CN108111409A - The method and apparatus for establishing disjoint paths - Google Patents

Abstract

The embodiment of the present application discloses a kind of method and apparatus for establishing disjoint paths, can solve the problems, such as topological trap when order calculates cross-domain disjoint paths.This method includes：High level controller determines to reach the domain sequence for at least two network domains composition that destination node needs pass through from source node；Disjoint paths group set in the domain that each network domains have in the true domain sequences of high level controller; disjoint paths group set includes disjoint paths group at least one domain in domain; disjoint paths group includes disjoint paths in two domains in each domain, and the paths in two domains in disjoint paths are that operating path and another paths are Protection path in domain in domain；High level controller is according to disjoint paths group set in the domain that each network domains have in the sequence of domain; determine at least one cross-domain disjoint paths group; each cross-domain disjoint paths group includes two cross-domain disjoint paths, in two cross-domain disjoint paths a paths are as cross-domain operating path and another paths are as cross-domain Protection path.

Description

The method and apparatus for establishing disjoint paths

Technical field

This application involves the communications field, and more particularly, to a kind of method and apparatus for calculating disjoint paths.

Background technology

Telecommunication network is usually made of the network equipment of magnanimity, and networking scale is very big.Therefore, handed over using multiprotocol label The telecommunication network of (MultiProtocol Label Switching, MPLS) technology structure is changed, it usually all can be according to certain rule Mould splits the network into multiple Autonomous Domains.Autonomous Domain be a group of labels Switch Router (Label Switching Router, LSR) the set of equipment composition.LSR equipment in one Autonomous Domain operate in a single distributed protocol example or It is managed by a central control unit.In the case where network is divided into multiple Autonomous Domains, a strip label switching path (Label switch Path, LSP) connection usually requires the Autonomous Domain through multiple MPLS.

In MPLS multiple domain network models, the reliability requirement of business is very high or equipment, optical fiber etc. break down In the case of, business needs to obtain quickly protection and recovers.For this purpose, usually it can map out the work and protect in an mpls network two Path is carrying same business.In the case that operating path breaks down, the business carried can be quickly switched into protection On path.In order to ensure the selection of this restoration path, operating path and Protection path need to follow path separation principle, will Operating path and Protection path are distributed to different nodes, different chain roads as far as possible.In order to establish follow path separation principle two Paths propose a kind of method that order calculates in the prior art.The main thought that order calculates is first to establish a job Then path resettles the separated Protection path of a paths in the case where excluding operating path.

But order computational methods do not consider Protection path at evaluation work path.It is had been calculated in operating path Cheng Houzai calculates Protection path, may result in a domain, can not calculate and the node of operating path, link disjoint Path (also referred to as topological trap), so as to cause the calculating failure of entire cross-domain disjoint paths.

The content of the invention

The application provides a kind of method and apparatus for establishing disjoint paths, topology when can solve to calculate cross-domain disjoint paths The problem of trap.

In a first aspect, this application provides a kind of method for establishing disjoint paths, logical including at least two network domains It is performed in letter system, which includes high level controller and at least two lower-level controllers.At least two low layer controls It is mapped as corresponding between device and at least two network domains.This method includes：High level controller determine from source node to The domain sequence that at least two network domains passed through up to destination node needs are formed.Each network domains include source boundary node With egress boundary node.Each disjoint paths group set in domain possessed by network domains in the true domain sequences of high level controller.It should Disjoint paths group set includes disjoint paths group at least one domain in domain.It is each in disjoint paths group at least one domain Disjoint paths group includes disjoint paths in two domains in domain, and the paths in two domains in disjoint paths are used as work in domain Path and another paths are as Protection path in domain.Wherein, each network domains have disjoint paths group at least one domain.It is high-rise Controller disjoint paths group set in domain according to possessed by each network domains in the sequence of domain, determines a cross-domain disjoint paths group Set.The cross-domain group of paths set includes at least one cross-domain disjoint paths group.Each cross-domain disjoint paths group includes each net Disjoint paths group in one domain in network domain.Each cross-domain disjoint paths group includes two cross-domain disjoint paths, this two cross-domain point From the paths in path as cross-domain operating path and another paths are as cross-domain Protection path.

It is to be understood that domain sequence is a sequence being made of in order multiple domains.From the network domains where source node, according to Each network domains in the secondary sequence by domain can finally reach the network domains where destination node.In fact, source node is the domain Node in sequence in first network domains, destination node are the node in the last one network domains in the domain sequence.

It should be noted that in the embodiment of the present application, when establishing disjoint paths group in the domain of network domains, only account for net The boundary node in network domain.The disjoint paths between node inside network domains can be calculated by the prior art.Existing skill Art is very ripe for the method for the disjoint paths between the internal node of calculating network domain.The embodiment of the present application is not made this in detail It states.

In the embodiment of the present application, high level controller is when establishing cross-domain disjoint paths, it is first determined is reached from source node Disjoint paths group in the domain of each network domains in the network domains that destination node needs pass through, then sequentially in the domain of these network domains Disjoint paths group is combined, and obtains cross-domain disjoint paths group.Since disjoint paths group includes Liang Tiaoyunei roads in each domain Footpath, one is used as operating path in domain, and another as Protection path in domain.In other words, when establishing operating path in domain In view of the foundation of Protection path in domain.So as to be wrapped in each cross-domain disjoint paths group determined according to disjoint paths group in domain Two cross-domain disjoint paths are included, wherein a paths are as cross-domain operating path and another paths are as cross-domain Protection path. Establishing the processes of cross-domain disjoint paths from the embodiment of the present application, to can be seen that cross-domain operating path and cross-domain Protection path be simultaneously It establishes.When can thus avoid order calculating disjoint paths in the prior art, resettled due to first establishing cross-domain operating path Cross-domain Protection path, after the completion of may causing cross-domain operating path calculating, the topology that cross-domain Protection path can not be calculated is fallen into The problem of trap.

In a kind of possible realization method, the first network domain at least two network domains includes a source border Node and multiple egress boundary nodes.At least two lower-level controllers include the first lower-level controllers.First low layer control Device processed is the corresponding lower-level controllers in first network domain.And each network domains that the true domain sequences of high level controller include Disjoint paths group set, specifically includes in possessed domain：High level controller receives first that the first lower-level controllers are sent and refers to Show information, and disjoint paths group set in the domain in first network domain is determined according to the instruction of the first indication information.The first network Disjoint paths group at least one domain of disjoint paths group set including first network domain in the domain in domain, in the domain in first network domain Disjoint paths group is that multiple egress boundary nodes in first network domain are combined to obtain at least two-by-two by the first lower-level controllers After one egress boundary node group, by the source boundary node in first network domain at least one egress boundary node group one by one It is combined.

It is to be understood that first network domain here can be any one network domains at least two network domains.Also, It can include one or more first network domains at least two network domains.That is, if there is portion at least two network domains The topological structure of the boundary node in subnetwork domain meet source boundary node be one and egress boundary node be it is multiple, then can be with The disjoint paths group of these network domains is built using the method provided in the present embodiment.

In a kind of possible realization method, the second network domains at least two network domains include multiple source borders Node and multiple egress boundary nodes.At least two lower-level controllers include the second lower-level controllers.Second low layer controls Device is lower-level controllers corresponding with the second network domains.And each network domains that the true domain sequences of high level controller include Disjoint paths group set, specifically includes in possessed domain：High level controller receives second that the second lower-level controllers are sent and refers to Show information, and disjoint paths group set in the domain of the second network domains is determined according to the instruction of the second indication information.Second network Disjoint paths group at least one domain of disjoint paths group set including the second network domains in the domain in domain, in the domain of the second network domains Disjoint paths group is that multiple source boundary nodes of the second network domains are combined by the second lower-level controllers two-by-two, is obtained at least One source boundary node group, and multiple egress boundary nodes of the second network domains are combined two-by-two, it obtains at least one After egress boundary node group, at least one source boundary node group and at least one egress boundary node group are carried out one by one What combination obtained.

Similar with first network domain, the second network domains here can be any one net at least two network domains Network domain.Also, one or more second network domains can be included at least two network domains.That is, if at least two net The topological structure for having the boundary node of multiple network domains in network domain meets that source boundary node is multiple and egress boundary node is It is multiple, then the method that is provided in the present embodiment may be employed to build the disjoint paths group of these network domains.

In a kind of possible realization method, the 3rd network domains at least two network domains include multiple source borders Node and an egress boundary node.At least two lower-level controllers include the 3rd lower-level controllers.3rd low layer controls Device is lower-level controllers corresponding with the 3rd network domains.And each network domains are had in the true domain sequences of high level controller Domain in disjoint paths group set, specifically include：High level controller receives the 3rd indication information that the 3rd lower-level controllers are sent, And disjoint paths group set in the domain of the 3rd network domains is determined according to the instruction of the 3rd indication information.In the domain of 3rd network domains Disjoint paths group set includes disjoint paths group at least one domain of the 3rd network domains, and road is separated in the domain of the 3rd network domains Footpath group is that multiple source boundary nodes of the 3rd network domains are combined by the 3rd lower-level controllers two-by-two, obtains at least one source After holding boundary node group, the egress boundary node of the 3rd network domains and at least one source boundary node group are subjected to group one by one Close what is obtained.

