CN100373866C - Method of network fault recovery crossing over connections in multiple domains - Google Patents

Method of network fault recovery crossing over connections in multiple domains Download PDF

Info

Publication number
CN100373866C
CN100373866C CNB2004100918310A CN200410091831A CN100373866C CN 100373866 C CN100373866 C CN 100373866C CN B2004100918310 A CNB2004100918310 A CN B2004100918310A CN 200410091831 A CN200410091831 A CN 200410091831A CN 100373866 C CN100373866 C CN 100373866C
Authority
CN
China
Prior art keywords
territory
connection
network
protection
domain
Prior art date
Application number
CNB2004100918310A
Other languages
Chinese (zh)
Other versions
CN1798051A (en
Inventor
刘建国
马恒
杨发明
高峰
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to CNB2004100918310A priority Critical patent/CN100373866C/en
Publication of CN1798051A publication Critical patent/CN1798051A/en
Application granted granted Critical
Publication of CN100373866C publication Critical patent/CN100373866C/en

Links

Abstract

The present invention relates to a method for recovering network fault of cross-multidomain connection. A system comprises a transmission plane and a control plane, wherein the transmission plane comprises a network formed by the link connection of a plurality of network element devices which are used for distributing or establishing the link connection and establishing sub-network connection, and the sub-network is divided into a plurality of network domains; the control plane comprises a plurality of control nodes, and each network element of the transmission plane uniquely corresponds to on control node. By establishing the normal connection of the cross-domain, determining protective mechanisms in the domains, determining the implementation and the protective mechanisms among the domains, and organically matching the implementation, the method for recovering network fault of cross-multidomain connection provided by the present invention automatically establishes static protective connection which is used for recovering faults among the domains, or the protective connection is dynamically established aiming at the faults among the domains when the faults occur. Thereby, the present invention ensures the recovery of business faults in the domains, and what is more, the technical method which can ensure the recovery of business faults among the domains is creatively proposed.

