就将节点 3和节点 6放入潜在下一跳集合中。 需要注意的是: 此时到达节点 3、 节点 6的链路是保护环上的链路。 由 于节点 2到节点 3或节点 6的链路与节点 1到节点 2的链路不是同一保 护环上的链路, 因此, 需要增加节点 2到节点 3或节点 6的链路代价, 设不同环链路的代价增量为 500000, 则节点 1到节点 3或节点 6的路径 代价均为 500200。
如图 7所示, 从潜在下一跳集合中选取路径代价最小的节点 6放入 最短路径树中, 同时选取节点 6的邻居节点放入潜在下一跳集合中。 因 为到达节点 6的路径是保护环中的路径, 所以需要在选取链路时有时隙 一致性的要求, 因为到达节点 5的链路与到达节点 6的链路都属于同一 个保护环上的链路, 所以选取相同的时隙, 并将节点 5放入潜在下一跳 集合中。 节点 1到节点 5的路径代价为 500360。
如图 8所示, 从潜在下一跳集合中选取路径代价最小的节点 3加入 最短路径树中, 同时选取节点 3的邻居节点放入潜在下一跳集合中。 因 为到达节点 3的路径是保护环中的路径, 所以需要在选取链路时有时隙 一致性的要求。 由于节点 3到达邻居节点 4有两条链路, 优先选择与到 达节点 3属于同一个保护环上的链路, 并选取链路中相同的时隙, 且将 节点 4和节点 7放入潜在下一跳集合中。 节点 1到节点 4的路径代价为 500360。 由于节点 3到节点 7的链路与节点 2到节点 3的链路不是同一 保护环上的链路, 因此, 需要增加节点 3到节点 7的链路代价, 则节点 1到节点 7的路径代价为 1000520。
如图 9所示, 从潜在下一跳集合中选取路径代价最小的节点 5放入 最短路径树中, 同时选取节点 5的邻居节点放入潜在下一跳集合中。 因 为节点 4已经在潜在下一跳集合中了, 且经过节点 5到达节点 4的新路 径比旧路径长, 因此忽略该新路径。
然后, 从潜在下一跳集合中选取路径代价最小的节点 4放入最小路 径树中, 同时选取节点 4的邻居节点放入潜在下一跳集合中。 因为节点 5 已经在潜在下一跳集合中, 且旧路径短于从节点 4到达的路径, 所以 维持旧路径不变。 因为到达部居节点 8或节点 9的链路与到达节点 4的 链路不在同一个保护环中, 所以没有时隙一致性的要求, 将节点 8、 节 点 9放入潜在下一跳集合中。 节点 1到节点 8的路径代价为 1000680。 节点 1到节点 9的路径代价为 1000400。
如图 10所示,从潜在下一跳集合中选取路径代价最小的节点 9放入 最短路径树中, 同时选取节点 9的邻居节点放入潜在下一跳集合中。 由 于节点 4已经放入最短路径树中了, 因此节点 9的邻居节点 4不可以加 入到潜在下一跳集合中。
然后, 从潜在下一跳集合中选取路径代价最小的节点 7放入最短路 径树中, 同时选取节点 7的邻居节点放入潜在下一跳集合中。 由于节点 7的邻居节点 8已经在潜在下一跳集合中, 且经过节点 7到达节点 8的 新路径比旧路径长, 因而保留旧路径。
然后, 从潜在下一跳集合中选取路径代价最小的节点 8加入最短路 径树中, 由于节点 8为目的节点, 则到达目的节点后计算结束。
下面再参照图 4对不能确定选取哪奈链路的情况进行举例说明。 例如, 在实际选择路径过程中, 可能存在这样一种情况: 如果从一 个节点 A到邻居节点 B有两条以上符合条件的链路,并且这些链路属于 不同的保护环中。 这时需要根据到达节点 A的路径确定选取哪条链路到 达节点 B, 但到达节点 A的链路不是任何环上的链路, 比如是线性复用 段, 此时并不对节点 A到节点 B的链路进行具体的选择, 而是等找到目 的节点后, 通过从目的节点回溯, 再确定具体的链路选择, 这样就能保 证选择到同一个保护环上的链路, 并确保时隙一致性, 图 11描述了此种 情况。
参照图 11所示, 图 11是图 4所示的智能光网络拓朴结构构建最短 路径树过程中, 通过回溯机制确定选取保护实体中链路的进程示意图。 这里, 要计算的是: 从节点 3到节点 5的 1保护的最短路径。
首先选择节点 3放到最短路径树中, 同时确定节点 3的邻居节点放 入潜在下一跳集合中,.其中, 节点 2和节点 7比较容易确定链路, 因为 到达节点 2、 节点 7只有一条链路。 但是, 到达节点 4有两条链路, 分 别属于两个保护环上的链路, 而且均符合约束条件, 由于不存在前一条 链路, 不能通过前一条链路来选择同环的链路, 因此不能确定选取哪条 链路。 由于从节点 3到节点 4的链路暂时不能确定, 因此在该节点 4上 作一个链路未确定的标记, 然后继续构建最短路径树。
最短路径树构建完成后, 从目的节点 5进行回溯。 回溯到节点 4, 发现节点 4中设置有链路未确定的标记, 那么, 就利用从节点 4到节点 5的路径确定从节点 3到节点 4选择哪个环上的链路, 即: 选取与从节 点 4到节点 5属于同一保护环的、 即复用段环 {2, 3 , 4, 5, 6}上从节点 3到节点 4的链路。 如果节点 3到节点 4的路径选取错误, 当节点 4失 效时, 业务就没有办法得到保护。
再如, 如果需要计算节点 3到节点 9的 1 : 1保护最短路径, 当节点 3加入到 PATHS中时, 由于不存在前一链路, 不能通过前一链路确定节 点 3到节点 4的链路。 因此, 需要在节点 4上作一个链路未确定标记, 在进行回溯处理时, 由于从节点 9到节点 4的链路是线性复用段链路, 不能从后一条链路确定前一链路, 因此, 应该选择路径代价较小的路径 所对应的 TE链路, 即应该选取节点 3到节点 4的链路代价最小的路径, 由于复用段环 {2, 3 , 4, 5, 6}上的各链路代价均比复用段环 {3, 4, 8, 7}的各链路代价小, 因此应该选择复用段环 {2, 3 , 4, 5, 6}上的节点 3 到节点 4的链路。
以上所述, 仅为本发明的较佳实施例而已, 本领域普通技术人员都 知道, 本发明有许多变形和变化而不脱离本发明的精神, 上述具体实施 例并非用于限定本发明的保护范围。