It is to be understood that the middle number " first " of the embodiment of the present application, " second " be just to distinguish different objects, for example, In order to distinguish different network domains, lower-level controllers etc..Therefore, above-mentioned first network domain, the second network domains and the 3rd network domains Only as three kinds of possible topological structures of a network domains.I.e. that is, any one network domains in communication system can Can be the topological structure in first network domain, it is also possible to be the topological structure of the second network domains, alternatively, can also be the 3rd network domains Topological structure.Based on different topological structures, lower-level controllers determine that process during disjoint paths group in domain is slightly different.Tool Body process is referring to the explanation in specification.

In a kind of possible realization method, this method further includes：High level controller calculates the cross-domain disjoint paths group collection The path cost of each cross-domain disjoint paths group in conjunction, and by the cross-domain disjoint paths group of path cost minimum be determined as target across Domain disjoint paths group.

Usually, path cost is alternatively referred to as path cost (path cost).

It is to be understood that high level controller can arbitrarily select a cross-domain disjoint paths group from cross-domain disjoint paths group set As the cross-domain disjoint paths group of target, and using the smaller paths of path cost in the cross-domain disjoint paths group of target as cross-domain Operating path, and using the larger another paths of path cost as cross-domain Protection path.

For the embodiment of the present application in the cross-domain disjoint paths group of selection target, high level controller can also be by cross-domain disjoint paths The cross-domain disjoint paths group of path Least-cost is determined as the cross-domain disjoint paths group of target in group set.Cross-domain point of target at this time It is global optimum from group of paths.

In a kind of possible realization method, high level controller calculates the path cost of each cross-domain disjoint paths group, bag It includes：High level controller obtains in the sequence of domain the path cost of disjoint paths group in each domain possessed by each network domains, wherein, In each domain of each network domains the path cost of disjoint paths group be with the lower-level controllers corresponding to each network domains according to What the weight calculation of default operating path and Protection path obtained；High level controller is according to path generation of each disjoint paths group Valency and default domain-to-domain link weight calculate the path cost of each cross-domain disjoint paths group.

Second aspect, this application provides a kind of method for establishing disjoint paths, logical including at least two network domains It is performed in letter system.The communication system includes high level controller and at least two lower-level controllers.At least two low layers control It is mapped as corresponding between device processed and at least two network domains.At least two lower-level controllers include and the first net Corresponding first lower-level controllers in network domain.First network domain is any one network domains at least two network domains.The party Method includes：First lower-level controllers determine disjoint paths group set in the domain in first network domain.Disjoint paths group set in the domain Including disjoint paths group at least one domain.It is wrapped in each domain at least one domain in disjoint paths group in disjoint paths group Include disjoint paths in two domains.A paths in two domains in disjoint paths are as operating path in domain and another paths As Protection path in domain.First lower-level controllers are used to indicate high level controller to high level controller transmission and determine first network The indication information of disjoint paths group set in the domain in domain, in order to which high level controller is according to disjoint paths in the domain in first network domain Disjoint paths group set in the domain of other network domains in group set and at least two network domains in addition to first network domain, really A fixed cross-domain disjoint paths group set.The cross-domain disjoint paths group set includes at least one cross-domain disjoint paths group.Often A cross-domain disjoint paths group includes two cross-domain disjoint paths, and the paths in two cross-domain disjoint paths are as cross-domain work Make path and another paths are as cross-domain Protection path.

In the embodiment of the present application, high level controller is when establishing cross-domain disjoint paths, it is first determined is reached from source node Disjoint paths group in the domain of each network domains in the network domains that destination node needs pass through, then sequentially in the domain of these network domains Disjoint paths group is combined, and obtains cross-domain disjoint paths group.And disjoint paths group is by low layer control in the domain of these network domains Device calculating processed indicates high level controller after determining.

When lower-level controllers determine disjoint paths in the domain of a network domains (the corresponding network domains of the lower-level controllers), always It is that disjoint paths group in the domain including two paths is calculated, the paths in this two paths are as the working ways in domain Footpath and another are as the Protection path in domain.That is, Protection path is calculated simultaneously in operating path and domain in domain.

Therefore, each cross-domain separation road that high level controller is determined according to disjoint paths group in the domains of each network domains In the group of footpath similarly include two cross-domain paths, and will a wherein paths as cross-domain operating path and another paths conduct Cross-domain Protection path.As it can be seen that cross-domain operating path and cross-domain Protection path are established simultaneously, it is thus possible to avoid the prior art When middle order calculates disjoint paths, cross-domain Protection path is resettled due to first establishing cross-domain operating path, cross-domain work may be caused After the completion of making path computing, the problem of topological trap of cross-domain Protection path can not be calculated.

In a kind of possible realization method, first network domain includes a source boundary node and multiple egress borders Node.And first lower-level controllers determine disjoint paths group set in the domain in first network domain, including：First low layer controls Multiple egress boundary nodes in first network domain are combined by device two-by-two, obtain at least one egress boundary node group；First Source boundary node and the plurality of egress boundary node group are combined by lower-level controllers one by one, obtain first network domain extremely Disjoint paths group in a few domain.

In a kind of possible realization method, first network domain includes multiple source boundary nodes and multiple egress borders Node.And first lower-level controllers determine disjoint paths group at least one domain in first network domain, including：First low layer Multiple source boundary nodes in first network domain are combined by controller two-by-two, obtain at least one source boundary node group, And be combined multiple egress boundary nodes in first network domain two-by-two, obtain at least one egress boundary node group；First At least one source boundary node group and at least one egress boundary node group are combined by lower-level controllers one by one, are obtained Disjoint paths group in at least one domain in first network domain.

In a kind of possible realization method, first network domain includes multiple source boundary nodes and an egress border Node and, the first lower-level controllers determine disjoint paths group at least one domain in first network domain, including：First low layer Multiple source boundary nodes in first network domain are combined by controller two-by-two, obtain at least one source boundary node group； First lower-level controllers by an egress boundary node in first network domain and at least one source boundary node group one by one into Row combination, obtains disjoint paths group at least one domain in first network domain.

In a kind of possible realization method, in the first lower-level controllers before high level controller sends indication information, This method further includes：First lower-level controllers calculate the path cost of disjoint paths group at least one domain in first network domain, Wherein, which is additionally operable to instruction high level controller according to disjoint paths group at least one domain in first network domain Possessed by each network domains of other network domains in path cost and at least two network domains in addition to first network domain It is cross-domain to be determined as target by the path cost of disjoint paths group at least one domain for the cross-domain disjoint paths group of path cost minimum Disjoint paths group.

In a kind of possible realization method, the first lower-level controllers are calculated at least one domain in first network domain and separated The path cost of group of paths, specifically includes：First lower-level controllers are according to the weight of default operating path and Protection path, meter Calculate at least one domain in the first network domain path cost of disjoint paths group in each domain in disjoint paths group.

It is to be understood that the first lower-level controllers calculate the path cost of disjoint paths in each domain in first network domain, and will The result for the path cost being calculated is indicated to high level controller.So as to which high level controller can be based on each network domains The path cost of disjoint paths group at least one domain, combination obtain the cross-domain disjoint paths group of path cost minimum, and should The cross-domain disjoint paths group of path cost minimum is as the cross-domain disjoint paths group of target.The cross-domain disjoint paths group of target at this time is Global optimum.Therefore, it is possible to save path cost, cost is reduced.

The third aspect, this application provides a kind of device for establishing cross-domain disjoint paths, for performing first aspect or Method in the arbitrary possible realization method of one side.Specifically, which includes performing first aspect or first aspect The unit of method in arbitrary possible realization method.

Fourth aspect, this application provides a kind of device for establishing cross-domain disjoint paths, for performing second aspect or Method in the arbitrary possible realization method of two aspects.Specifically, which includes performing second aspect or second aspect The unit of method in arbitrary possible realization method.

5th aspect, this application provides a kind of network equipment, which includes processor and memory.Memory For storing computer program, processor is used to call from memory and run computer program.It, should when program is run Processor performs the method in the arbitrary possible realization method of above-mentioned first aspect or first aspect.

6th aspect, this application provides a kind of network equipment, which includes processor and memory.Memory For storing computer program, processor is used to call from memory and run computer program.It, should when program is run Processor performs the method in the arbitrary possible realization method of above-mentioned second aspect or second aspect.

7th aspect, this application provides a kind of communication system, including the device and fourth aspect described in the third aspect The device.

Eighth aspect, this application provides a kind of communication system, including the network equipment and the 6th described in the 5th aspect The network equipment described in aspect.

9th aspect, this application provides a kind of computer-readable mediums, for storing computer program, the computer journey Sequence includes the instruction of the method in the arbitrary possible realization method for performing first aspect or first aspect.

Tenth aspect, this application provides a kind of computer-readable mediums, for storing computer program, the computer journey Sequence includes the instruction of the method in the arbitrary possible realization method for performing second aspect or second aspect.

In the embodiment of the present application, high level controller is when establishing cross-domain disjoint paths, it is first determined is reached from source node Disjoint paths group in the domain of each network domains in the network domains that destination node needs pass through, then sequentially in the domain of these network domains Disjoint paths group is combined, and obtains cross-domain disjoint paths group.Since disjoint paths group includes Liang Tiaoyunei roads in each domain Footpath, one is used as operating path in domain, and another as Protection path in domain.So as to be determined according to disjoint paths group in domain Each cross-domain disjoint paths group includes two cross-domain disjoint paths, wherein a paths are as cross-domain operating path and another Path is as cross-domain Protection path.That is, cross-domain operating path and cross-domain Protection path are established simultaneously, it is thus possible to be avoided existing Have when disjoint paths are sequentially calculated in technology, resettle cross-domain Protection path due to first establishing cross-domain operating path, may cause After the completion of cross-domain operating path calculates, the problem of topological trap of cross-domain Protection path can not be calculated.

Description of the drawings

Fig. 1 shows the Organization Chart for the MPLS network for being layered deployment in the prior art.

Fig. 2 is shown follows the separated work LSP paths in path and protection LSP paths in the prior art.

Fig. 3 is a kind of schematic diagram of network topology structure.

Fig. 4 is the schematic flow chart of the method provided by the embodiments of the present application for establishing cross-domain disjoint paths.

Fig. 5 is a kind of example of topological structure of path separation group in domain.

Fig. 6 is a kind of example of topological structure of path separation group in domain.

Fig. 7 is the centralized control type network structure disposed suitable for a kind of layering of the embodiment of the present application.

Fig. 8 is the schematic diagram of the device 400 for establishing disjoint paths of one embodiment of the application.

Fig. 9 is the schematic diagram of the device 500 for establishing disjoint paths of another embodiment of the application.

Figure 10 is the schematic diagram of the equipment 600 for establishing disjoint paths of one embodiment of the application.

Figure 11 is the schematic diagram of the equipment 700 for establishing disjoint paths of another embodiment of the application.

Specific embodiment

Below in conjunction with the accompanying drawings, the technical solution of the embodiment of the present application is illustrated.

First, to the invention relates to related notion simply introduced.

Multiprotocol label switching (Multi-Protocol Label Switching, MPLS) is to carry out data using mark The technology of forwarding.When packet message enters network, the short mark of regular length is distributed for it, and will mark and packet message It is packaged together.In entire repeating process, switching node is forwarded to according only to mark.In an mpls network, data transmission It is happened on label switched path (Label switch Path, LSP), LSP is each along the path from source to egress On node sequence label.

MPLS network generally include distributed AC servo system MPLS network and with two kinds of centerized fusion MPLS network.Telecommunication network It is usually made of the network equipment of magnanimity, either distributed or centralization, the networking scale that single example can be supported is all It is limited.So telecommunication network built using MPLS technology, it will usually be splitted the network into according to certain scale multiple Autonomous Domain.Autonomous Domain is the set of one group of LSR equipment.LSR equipment in one domain operates in a single distributed protocol It is managed in example or by a central control unit.In the case where network is divided into multiple Autonomous Domains, a LSP connects Connect the Autonomous Domain usually required through multiple MPLS.

Since the presence of multiple domain networking mode is, it is necessary to have the synergistic mechanism in multiple domains, to guarantee to create and manage across more The service connection in a domain is realized in the prior art by being layered the control unit of deployment mostly.

Fig. 1 shows the Organization Chart for the MPLS network for being layered deployment in the prior art.As shown in Figure 1, single domain controller is (such as Controller A and controller B in Fig. 1) directly and MPLS network in LSR interact, realize the centralized Control to individual domain.It is high-rise The single domain controller of controller (the controller AB in such as Fig. 1) and low layer interacts, and completes the Collaborative Control to multiple domains.

It is to be understood that controller can a point multilayer disposed.Two layers of deployment framework shown in FIG. 1, which is only used as, illustratively to be said It is bright.For example, the controller A and controller B in Fig. 1 are 0 layer of Lever, high level controller AB is 1 layer of Lever.

The telecommunication network built using MPLS technology, it is very high to the reliability requirement of business.In equipment, fiber failure In the case of, business needs to obtain quickly protection and recovers.General recovery time is required within 50ms.

In the prior art, in order to meet the needs of business fast quick-recovery, usually map out the work in an mpls network path and Protection path two LSP carry same business.And the corresponding Operations, Administration and Maintenance of use (operation, Administration and maintenance, OAM) mechanism detects the failure of LSP.For example, two-way converting detects (Bidirectional Forwarding Detection, BFD), multiprotocol label switching transmission subset (Multiprotocol Label Switching transport profile, MPLS-TP) etc..In the case that work is broken down, work LSP roads The business carried on footpath can be rapidly switched on protection LSP paths.

In order to ensure this mechanism of protection and restoration effective operation, need to follow in the selection for protecting and working LSP paths The separated principle in path.Path separation of the prior art, refers to node and link that operating path and Protection path are passed through It is entirely different.

Fig. 2 shows the work LSP paths for following path separation principle in the prior art and protection LSP paths.Such as Fig. 2 institutes Show, networking topology includes 2 network domains (for example, domain A and domain B).Wherein, work LSP and protection LSP extend through domain A and Domain B.But work LSP and protection LSP are respectively distributed to different chain roads, and the node of chain road does not overlap.

Based on the separated concept in such path, a kind of sequential path is proposed in the prior art and calculates (Sequential Path computation) method follows the operating path and Protection path of path separation principle to calculate.Sequential path calculates The main thought of method is：First establish a work LSP path.Then, one is resettled in the case where excluding work LSP paths Item protection LSP paths separated with work LSP paths.It is as follows specifically to establish process：

(1) a cross-domain operating path is calculated.

(2) existing operating path (including excluding node, link on operating path), weight are excluded in route information table Newly establish a Protection path.

In the case of cross-domain MPLS networkings, the method for building up in wall scroll LSP paths is very ripe.Layering portion can be passed through Layered method and collaboration between the centerized fusion unit of administration realize the calculating and foundation in LSP paths.

But since order computational methods are when the first paths (for example, operating path) is established, is not accounted for The requirement of two paths (for example, Protection path).Therefore, under many network topology structures, order computational methods cannot be guaranteed It calculates and follows the separated work LSP paths in path and protection LSP paths.Alternatively, after work LSP paths are calculated, it can not Calculate protection LSP paths.

Fig. 3 is a kind of schematic diagram of network topology structure.As shown in figure 3, node A and node Z are respectively source node and mesh Node (alternatively, referred to as destination node).When calculating two disjoint paths from node A to node Z, if calculating side in sequence Method, the optimization task route footpath calculated are A-B-F-H-J-Z.When calculating Protection path, after excluding operating path, will be unable to look for To the Protection path being kept completely separate with operating path.

However, it can be seen from fig. 3 that in actual network topology, there are separated service paths.For example, A-C-E-F-H-J-Z and A-B-D-G-I-Z.

As it can be seen that the prior art propose order computational methods there are it is certain the defects of, do not examined when calculating the first paths Consider the second paths, may result in not calculating and follow the separated operating path in path and Protection path.

Below in conjunction with Fig. 4 to Fig. 7, the method for establishing cross-domain disjoint paths of the embodiment of the present application is described in detail.

Firstly, it is necessary to explanation, the method provided by the embodiments of the present application for establishing cross-domain disjoint paths is applied to concentrate In the network of control mode, include but not limited to MPLS network.In other networks using interconnection technique, this can also be used Apply for that the method for the cross-domain disjoint paths of calculating of embodiment calculates cross-domain disjoint paths, for example, Synchronous Digital Hierarchy The VC connections of (Synchronous Digital Hierarchy, SDH), optical transfer network (Optical Transport Network, OTN) in optical channel data cell connection, wavelength connection in WDN etc..

It should be noted that disjoint paths described in the embodiment of the present application are (including disjoint paths in domain described hereinafter With cross-domain disjoint paths) with described disjoint paths are different in the prior art above.In the embodiment of the present application, multiple domain is worked as When the source boundary node and egress boundary node of each network domains included by network are multiple, according to the embodiment of the present application The two cross-domain disjoint paths (that is, cross-domain operating path and cross-domain Protection path) established are without going past same node, i.e., two The cross-domain disjoint paths of item strictly disjoint.And in the multiple-domain network some network domains only there are one source boundary node or only there are one During egress boundary node, it can then pass through same node (that is, two according to two cross-domain disjoint paths that the embodiment of the present application is established The cross-domain disjoint paths of item can pass through the only source boundary node of the network domains or an only egress border is saved Point).It can be seen that in the embodiment of the present application, at least two network domains for reaching that destination node needs pass through from source node The source boundary node for having subnetwork domain is only one or when egress boundary node is only one, and high level controller is finally established Cross-domain disjoint paths be not what is strictly disjointed.