Claims

权利要求书
1、一种在智能光网络中获得具有保护实体路径的方法,其特征在于, 该方法包括以下步骤:
a. 为不同保护类型的数据链路建立不同的流量工程 TE链路, 将具 有相同保护属性且属于相同保护实体中的数据链路建立为同一 TE链路; b. 对所建立的 TE链路的保护实体信息进行洪泛, 扩散整个路由区 域内的保护实体信息;
c根据洪泛后的保护实体信息, 获得所述网络中具有保护实体的路 径。
2、 如权利要求 1所述的方法, 其特征在于, 步骤 a中所述 TE链路 的建立为:按所述智能光网络中时隙保护类型的不同属性建立不同的 TE 链路。
3、 如权利要求 2所述的方法, 其特征在于, 按所述智能光网络中时 隙保护类型的不同属性建立不同的 TE链路具体包括:
将用于工作业务且有对应的保护链路对其进行保护的 TE链路作为 被保护的 TE链路;
将用于保护所;
将用于工作业务且没有对应的保护链路对其进行保护的 TE链路作 为无保护的 TE链路。
4、 如权利要求 1所述的方法, 其特征在于, 步骤 b所述对保护实体 信息进行洪泛, 扩散整个路由区域内的保护实体信息具体包括:
bl. 进行一种或一种以上不同 TE链路类型的链路状态发布 LSA,所 述 LSA中带有所述 TE链路的保护类型和时隙占用信息以及该 TE链路 所属保护环信息; b2.根据 LSA中携带的 TE链路的保护类型和时隙占用信息以及该 TE链路所属保护环信息获得整个路由区域内的保护实体信息。
5、 如权利要求 4所述的方法, 其特征在于, 所述 TE链路所属保护 环信息包括: 每个保护环实体经过所述 TE链路时的本链路的同环入接 口索引。
6、 如权利要求 1所述的方法, 其特征在于, 所述步骤 c具体包括: cl. 获取所述网络路径的源节点放入潜在下一跳集合中;
c2. 从所述潜在下一跳集合中选取路径代价最小的节点放入最短路 径树中, 判断该节点是否为目的节点, 如果是, 则输出路径计算结果, 路由计算结束; 否则, 根据约束条件和时隙一致性匹配 TE链路, 并根 据该路径代价最小的节点的邻居节点所处的状态, 将所述路径代价最小 节点的邻居节点放入潜在下一跳集合中;
c3. 重复上述步骤 c2, 直到所述路径代价最小节点为目的节点。
7、 如权利要求 6所述的方法, 其特征在于, 步骤 c2中, 所述根据 约束条件和时隙一致性匹配 TE链路包括:
判断所述路径代价最小节点到达其邻居节点的 TE链路是否匹配约 束条件, 如果不匹配, 则忽略该 TE链路; 如果匹配, 则检验该 TE链路 是否满足与前一条链路的时隙一致性, 如果满足, 则选取该 TE链路, 计算通过该 TE链路到达所述路径代价最小节点的邻居节点的路径代价, 否则忽略该 TE链路。
8、 如权利要求 7所述的方法, 其特征在于, 所述检脸该 TE链路是 否满足与前一条链路的时隙一致性的方法为:判断到达该邻居节点的 TE 链路与路径上前一条链路是否属于同一保护环实体, 如果不属于同一保 护环实体, 则满足时隙一致性;
如果属于同一保护环实体, 则查找出路径上与所述 TE链路属于同 一保护环实体的所有 TE链路,然后判断查到的所有 TE链路上是否具有 相同通道号的时隙供使用, 如果具有相同通道号的时隙供使用, 则满足 时隙一致性; 否则不满足时隙一致性。
9、 如权利要求 7所述的方法, 其特征在于, 所述计算到达邻居节点 的路径代价包括:
判断所述 TE链路与路径上的前一条 TE链路是否属于同一保护环实 体, 如果属于同一保护环实体, 则增大所述 TE链路的代价, 如果不属 于同一保护环实体, 则保持所述 TE链路的代价不变; 然后计算到达邻 居节点的路径代价。
10、 如权利要求 6所述的方法, 其特征在于, 步骤 c2中, 所述将所 述邻居节点放入潜在下一跳集合具体包括:
如果所述邻居节点既不在所述潜在下一跳集合中, 也不在所述最短 路径树中, 则将所述邻居节点放入所述潜在下一跳集合中;
如果所述邻居节点在所述最短路径树中, 则忽略该邻居节点; 如果所述邻居节点在所述潜在下一跳集合中, 则根据到达所述邻居 节点新、 旧路径大小以及新、 旧链路的节点状态确定到达所述邻居节点 的路径。
11、 如权利要求 10所述的方法, 其特征在于, 所述根据到达所述邻 居节点新、 旧路径以及新、 旧链路确定到达所述邻居节点的路径的具体 为:
判断所述的新旧两条链路是否属于相同的节点, 如果属于相同的节 点, 则通过路径上的前一条链路从所述的新旧两条链路中选取链路; 如 果所述的新旧两条链路属于不同的节点, 则比较新路径与潜在下一跳集 合中的旧路径的路径代价大小, 如果新路径的路径代价小于旧路径的路 径代价, 则用新路径代替旧路径; 如果新路径的路径代价大于或等于旧 路径的路径代价, 则忽略该新路径。
12、 如权利要求 11所述的方法, 其特征在于, 所述通过路径上的前 一条链路选取链路的方法具体为: 如果新链路与路径上的前一条链路属 于同一保护环, 则选取新链路, 并用新路径代替旧路径; 如果旧链路与 路径上的前一条链路属于同一保护环, 则选取旧链路, 忽略新路径; 否 则, 如果不能确定选取哪条链路, 则对所述链路到达的邻居节点设置链 路未确定标志, 并在该邻居节点上记录所述的多条 TE链路的检索信息 和对应的路径代价。
13、 如权利要求 12所述的方法, 其特征在于, 步骤 c进一步包括: c4. 所述最短路径树构建成功后, 从所述最短路径树的目的节点回 溯, 通过后一条链路信息确定前一条链路。
14、 如权利要求 13所述的方法, 其特征在于, 所述步骤 c4具体包 括:
从所述最短路径树的目的节点向前检查路径上的每一个节点, 如果 节点上没有设置链路未确定标志, 则到达所述节点的链路已经确定; 如 果节点上设置了链路未确定标志, 则从记录在所述节点上的多条 TE链 路中选取与路径上后一条链路属于同一保护环的 TE链路; 如果不能选 出与路径上后一条链路属于同一保护环的 TE链路, 则选择路径代价最 小的路径对应的 TE链路。
PCT/CN2004/001009 2003-09-02 2004-09-01 Procede pour realiser une voie avec entite de protection dans un reseau optique intelligent WO2005022823A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN03159210.4 2003-09-02
CNB031592104A CN1306738C (zh) 2003-09-02 2003-09-02 一种在智能光网络中获得具有保护实体的路径的方法

Publications (1)

Publication Number Publication Date
WO2005022823A1 true WO2005022823A1 (fr) 2005-03-10

Family

ID=34240937

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2004/001009 WO2005022823A1 (fr) 2003-09-02 2004-09-01 Procede pour realiser une voie avec entite de protection dans un reseau optique intelligent

Country Status (2)