It is thus understood that according to the method for establishing cross-domain disjoint paths of the embodiment of the present application, for multiple domain net The topological structure of network does not have any restrictions.I.e. that is, no matter what kind of the real topology of multiple-domain network is, according to this Shen Please embodiment can calculate two cross-domain disjoint paths.Although under some topological structures, two calculated are cross-domain Disjoint paths are not to strictly disjoint and (that is, may pass through identical node).But compared in the prior art in such case Under can not calculate cross-domain disjoint paths and compare, the performance of algorithm is improved.Also, when operating path breaks down, if The node of failure is not the node that the operating path and Protection path pass through jointly, then, Protection path can't be subject to work Make the influence of path failure, so as to realize that business is switched to Protection path from operating path, to play fast quick-recovery business Effect.

Fig. 4 is the schematic flow chart of the method 200 provided by the embodiments of the present application for establishing disjoint paths.It is to be understood that side Method 200 can be performed by the high level controller in centralized control type network.It more specifically, can be by configuring in high level controller Processor perform.As shown in figure 4, this method mainly includes step 210-230.

210th, high level controller determines the domain sequence formed from the network domains that source node arrival destination node needs pass through, The domain sequence includes at least two network domains, and each network domains include source boundary node and egress boundary node.

It is to be understood that domain sequence here is a sequence being made of in order multiple domains.From the network domains where source node It sets out, successively by each network domains in the sequence of domain, can finally reach the network domains where destination node.In fact, source is saved Point is the node in first network domains in the domain sequence, and destination node is the section in the last one network domains in the domain sequence Point.

In the embodiment of the present application, domain sequence can be preset.In other words, from source node reach destination node need through Which network domains is crossed, can be specified by user.

Optionally, domain sequence can also automatically be calculated according to cross-domain topological relation by high level controller and generated.

In addition, the connection relation between network domains can be specified by user and (in other words, preset).Alternatively, between network domains Connection relation can also automatically be found based on the agreement between network domains by high level controller.

220th, disjoint paths group set in domain possessed by each network domains that the true domain sequences of high level controller include, Disjoint paths group set includes disjoint paths group at least one domain in the domain, and disjoint paths group is included in two domains in each domain Disjoint paths, the paths in two domains in disjoint paths are protected as operating path in domain and another paths as in domain Road maintenance footpath, wherein, each network domains have at least one disjoint paths group.

Specifically, high level controller can be by receiving being used to indicate for the corresponding lower-level controllers transmission of each network domains The indication information of disjoint paths group set in the domain of each network domains determines the domain of each network domains included in the domain sequence Interior disjoint paths group set.

In embodiment is applied for, disjoint paths group is by the corresponding lower-level controllers of the network domains in the domain of each network domains It is calculated.Each lower-level controllers are in the domain for calculating corresponding network domains during disjoint paths group, according to network domains side The difference of boundary's node topology structure, specific computational methods are also different.

In fact, the topology of network domains boundary node point can be three kinds of situations (for convenience of description, is denoted as feelings individually below Condition 1, situation 2 and situation 3).

1, source boundary node of situation, multiple egress boundary nodes

Under the topological structure of situation 1, disjoint paths group includes disjoint paths in two domains and (remembers individually below in each domain Make path 1 and path 2), the source boundary node of disjoint paths is identical in this two domains and egress boundary node is different.

Optionally, as one embodiment, the first network domain at least two network domains includes a source border Node and multiple egress boundary nodes, at least two lower-level controllers include the first lower-level controllers, the first low layer control Device processed is the corresponding lower-level controllers in first network domain,

And disjoint paths group collection in domain possessed by each network domains for including of the true domain sequences of high level controller It closes, including：

High level controller receives the first indication information that the first lower-level controllers are sent, and according to the finger of the first indication information Show disjoint paths group set in the domain in definite first network domain, disjoint paths group set includes first in the domain in the first network domain Disjoint paths group at least one domain of network, disjoint paths group is the first lower-level controllers by the domain in the first network domain Multiple egress boundary nodes of one network domains are combined two-by-two after obtaining multiple egress boundary node groups, by first network domain What source boundary node was combined one by one with the plurality of egress boundary node group.

Specifically, if the source boundary node of some network domains is one in the sequence of domain and egress boundary node is more It is a, the corresponding lower-level controllers of the network domains (that is, the first lower-level controllers) disjoint paths group in the domain for calculating the network domains When, first by egress boundary node combination of two, multipair egress boundary node group is formed, then by source boundary node and is owned " egress boundary node group " is combined, and obtains disjoint paths group in the multipair domain of the network domains.

Optionally, lower-level controllers are calculated at least one domain of corresponding network domains after disjoint paths group, to High level controller transmission is used to indicate the indication information of disjoint paths group set in the domain of corresponding network domains.Wherein, the domain Interior disjoint paths group set includes at least one in disjoint paths group in the domain of corresponding network domains.That is, lower-level controllers It can not have to separation group in whole domains of the corresponding network domains calculated informing higher layer server.

Fig. 5 is a kind of example of topological structure of disjoint paths group.As shown in figure 5, the source border that node A is domain A is saved Point, node B1, B2, B3 are the egress boundary node of domain A.Topological structure shown in Fig. 5 is commonly used in service access domain, node A is LSP access nodes, and node B1, B2, B3 are cross-domain node.

According to described previously, the corresponding lower-level controllers of domain A first by node B1, B2, B3 combination of two, formed (B1, B2), (B1, B3), (B2, B3) three kinds of egress boundary node groups.Again by source boundary node and these three egress boundary node groups It is respectively combined, ultimately forms [A- (B1, B2)], [A- (B1, B3)], disjoint paths group in [A- (B2, B3)] three domains. That is, disjoint paths group in three domains of domain A is generated.

Situation 2, multiple source boundary nodes, multiple egress boundary nodes

Under the topological structure of situation 2, disjoint paths group is also comprising disjoint paths in two domains in each domain, different from feelings Condition 1, the source boundary node in path is different in this two domains, and egress boundary node is also different.

Optionally, as one embodiment, the second network domains at least two network domains include multiple source borders Node and multiple egress boundary nodes, at least two lower-level controllers include the second lower-level controllers, second low layer Controller is the corresponding lower-level controllers of the second network domains,

And disjoint paths group collection in domain possessed by each network domains for including of the true domain sequences of high level controller It closes, including：

High level controller receives the second indication information that the second lower-level controllers are sent, and according to the finger of the second indication information Show disjoint paths group set in the domain of definite second network domains, disjoint paths group set includes second in the domain of second network domains Disjoint paths group at least one domain of network domains, disjoint paths group is the second lower-level controllers by the domain of the second network domains Multiple source boundary nodes of two network domains are combined two-by-two, obtain at least one source boundary node group, and by the second net Multiple egress boundary nodes in network domain are combined two-by-two, after obtaining at least one egress boundary node group, this is at least one What source boundary node group was combined one by one at least one egress boundary node group.

Specifically, if the source boundary node of some network domains is multiple in the sequence of domain and egress boundary node is multiple When, the corresponding lower-level controllers of the network domains are in the domain for calculating the network domains during disjoint paths group, lower-level controllers (that is, the Two lower-level controllers) first by source boundary node combination of two, form at least one source boundary node group (that is, source side The combination of boundary's node).Egress boundary node combination of two is formed at least one egress boundary node group by lower-level controllers simultaneously (that is, the combination of egress boundary node).Last lower-level controllers are at least one with this by least one source boundary node group Egress boundary node group combination of two, obtains disjoint paths group at least one domain of the network domains.

Fig. 6 is the example of another topological structure of disjoint paths group.As shown in fig. 6, node C1, C2, C3 are the source of domain B Boundary node is held, node D1, D2, D3 are the egress boundary node of domain B.B corresponding lower-level controllers in domain are first by source border Node C1, C2, C3 are combined two-by-two, obtain source boundary node group, are respectively (C1, C2), (C1, C3), (C2, C3).Together When, D1, D2, D3 are combined two-by-two, obtain egress boundary node group, is respectively (D1, D2), (D1, D3), (D2, D3).Most Afterwards, source boundary node group and egress boundary node group are combined by the corresponding lower-level controllers of domain B again two-by-two, obtain domain B Domain in disjoint paths group.

If it is understood that a network domains have n source boundary node, m egress boundary node, then, generation The domain domain in the quantity of disjoint paths group be：

Wherein, n is the integer more than or equal to 2, and m is the integer more than or equal to 2.

By taking the domain boundary topology structure shown in Fig. 6 as an example, the quantity of disjoint paths group is in the domain of obtained domain B：

Respectively：

[C1-D1、C2-D2]、[C1-D2、C2-D1]

[C1-D1、C2-D3]、[C1-D3、C2-D1]

[C1-D2、C2-D3]、[C1-D3、C2-D2]

[C1-D1、C3-D2]、[C1-D2、C3-D1]

[C1-D1、C3-D3]、[C1-D3、C3-D1]

[C1-D2、C3-D3]、[C1-D3、C3-D2]

[C2-D1、C3-D2]、[C2-D2、C3-D1]

[C2-D1、C3-D3]、[C2-D3、C3-D1]

[C2-D2、C3-D3]、[C2-D3、C3-D2]。

Situation 3, multiple source boundary nodes, an egress boundary node

It is to be understood that in case 3, lower-level controllers calculate the process and feelings of disjoint paths in the domain of corresponding network domains Condition 1 is similar.That is, lower-level controllers obtain at least one source boundary node first by multiple source boundary node combination of two Group, then egress boundary node is combined one by one with the source boundary node group, so as to obtain disjoint paths group at least one domain.

Each lower-level controllers are according to the topological structure of corresponding network domains boundary node, according in above-mentioned three kinds of situations Method, calculate disjoint paths group at least one domain of corresponding network domains, and calculate complete after to high level controller Indication information is sent, which is used to indicate disjoint paths group at least one domain to high level controller.

Correspondingly, high level controller receives the indication information that each lower-level controllers are sent, and according to indication information It indicates to determine disjoint paths group set in the domain of each network domains corresponding to each lower-level controllers, in the domain of each network domains Disjoint paths group set includes disjoint paths group at least one domain of the network domains.So as to which high level controller obtains domain sequence In each network domains at least one domain in disjoint paths group.

Here, the mode that lower-level controllers are sent with indication information to high level controller is not particularly limited.For example, low layer Controller can be by sending Simple Network Management Protocol (Simple Network to high level controller to high level controller Management Protocol, SNMP) message, Socket protocol messages and transmission control protocol (Transmission Control Protocol, TCP) protocol message mode, to high level controller transmission be used to indicate disjoint paths group in domain Indication information.

By taking snmp message as an example, lower-level controllers carry the indication information in snmp message.Specifically, the indication information It can indicate two mark sequences.Each mark sequence is made of at least two device identifications.Wherein, each in mark sequence Each device identification is for identifying an equipment, also, separation road in each device identification difference identification field identified in sequence The equipment (in other words, node) that footpath is sequentially passed through.In addition, the interface message between these equipment is further included in the indication information.

High level controller receives the snmp message that lower-level controllers are sent, and the indication information, base are obtained from snmp message In the indication information, high level controller is believed according to the interface in each mark sequence between the order and equipment of device identification Breath can determine possessed by the network domains disjoint paths group in domain.

It should be noted that it in the embodiment of the present application, in the domain in calculating network domain during disjoint paths group, only accounts for The boundary node (including source boundary node and egress boundary node) of network domains, without considering the node inside network domains. In fact, the method for the disjoint paths of calculating network domain internal node is highly developed in the prior art, the embodiment of the present application This is not described further.

230th, high level controller disjoint paths group set in domain according to possessed by each network domains that domain sequence includes, Determine a cross-domain disjoint paths group set, which includes at least one cross-domain disjoint paths group, often A cross-domain disjoint paths group includes disjoint paths group in a domain of each network domains, and each cross-domain disjoint paths group includes two Cross-domain disjoint paths, paths in two cross-domain disjoint paths as cross-domain operating path and another paths be used as across Domain Protection path.

According to described previously, cross-domain operating path and cross-domain Protection path are two and are separated from each other and sequentially pass through domain sequence In each network domains path.According to the calculating of step 220, the disjoint paths group in each domain has been obtained.

Specifically, in step 230, select to separate in a domain in disjoint paths group out of multiple domains of each network domains Group of paths is combined, and obtains a cross-domain disjoint paths group set, which includes multiple cross-domain points From group of paths.

It is easily understood that disjoint paths group and domain sequence in the domain that each cross-domain disjoint paths group being combined into includes In network domains correspond, disjoint paths group is disjoint paths group at least one domain of corresponding network domains in each domain In one.

It can be seen that in the embodiment of the present application according to above-mentioned calculating process, high level controller is establishing cross-domain separation road During footpath, it is first determined disjoint paths in the domain of each network domains in the network domains passed through from source node arrival destination node needs Group, then sequentially disjoint paths group in the domain of these network domains is combined, obtain cross-domain disjoint paths group.Due in each domain Disjoint paths group includes path in two domains, and one is used as operating path in domain, and another as Protection path in domain.Change sentence It talks about, the foundation of Protection path in domain is also allowed for when establishing operating path in domain.So as to be determined according to disjoint paths group in domain Each cross-domain disjoint paths group include two cross-domain disjoint paths, wherein a paths are as cross-domain operating path and another Paths are as cross-domain Protection path.The process that cross-domain disjoint paths are established from the embodiment of the present application can be seen that cross-domain work Path and cross-domain Protection path are established simultaneously, it is thus possible to when avoiding order calculating disjoint paths in the prior art, due to It first establishes cross-domain operating path and resettles cross-domain Protection path, after the completion of cross-domain operating path calculating may be caused, can not calculate The problem of drawing the topological trap of cross-domain Protection path.

Optionally, as one embodiment, this method further includes：

High level controller calculates the path cost of each cross-domain disjoint paths group in the cross-domain disjoint paths group set, and will The cross-domain disjoint paths group of path cost minimum is determined as the cross-domain disjoint paths group of target.

It is easily understood that according to above-mentioned steps 210-230, high level controller has determined and (in other words, has been calculated) one A cross-domain separation group set, and cross-domain disjoint paths group set includes at least one cross-domain disjoint paths group, it is cross-domain point each From in group of paths actually all include two cross-domain disjoint paths.Therefore, can be determined from step 230 at least two are cross-domain A cross-domain disjoint paths group is arbitrarily selected in disjoint paths group, and one in selected cross-domain disjoint paths group is cross-domain Path is as cross-domain operating path, and in addition a cross-domain path is as cross-domain Protection path.

In the present embodiment, can path cost be selected most from the cross-domain disjoint paths group set according to path cost Small cross-domain disjoint paths group is as the cross-domain disjoint paths group of target.

It is possible to further using the smaller cross-domain path of path cost in the cross-domain disjoint paths group of target as cross-domain Operating path, and using another larger cross-domain path of path cost as cross-domain Protection path.

Optionally, as one embodiment, high level controller obtains each domain possessed by each network domains in the sequence of domain The path cost of interior disjoint paths group, wherein, in each domain of each network domains the path cost of disjoint paths group be with it is each What the lower-level controllers corresponding to network domains were obtained according to the weight calculation of default operating path and Protection path；

Path cost and default domain-to-domain link weight of the high level controller according to each disjoint paths group, calculate each across The path cost of domain disjoint paths group.

Specifically, with previously described lower-level controllers disjoint paths group in domain is used to indicate to high level controller transmission Indication information is similar, lower-level controllers can by high level controller send snmp protocol message, Socket protocol messages and Transmission Control Protocol message reports the path cost of disjoint paths group in domain to high level controller.For example, snmp protocol message, It carries to be used to indicate in the domain in the network domain corresponding to the lower-level controllers in Socket protocol messages or Transmission Control Protocol message and separate The information of the path cost of group of paths.

It is contemplated that when disjoint paths group has multiple in the domain of a network domains, which includes more The information of a path cost.Obviously, the path cost of disjoint paths group in whole domains can be reported to high level by lower-level controllers Controller can also select the path cost of disjoint paths group in the smaller part domain of path cost to be reported to high level controller.

In the embodiment of the present application, high level controller is to combine each network in the cross-domain disjoint paths group of definite target The path cost of disjoint paths group in the domain in domain, the path cost of disjoint paths group and the power of domain-to-domain link cost in default domain What re-computation obtained.

Lower-level controllers can substantially include for the calculating of the path cost of disjoint paths group [C1-D1, C2-D2] in domain Following process：

(1) path cost of calculate node C1 to node D1 is denoted as Cost [C1-D1] below.

(2) path cost of calculate node C2 to node D2 is denoted as Cost [C2-D2] below.

After the path cost for calculating two disjoint paths in domain in disjoint paths group [C1-D1, C2-D2], by path The small paths of cost are determined as operating path in domain, and the big another paths of path cost are determined as protecting road in domain Footpath.

(3) path cost of disjoint paths group in the domain is calculated.

Specifically, operating path in path cost × domain of operating path in path cost=domain of disjoint paths group in domain Path cost weight k+ domains in Protection path path cost × domain in Protection path path cost weight j.

It should be noted that in domain here in the path cost weight of operating path and domain Protection path path cost Weight can be specified by user and (in other words, preset).

(4) path cost of cross-domain disjoint paths is calculated.

The path cost of cross-domain disjoint paths is each network domains that high level controller is determined in each lower-level controllers It is calculated in domain on the basis of the path cost of disjoint paths.

Specifically, high level controller can calculate the path cost of cross-domain disjoint paths group as follows.That is, it is cross-domain Disjoint paths in weight+domain of the path cost of disjoint paths group #1 × network domains #1 in path cost=domain of disjoint paths The weight of path cost × network domains #2 of group #2+...+(path cost of the domain-to-domain link of network domains #1 and network domains #2) × (the path cost weight of the domain-to-domain link of network domains #1 and network domains #2)+(domain-to-domain link of network domains #2 and network domains #3 Path cost) × (the path cost weight of the domain-to-domain link of network domains #2 and network domains #3)+....

Similarly, (in other words, in advance the path cost weight of domain-to-domain link and the weight of network domains can be specified by user If).