Country Link
CN (1) CN1306738C (zh)
WO (1) WO2005022823A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103647598A (zh) * 2013-11-05 2014-03-19 国家电网公司 一种基于智能光网络的通信多等级保护方法
CN111651844A (zh) * 2019-12-04 2020-09-11 台州学院 一种用于计算两个公交站点间多条乘车路径的算法
CN111787436A (zh) * 2020-07-02 2020-10-16 中国电子科技集团公司第三十四研究所 一种光网络通信路径自动搜索规划的方法

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100382504C (zh) * 2005-07-14 2008-04-16 广东省电信有限公司研究院 自动交换光网络中主备保护的跨域端到端连接的建立方法
CN100389578C (zh) * 2006-01-25 2008-05-21 华为技术有限公司 传统光传送网和智能光网络组网时的资源分配方法
CN101155064A (zh) * 2006-09-26 2008-04-02 华为技术有限公司 流量工程链路资源信息的处理方法
CN101163090B (zh) * 2006-10-09 2010-08-04 华为技术有限公司 一种业务路径的计算方法
CN100454848C (zh) * 2007-03-07 2009-01-21 华为技术有限公司 一种网络结构拓扑布局的方法及网管设备
CN101146099B (zh) * 2007-06-12 2011-12-28 中兴通讯股份有限公司 域间路由协议流量工程链路的自动生成和维护方法
CN101145963B (zh) * 2007-07-27 2010-08-18 中兴通讯股份有限公司 一种保护子网的拓扑发现方法

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002082720A2 (en) * 2001-04-04 2002-10-17 Roke Manor Research Limited Automated network with tunnels and method of data routing
US6515966B1 (en) * 2000-05-05 2003-02-04 Fujitsu Network Communications, Inc. System and method for application object transport

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6515966B1 (en) * 2000-05-05 2003-02-04 Fujitsu Network Communications, Inc. System and method for application object transport
WO2002082720A2 (en) * 2001-04-04 2002-10-17 Roke Manor Research Limited Automated network with tunnels and method of data routing

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103647598A (zh) * 2013-11-05 2014-03-19 国家电网公司 一种基于智能光网络的通信多等级保护方法
CN103647598B (zh) * 2013-11-05 2016-01-06 国家电网公司 一种基于智能光网络的通信多等级保护方法
CN111651844A (zh) * 2019-12-04 2020-09-11 台州学院 一种用于计算两个公交站点间多条乘车路径的算法
CN111651844B (zh) * 2019-12-04 2023-04-11 台州学院 一种用于计算两个公交站点间多条乘车路径的方法
CN111787436A (zh) * 2020-07-02 2020-10-16 中国电子科技集团公司第三十四研究所 一种光网络通信路径自动搜索规划的方法

Also Published As

Publication number Publication date
CN1592160A (zh) 2005-03-09
CN1306738C (zh) 2007-03-21

Similar Documents

Publication Publication Date Title
US6728205B1 (en) Method and apparatus for automatic protection switching
US7729337B2 (en) Protocol for the determination of network topology
JP2737828B2 (ja) 通信ネットワークおよび該ネットワークでの経路選択方法
US7248561B2 (en) Path establishment method for establishing paths of different fault recovery types in a communications network
JP3700596B2 (ja) 通信ネットワーク及びパス設定方法並びにパス設定用プログラム
US8155522B2 (en) System and method for network planning
US8467288B2 (en) Method of path switching and node apparatus
KR101343596B1 (ko) 다계층 자원 전송망 경로 계산에 필요한 자원 관리 및 재귀적 경로 계산 방법 및 장치
US7830789B2 (en) Method for accommodating signal paths to ring network
WO2005022823A1 (fr) Procede pour realiser une voie avec entite de protection dans un reseau optique intelligent
CN101646105B (zh) 业务恢复的方法、系统和设备
CA2440515A1 (en) Constraint based routing with non-transitive exceptions
CN102143410B (zh) 一种光网络中的路径计算方法及路径计算单元
CN100536405C (zh) 有效使用网络资源的保护路径获取方法
CN113242176A (zh) 一种端到端多路径快速计算方法及装置
Hwang et al. Multiple shared backup cycles for survivable optical mesh networks
CN101243723B (zh) 自动交换光网络组播业务的路由选择方法
US7564859B1 (en) Method and system for optimized routing and protection of VT traffic in SONET networks
CN108337174B (zh) 一种传输网端到端业务路由的搜索方法和装置、存储介质
US7710878B1 (en) Method and system for allocating traffic demands in a ring network
Morley et al. A comparative survey of methods for automated design of ring-based transport networks
EP0939560A1 (en) Method and system for restoring a distributed telecommunications network
Zhang et al. Different implementations of token tree algorithm for DWDM network protection/restoration
Kos et al. Topological planning of communication networks
Zhang Distributed tree schemes for DWDM network protection and restoration.

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
122 Ep: pct application non-entry in european phase