The path of all end-to-end (that is, source node to destination node) cross-domain disjoint paths is calculated according to above-mentioned steps After cost, by comparing, the cross-domain disjoint paths group of path cost minimum is determined as optimal (that is, target) cross-domain disjoint paths Group.

Finally, the optimal cross-domain disjoint paths group is established, completes the foundation of cross-domain disjoint paths.

As it was noted above, the method for the cross-domain disjoint paths of calculating of the embodiment of the present application is suitable for the concentration control of layering deployment Network system processed, for establishing the separated two strip labels switching path (LSP) in path in the case of multiple network domains.By each Interaction between layer, topology trap when can solve the problems, such as to establish cross-domain disjoint paths in the prior art.

Further, according to the method for establishing cross-domain disjoint paths of the embodiment of the present application, it is ensured that establish out across Domain disjoint paths global optimum.

With reference to the centrally controlled network system of layering deployment shown in Fig. 7, the embodiment of the present application illustrate It is bright.

Fig. 7 is the centralized control type network system disposed suitable for a kind of layering of the embodiment of the present application.As shown in fig. 7, Two layers of deployment of network system point is denoted as layer 1 and layer 0 (L1 and L0 as illustrated in the drawing respectively) respectively.Wherein, controller A is height Layer controller, controller B, controller C, controller D and controller E are lower-level controllers.Controller B, controller C, controller D, controller is managed and controls to domain 1, domain 2, domain 3 and domain 4 respectively.Pass through friendship between lower-level controllers and high level controller Mutually realize the coordinated management and control to entire network system.

301st, routing iinformation is generated.

Specifically, lower-level controllers (controller B, controller C, controller D and controller E as illustrated in the drawing) generation is each From the routing iinformation in domain.High level controller forms the contact relation of boundary node between domain by way of being manually specified, generation Cross-domain routing iinformation.

302nd, true domain sequences.

After high level controller (that is, controller A) receives cross-domain service foundation request, opened up according to cross-domain be abstracted of this layer Flutter, calculate needed successively from source node to destination node by network domains sequential combination, obtain a domain sequence.Cross-domain pumping As the topological connection relation comprising between the boundary node of domain, wherein, cross-domain boundary point connection relation can be specified by user, Alternatively, it can also automatically be found based on the agreement between network domains by high level controller.

303rd, disjoint paths group in the domain of each network domains is generated.

Controller A sends request to lower-level controllers and disappears according to the LSP domains passed through and the boundary node of the domain and adjacent domains Breath, the request message are used for disjoint paths group in the domain that lower-level controllers is asked to calculate corresponding network domains.

Lower-level controllers receive the request message that high level controller is sent, and determine at least one domain of corresponding network domains Interior disjoint paths group.

Lower-level controllers determine that the detailed process of disjoint paths group in domain may refer to previously described situation 1 and situation 2, Details are not described herein again.

304th, the path cost of disjoint paths group in the domain in each domain is calculated.

Specifically, each lower-level controllers calculate the path generation of disjoint paths group in all domains of corresponding network domains Valency.After the completion of calculating, to high level controller A send comprising the service aisle cost of path separation group in domain, protection passage cost, The information of the path cost of path separation group in domain.

305th, the path cost of cross-domain disjoint paths group is determined.

What high level controller A received that each lower-level controllers send is used to indicate corresponding each network domains at least In one domain after the indication information of the path cost of disjoint paths group, according to the instruction of the indication information and cross-domain boundary node Connection relation, combination obtain multiple cross-domain disjoint paths groups (that is, end-to-end separation LSP group of paths) to get to one cross-domain point From group of paths set.Then high level controller calculates the path cost of each cross-domain disjoint paths group.

Similarly, specific calculating process may be referred to above.Which is not described herein again.

306th, optimal cross-domain disjoint paths group is determined.

It is to be understood that high level controller can arbitrarily select a cross-domain disjoint paths group from cross-domain disjoint paths group set As the cross-domain disjoint paths group of target, and using the smaller paths of path cost in the cross-domain disjoint paths group of target as cross-domain Operating path, and using the larger another paths of path cost as cross-domain Protection path.

Optionally, high level controller A selects path cost by comparing the path cost of all cross-domain disjoint paths groups Minimum cross-domain disjoint paths group is as the cross-domain path separation group of target.The cross-domain disjoint paths group of target at this time for it is global most It is excellent.

307th, optimal cross-domain disjoint paths group is established.

Path in domain in the optimal cross-domain disjoint paths group selected is issued to corresponding low layer control by high level controller A Device processed is completed the foundation of cross-domain disjoint paths by lower-level controllers.

It should be noted that the embodiment of the present application only with the situation using two layers controller as an example, more being deployed with In the case of layer controller, by being successively abstracted, the method for calculating cross-domain path of the embodiment of the present application can also be used.

It is to be understood that mentioned here be successively abstracted the topology for referring to the network that lower-level controllers are managed, high level is controlled Device is abstracted as the only network topology comprising domain intermediate node and is abstracted.For example, L0 layer networks are abstracted as domain boundary point by L0 layers of controller Connection relation is to L1 layers of control unit.L1 layer networks are abstracted as domain boundary point connection relation to L2 layers of controller by L1 layers of controller Deng.

The method for establishing disjoint paths of the embodiment of the present application, by calculating cross-domain operating path and cross-domain protection road simultaneously Footpath, when can solve to calculate cross-domain disjoint paths in existing scheme, first establish cross-domain Protection path resettle cross-domain Protection path and The problem of topological trap that cross-domain Protection path caused by possible can not be calculated.

Further, in the cross-domain disjoint paths group of selection target, high level controller will calculate the embodiment of the present application To multiple cross-domain disjoint paths groups in the cross-domain disjoint paths group of path Least-cost be determined as the cross-domain disjoint paths group of target, At this point, the cross-domain disjoint paths group of target is actually global optimum, therefore, it is possible to solve the overall situation calculated in the prior art The problem of disjoint paths are non-optimal.

The method for establishing disjoint paths of the embodiment of the present application is described in detail above in association with Fig. 4 to Fig. 7.Below With reference to Fig. 8 to figure, the device for establishing disjoint paths of the embodiment of the present application is illustrated.

Fig. 8 shows the schematic diagram of the device 400 for establishing disjoint paths of the embodiment of the present application.Wherein, device 400 configures In the communication system including at least two lower-level controllers, which includes at least two network domains, this at least two Mapping relations between lower-level controllers and at least two network domains are one-to-one corresponding.As shown in figure 8, device 400 includes：

First processing units 410, for determining that reaching destination node from source node needs at least two net passed through The domain sequence that network domain is formed, each network domains include source boundary node and egress boundary node；

Second processing unit 440 is used for, each disjoint paths group set in domain possessed by network domains in true domain sequences, Disjoint paths group set includes disjoint paths group at least one domain in the domain, and disjoint paths group is included in two domains in each domain Disjoint paths, the paths in two domains in disjoint paths are protected as operating path in domain and another paths as in domain Road maintenance footpath, wherein, each network domains have at least one disjoint paths group；

Second processing unit 440 is additionally operable to, according to disjoint paths group collection in domain possessed by each network domains in the sequence of domain It closes, determines a cross-domain disjoint paths group set, which includes at least one cross-domain disjoint paths group, each Disjoint paths group in one domain of cross-domain disjoint paths group including each network domains, each cross-domain disjoint paths group include two across Domain disjoint paths, in two cross-domain disjoint paths a paths are as cross-domain operating path and another paths are as cross-domain Protection path.

Specifically, device 400 further includes transmitting element 420 and receiving unit 430.

Optionally, as one embodiment, the first network domain at least two network domains includes a source border Node and multiple egress boundary nodes, at least two lower-level controllers include the first lower-level controllers, the first low layer control Device processed is the corresponding lower-level controllers in first network domain,

The receiving unit 430, for receiving the first indication information that the first lower-level controllers are sent；

And second processing unit 440 is specifically used for the instruction according to the first indication information, determines the domain in first network domain Interior disjoint paths group set, disjoint paths group set is included at least one domain in first network domain in the domain in the first network domain Disjoint paths group, disjoint paths group is the first lower-level controllers by multiple egress sides in first network domain in the domain in first network domain Boundary's node is combined two-by-two after obtaining at least one egress boundary node group, by the source boundary node in first network domain with being somebody's turn to do What at least one egress boundary node group was combined one by one.

Optionally, as one embodiment, the second network domains at least two network domains include multiple source borders Node and multiple egress boundary nodes, at least two lower-level controllers include the second lower-level controllers, the control of the second low layer Device is lower-level controllers corresponding with the second network domains,

The receiving unit 430 is additionally operable to receive the second indication information that the second lower-level controllers are sent；

And second processing unit 440 is specifically used for the instruction according to the second indication information, determines the domain of the second network domains Interior disjoint paths group set, disjoint paths group set is included at least one domain of the second network domains in the domain of second network domains Disjoint paths group, disjoint paths group is the second lower-level controllers by multiple source sides of the second network domains in the domain of the second network domains Boundary's node is combined two-by-two, obtains at least one source boundary node group, and multiple egress borders of the second network domains are saved Point is combined two-by-two, after obtaining at least one egress boundary node group, by least one source boundary node group with this extremely What a few egress boundary node group was combined one by one.

Optionally, as one embodiment, the second processing unit 440 is additionally operable to calculate the cross-domain disjoint paths combination The path cost for each cross-domain disjoint paths group that set includes, and the cross-domain disjoint paths group of path cost minimum is determined For the cross-domain disjoint paths group of target.

Optionally, as one embodiment, second processing unit 440 is specifically used for：

The path cost of disjoint paths group in each domain possessed by each network domains is obtained in the sequence of domain, wherein, each The path cost of disjoint paths group is according to default with the lower-level controllers corresponding to each network domains in each domain of network domains Operating path and the weight calculation of Protection path obtain；

According to the path cost of each disjoint paths group and default domain-to-domain link weight, each cross-domain disjoint paths are calculated The path cost of group.

Each unit in the device 400 of the embodiment of the present application and above-mentioned other operations or function are respectively in order to realize this Apply for the corresponding flow performed in the method 200 for establishing disjoint paths of embodiment by high level controller.For sake of simplicity, herein not It repeats again.

Fig. 9 shows the schematic diagram of the device 500 for establishing disjoint paths of the embodiment of the present application.Wherein, device 500 configures In the communication system including at least two network domains, which includes high level controller and the control of at least two low layers Device is mapped as corresponding, at least two low layer between at least two lower-level controllers and at least two network domains Controller includes device 500 corresponding with first network domain, and first network domain is any one at least two network domains A network domains.As shown in figure 9, device 500 includes：

Processing unit 510, for determining disjoint paths group set in the domain in first network domain, disjoint paths group collection in the domain Conjunction includes disjoint paths group at least one domain, disjoint paths group bag in each domain at least one domain in disjoint paths group Disjoint paths in two domains are included, the paths in two domains in disjoint paths are as operating path in domain and another paths As Protection path in domain；

Transmitting element 520, for sending the domain for being used to indicate high level controller and determining first network domain to high level controller The indication information of interior disjoint paths group set, in order to which high level controller is according to disjoint paths group set in the domain in first network domain With disjoint paths group set in the domain of other network domains at least two network domains in addition to first network domain, one is determined Cross-domain disjoint paths group set, which includes at least one cross-domain disjoint paths group, cross-domain point each Include two cross-domain disjoint paths from group of paths, the paths in two cross-domain disjoint paths as cross-domain operating path and Another paths are as cross-domain Protection path.

Specifically, device 500 further includes receiving unit 530.

Optionally, as one embodiment, first network domain includes a source boundary node and multiple egress borders Node and, processing unit 510 is specifically used for：

The multiple egress boundary node in the first network domain is combined two-by-two, obtains at least one egress side Boundary's node group；

The source boundary node in first network domain and at least one egress boundary node group are combined one by one, Obtain disjoint paths group at least one domain in first network domain.

Optionally, as one embodiment, first network domain includes multiple source boundary nodes and multiple egress borders Node and, processing unit 510 is specifically used for：

The multiple source boundary node in first network domain is combined two-by-two, obtains at least one source border section Point group, and multiple egress boundary nodes in first network domain are combined two-by-two, obtain at least one egress boundary node group；

At least one source boundary node group and at least one egress boundary node group are combined one by one, obtained Disjoint paths group at least one domain in first network domain.

Optionally, as one embodiment, processing unit 510 is additionally operable to：

The path cost of disjoint paths group at least one domain in first network domain is calculated, which is additionally operable to indicate High level controller determines the path cost of disjoint paths group at least one domain in first network domain, in order to high level controller root According to disjoint paths group at least one domain in first network domain path cost and at least two network domains in remove first network Possessed by each network domains of other network domains outside domain at least one domain disjoint paths group path cost, by path The cross-domain disjoint paths group of Least-cost is determined as the cross-domain disjoint paths group of target.

Optionally, as one embodiment, processing unit is specifically used for according to default operating path and Protection path Weight calculates at least one domain in first network domain the path cost of disjoint paths group in each domain in disjoint paths group.

Each unit in the device 500 of the embodiment of the present application and above-mentioned other operations or function are respectively in order to realize this Apply for the phase performed in the method 200 for establishing disjoint paths of embodiment by lower-level controllers (for example, first lower-level controllers) Answer flow.For sake of simplicity, details are not described herein again.

Figure 10 is the schematic diagram of the equipment 600 for establishing disjoint paths of the embodiment of the present application.As shown in Figure 10, if Standby 600 include：Memory 610, processor 620 and communication interface 630.Wherein, memory 610, processor 620 and communication interface 630 are connected with each other by communication bus 640.

Memory 610 performs application program, code or the instruction of the present invention program for storing.Processor 620 is used to hold Application program, code or the instruction stored in line storage 610, to complete to establish the method 200 of disjoint paths and each implementation Corresponding flow by high level controller execution and/or operation in example.For sake of simplicity, details are not described herein again.

It is to be understood that is provided in Fig. 8 establishes the device 400 of disjoint paths, can be separated by the foundation shown in Figure 10 The equipment 600 in path is realized.For example, the first processing units 410 and second processing unit 440 in Fig. 8 can be by Figure 10 In processor 620 realize that transmitting element 420 and receiving unit 430 can be connect by one or more communications shown in Figure 10 Mouth 630 is realized.

Figure 11 is the schematic diagram of the equipment 700 for establishing disjoint paths of the embodiment of the present application.As shown in figure 11, if Standby 700 include：Memory 710, processor 720 and communication interface 730.Wherein, memory 610, processor 620 and communication interface 730 are connected with each other by communication bus 740.

Memory 710 performs application program, code or the instruction of the present invention program for storing.Processor 720 is used to hold Application program, code or the instruction stored in line storage 710, to complete to establish the method 200 of disjoint paths and each implementation Corresponding flow by lower-level controllers (for example, first lower-level controllers) execution and/or operation in example.For sake of simplicity, herein not It repeats again.

Likewise, what is provided in Fig. 9 establishes the device 500 of disjoint paths, can be separated by the foundation shown in Figure 11 The equipment 700 in path is realized.For example, the processing unit 510 in Fig. 9 can be realized by the processor 720 in Figure 11, send single Member 520 and receiving unit 530 can be realized by one or more of Figure 11 communication interfaces 730.

Processor shown in Figure 10 and Figure 11 is (for example, the processing shown in processor 620 and Figure 11 shown in Figure 10 Device 720) can be central processing unit (CPU), microprocessor, application-specific integrated circuit (application-specific Integrated circuit, ASIC) or one or more integrated circuits for being used to that the present invention program program to be controlled to perform.

Memory shown in Figure 10 and Figure 11 is (for example, the storage shown in memory 610 and Figure 11 shown in Figure 10 Device 710) it can be read-only memory (read-only memory, ROM) or the other types of static information and instruction can be stored Static storage device, random access memory (random access memory, RAM) or can store information and instruction Other kinds of dynamic memory or Electrically Erasable Programmable Read-Only Memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), read-only optical disc (Compact Disc Read-Only Memory, CD-ROM) or other optical disc storages, optical disc storage are (including compression optical disc, laser disc, optical disc, Digital Versatile Disc, Blu-ray Disc Deng), magnetic disk storage medium or other magnetic storage apparatus or can be used in carrying or store with instruction or data structure shape The desired program code of formula simultaneously can by any other medium of computer access, but not limited to this.Memory can be only It is vertical to exist, pass through communication bus (for example, communication bus 740 in communication bus 640 and Figure 11 in Figure 10) and processor phase Connection.Memory can also be integrated with processor.

Communication bus can also include power bus, controlling bus and status signal bus in addition in addition to including data/address bus Deng.For the sake of clear explanation, various buses are all designated as communication bus in figure.

Communication interface can be wireline interface, such as Fiber Distributed Data Interface (Fiber Distributed Data Interface, abbreviation FDDI), gigabit Ethernet (Gigabit Ethernet, abbreviation GE) interface etc. or wirelessly connect Mouthful.The embodiment of the present application is not particularly limited this.

It is to be understood that in the various embodiments of the application, the size of the sequence number of above-mentioned each process is not meant to perform suitable The priority of sequence, the execution sequence of each process should be determined with its function and internal logic, without the implementation of reply the embodiment of the present application Process forms any restriction.

Those of ordinary skill in the art may realize that each exemplary lists described with reference to the embodiments described herein Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually It is performed with hardware or software mode, specific application and design constraint depending on technical solution.Professional technician Described function can be realized using distinct methods to each specific application, but this realization is it is not considered that exceed Scope of the present application.

It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description, The specific work process of device and unit may be referred to the corresponding process in preceding method embodiment, and details are not described herein.

In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit Division is only a kind of division of logic function, can there is other dividing mode, such as multiple units or component in actual implementation It may be combined or can be integrated into another system or some features can be ignored or does not perform.It is another, it is shown or The mutual coupling, direct-coupling or communication connection discussed can be the indirect coupling by some interfaces, device or unit It closes or communicates to connect, can be electrical, machinery or other forms.

The unit illustrated as separating component may or may not be physically separate, be shown as unit The component shown may or may not be physical location, you can be located at a place or can also be distributed to multiple In network element.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs 's.

In addition, each functional unit in each embodiment of the application can be integrated in a processing unit, it can also That unit is individually physically present, can also two or more units integrate in a unit.

If the function is realized in the form of SFU software functional unit and is independent production marketing or in use, can be with It is stored in a computer read/write memory medium.Based on such understanding, the technical solution of the application is substantially in other words The part contribute to the prior art or the part of the technical solution can be embodied in the form of software product, the meter Calculation machine software product is stored in a storage medium, is used including some instructions so that a computer equipment (can be People's computer, server or network equipment etc.) perform each embodiment the method for the application all or part of step. And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic disc or CD.

It is to be understood that in the various embodiments of the application, the size of the sequence number of above-mentioned each process is not meant to perform suitable The priority of sequence, the execution sequence of each process should be determined with its function and internal logic, without the implementation of reply the embodiment of the present application Process forms any restriction.

The above is only the specific embodiment of the application, but the protection domain of the application is not limited thereto, any In the technical scope that those familiar with the art discloses in the application, change or replacement can be readily occurred in, should all be contained It covers within the protection domain of the application.Therefore, the protection domain of the application should be based on the protection scope of the described claims.

Claims (10)

1. A kind of 1. method for establishing disjoint paths, which is characterized in that it is performed in the communication system including at least two network domains, The communication system include high level controller and at least two lower-level controllers, at least two lower-level controllers with it is described extremely It is mapped as corresponding between few two network domains, the described method includes：

   The high level controller determines that reaching destination node from source node needs at least two network domains passed through to be formed Domain sequence, each network domains include source boundary node and egress boundary node；

   The high level controller determines each disjoint paths group set, the domain in domain possessed by network domains in the domain sequence Interior disjoint paths group set includes disjoint paths group at least one domain, and disjoint paths group includes separating in two domains in each domain Path, the paths in two domains in disjoint paths are protected as operating path in domain and another paths as in domain Path, wherein, each network domains have disjoint paths group at least one domain；

   High level controller disjoint paths group set in domain according to possessed by each network domains in the domain sequence, determines one A cross-domain disjoint paths group set, the cross-domain group of paths set include at least one cross-domain disjoint paths group, it is each described across Domain disjoint paths group includes disjoint paths group in a domain of each network domains, and each cross-domain disjoint paths group includes Two cross-domain disjoint paths, the paths in two cross-domain disjoint paths are as cross-domain operating path and another paths As cross-domain Protection path.
2. 2. according to the method described in claim 1, it is characterized in that, the first network domain at least two network domains includes One source boundary node and multiple egress boundary nodes, at least two lower-level controllers include the control of the first low layer Device, first lower-level controllers are the corresponding lower-level controllers in the first network domain,

   And the high level controller determines possessed by each network domains that the domain sequence includes disjoint paths group in domain Set, specifically includes：

   The high level controller receives the first indication information that first lower-level controllers are sent, and according to the described first instruction The instruction of information determines disjoint paths group set in the domain in the first network domain, disjoint paths in the domain in the first network domain Group, which is gathered, includes disjoint paths group at least one domain in the first network domain, disjoint paths in the domain in the first network domain Group is that the multiple egress boundary node in the first network domain is combined to obtain two-by-two by first lower-level controllers After at least one egress boundary node group, by the source boundary node in the first network domain and at least one egress border Node group is combined one by one.
3. 3. method according to claim 1 or 2, which is characterized in that the second network domains at least two network domains Including multiple source boundary nodes and multiple egress boundary nodes, at least two lower-level controllers include the second low layer control Device processed, second lower-level controllers are lower-level controllers corresponding with second network domains,

   And the high level controller determines possessed by each network domains that the domain sequence includes disjoint paths group in domain Set, specifically includes：

   The high level controller receives the second indication information that second lower-level controllers are sent, and according to the described second instruction The instruction of information determines disjoint paths group set in the domain of second network domains, disjoint paths in the domain of second network domains Group, which is gathered, includes disjoint paths group at least one domain of second network domains, disjoint paths in the domain of second network domains Group is that the multiple source boundary node of second network domains is combined by second lower-level controllers two-by-two, is obtained At least one source boundary node group, and the multiple egress boundary node of second network domains is combined two-by-two, After obtaining at least one egress boundary node group, by least one source boundary node group and at least one egress side Boundary's node group is combined one by one.
4. 4. according to the method in any one of claims 1 to 3, which is characterized in that the method further includes：

   The high level controller calculates the road for each cross-domain disjoint paths group that the cross-domain disjoint paths composite set includes Footpath cost, and the cross-domain disjoint paths group of path cost minimum is determined as the cross-domain disjoint paths group of target.
5. 5. according to the method described in claim 4, it is characterized in that, the high level controller calculates each cross-domain separation road The path cost of footpath group, specifically includes：

   The high level controller obtains in the domain sequence path of disjoint paths group in each domain possessed by each network domains Cost, wherein, the path cost of disjoint paths group is right with each network domains institute in each domain of each network domains The lower-level controllers answered are obtained according to the weight calculation of default operating path and Protection path；

   Path cost and default domain-to-domain link weight of the high level controller according to each disjoint paths group calculate institute State the path cost of each cross-domain disjoint paths group.
6. 6. a kind of device for establishing disjoint paths, which is characterized in that it configures in the communication system including at least two network domains, The communication system includes at least two lower-level controllers, at least two lower-level controllers and at least two network domains Between mapping relations to correspond, described device includes：

   First processing units, for determining that reaching destination node from source node needs at least two network domains institute group passed through Into domain sequence, each network domains include source boundary node and egress boundary node；

   Second processing unit, for determining each disjoint paths group set, institute in domain possessed by network domains in the domain sequence Stating disjoint paths group set in domain includes disjoint paths group at least one domain, and disjoint paths group includes two in each domain Disjoint paths in domain, the paths in two domains in disjoint paths are as operating path in domain and another paths conduct Protection path in domain, wherein, each network domains have at least one disjoint paths group；

   The second processing unit is additionally operable to according to possessed by each network domains in the domain sequence disjoint paths group collection in domain It closes, determines a cross-domain disjoint paths group set, the cross-domain group of paths set includes at least one cross-domain disjoint paths group, often A cross-domain disjoint paths group includes disjoint paths group in a domain of each network domains, each cross-domain separation road Footpath group includes two cross-domain disjoint paths, and the paths in two cross-domain disjoint paths are as cross-domain operating path and another One paths are as cross-domain Protection path.
7. 7. device according to claim 6, which is characterized in that the first network domain at least two network domains includes One source boundary node and multiple egress boundary nodes, at least two lower-level controllers include the control of the first low layer Device, first lower-level controllers are the corresponding lower-level controllers in the first network domain, and described device further includes receiving unit, The receiving unit is used to receive the first indication information that first lower-level controllers are sent；

   The second processing unit is specifically used for the instruction according to first indication information, determines the domain in the first network domain Interior disjoint paths group set, disjoint paths group set includes at least the one of the first network domain in the domain in the first network domain Disjoint paths group in a domain, disjoint paths group is first lower-level controllers by described first in the domain in the first network domain The multiple egress boundary node of network domains is combined two-by-two after obtaining at least one egress boundary node group, by described What the source boundary node of one network domains was combined one by one at least one egress boundary node group.
8. 8. device according to claim 6, which is characterized in that the second network domains at least two network domains include Multiple source boundary nodes and multiple egress boundary nodes, at least two lower-level controllers are controlled including the second low layer Device, second lower-level controllers are lower-level controllers corresponding with second network domains, and it is single that described device further includes reception Member, the receiving unit are used to receive the second indication information that second lower-level controllers are sent；

   The second processing unit is specifically used for the instruction according to second indication information, determines the domain of second network domains Interior disjoint paths group set, disjoint paths group set includes second network domains at least in the domain of second network domains Disjoint paths group in one domain, disjoint paths group is second lower-level controllers by described in the domain of second network domains The multiple source boundary node of two network domains is combined two-by-two, obtains at least one source boundary node group, and by institute The multiple egress boundary node for stating the second network domains is combined two-by-two, after obtaining at least one egress boundary node group, At least one source boundary node group and at least one egress boundary node group are combined one by one.
9. 9. the device according to any one of claim 6 to 8, which is characterized in that the second processing unit is additionally operable to count The path cost for each cross-domain disjoint paths group that the cross-domain disjoint paths composite set includes is calculated, and by path cost most Small cross-domain disjoint paths group is determined as the cross-domain disjoint paths group of target.
10. 10. device according to claim 9, which is characterized in that the second processing unit is specifically used for：

    The path cost of disjoint paths group in each domain possessed by each network domains is obtained in the domain sequence, wherein, it is described The path cost of disjoint paths group is the lower-level controllers corresponding to each network domains in each domain of each network domains It is obtained according to the weight calculation of default operating path and Protection path；

    According to the path cost of each disjoint paths group and default domain-to-domain link weight, each cross-domain separation is calculated The path cost of group of paths.