Description

Cross over the method for the network disaster recovery of multiple domain connection
Technical field
The present invention relates to general transmission network; comprise that IP transmits the method that network, Ethernet, atm network, SDH and OTN etc. transmit the network disaster recovery of network field; relate in particular to a kind of network bandwidth resources of effectively utilizing and set up the method that cross-domain normal connection is connected with protection; business recovery when not only guaranteeing the territory internal fault; the more important thing is when influencing business when breaking down between the territory; recover professional effectively, and fully optimize and use Internet resources.
Background technology
According to current I TU standard recommendation G.841 and G.842,, transmit gateway in the restoration methods of fault, adopt the static protection method, these methods are set up corresponding protection and are connected at an end-to-end connection, or distribute resource for protection connects.The end-to-end connection here can be the single single connection that does not comprise a plurality of client layers, also can comprise service layer's connection that a plurality of client layers connect, and end-to-end connection can be striden two or more network nodes, and this connection can be used for carrying single or multiple service signals.
The patent of present related network fault recovery technology has generally adopted intelligent control software to realize the static protection technology automatically or has adopted dynamic heavy-route technology to set up protection when network breaks down to connect.For example China Patent No. promptly adopts such method for the patent that the CN02150136 name is called " a kind of ASON protection and method of recovering of realizing ": utilize the upper strata intelligence software to set up operating path and protection path automatically; When fault happens in operational, switch the business to the protection path according to the APS agreement; After fault recovery, business recovery is arrived operating path according to the APS agreement; When needs are deleted this business, initiate delete command by described upper strata intelligence software, deletion protection simultaneously and operating path.This method had both embodied the flexible of ASON, had absorbed the fast length of conventional optical network protected mode again, can satisfy the international standard that realizes protection in the 50ms.
But what above technical method was mainly considered is the interior network disaster recovery of individual domain.Owing to when network size is bigger,, also comprises and consider to connect the efficient of setting up in order to consider to recover efficient, network must be divided into some recoveries territory (or claiming protected field), can use present recovery technology in single recovery territory.But still do not have the fault recovery technology patent between interconnection recovery territory at present, fault comprises two kinds of situations between the recovery territory here: (1) domain-to-domain link breaks down; (2) node at the two ends of domain-to-domain link breaks down (also often claiming the Gateway Network Element fault).
Restoration methods about fault between the territory; can see disclosed method up till now from ITU G8080 etc.: (1) realizes fault recovery between the territory by SNCP; this method can be guaranteed professional recovery when link failure takes place; as shown in Figure 1; figure wherein (a) part is not considered the service failure recovery mechanism between the territory; between the territory, introduced and is connected the w corresponding protection with work and connects p and scheme (b) part, can guarantee the business of recovering when domain-to-domain link breaks down.Obviously recovery technology can not recover professional between this territory when Gateway Network Element breaks down; (2) gateway fails and when need recovering adopts and recovers territories with two and regard a big recovery territory as, in big recovery territory, avoids the Gateway Network Element fault by dynamic heavy-route is carried out in the end-to-end connection in this big territory.
It has following distinct disadvantage as can be known from present known recovery technology:
(1) the SNCP technology between the territory can only be recovered at link failure, does not consider protection gateway network element failure.
(2) when gateway breaks down, in bigger recovery territory, carry out fault recovery if adopt, route is calculated with being connected at each network element and is set up that to operate the network size that relates to very big, calculation of complex, thus influenced recovery efficient.
(3) existing recovery technology document patent, do not consider how effectively to recover at various faults between the territory, though these technology can be used for carrying out fault recovery between the territory, but connect in the system that sets up and recover at real-time supporting business, it is a full automatic process that cross-domain connection is set up, therefore how these methods are applied to fault between the territory, the systematic procedure that needs a complete automatic realization, this systematic procedure that realizes automatically comprises: realize that the operation that recovers between the territory should organically combine with the operation in the territory.If there is not this automatic realization system, just can not guarantee to connect the real-time of foundation.
Therefore, prior art is defectiveness also, still awaits continuing to improve and development.
Summary of the invention
The purpose of this invention is to provide a kind of method of crossing over the network disaster recovery of multiple domain connection, for overcoming all above-mentioned defectives of the prior art, when network is divided into a plurality of recoveries territory, to striding a plurality of recoveries territory and the connection of business recovery requirement being arranged, consider to use existing recovery technology, and further consider the virtual connections technology, when setting up the Restoration Mechanism that guarantees the territory internal fault, at inter-domain link fault and two kinds of situations of Gateway Network Element fault, set up service failure recovery mechanism between a kind of territory, guarantee the recovery of fault between the territory, and fully optimize and use Internet resources.
Technical scheme of the present invention is:
A kind of method of crossing over the network disaster recovery of multiple domain connection, its system comprises that one transmits a plane and a control plane, described transmission plane comprises a network that is connected and composed by link by a plurality of network element devices, this transmission network element is used for distributing or sets up link and connects, and set up the subnet connection, this network is divided into a plurality of territories; Described control plane comprises a plurality of Control Node, the unique corresponding Control Node of network element on described each transmission plane; Wherein, described method is further comprising the steps of:
A) described Control Node receives the normal connection request that the fault recovery from other Control Node or network management system equipment requires, the source transmission node of the connection that this connection request indicates and purpose transmission node be not on same net territory, this Control Node determines normally to be connected the outlet gateway location in this territory in the 1st territory at described source node place: borde gateway network element, the outbound that leaves this territory are connected with outbound;
B) the normal route that connects in calculating from the source transmission node to the territory that normally is connected inner outlet position, described the 1st territory, normal corresponding Control Node distribute links connection resource and the network element interconnection resource that connects the transmission network element of required process in the territory, and order then transmits the plane and sets up the actual physical connection;
C) to normally connecting in the territory of setting up, normally connect the Control Node of head-end network element and the Control Node of distal point network element and be responsible for setting up territory internal fault Restoration Mechanism.
D) Control Node of the above-mentioned outlet network element that is connected the 1st territory to outbound process next adjacent domains promptly the Control Node of the transmission Gateway Network Element in the 2nd territory send and connect the request of foundation, after Control Node received this connection request, the outlet network element in the 1st territory that connects earlier was connected with the normal of transmission Gateway Network Element in the 2nd territory; If the purpose of initial connection request transmits network element not on the 2nd territory, then set up normally connection in the territory of striding the 2nd territory by above-mentioned steps; Otherwise, calculate from connect through the 2nd territory transmit Gateway Network Element transmit to initial connection request purpose in the territory of network element normal link road by, the corresponding Control Node distribute links connection resource and the network element interconnection resource that normally connect the transmission network element of required process then in the territory, and order the transmission plane to set up the actual physical connection, thereby set up normal connection the in the territory;
E) to normally connecting in the 2nd territory of setting up, the method identical with described step c) set up territory internal fault Restoration Mechanism, when application domain internal fault Restoration Mechanism when network element or link occurs fault in the territory that connects process in the territory;
F) finish and protect establishment of connection between the territory: intermediate node selecting normally to be connected connecting portion branch leap first territory in and the cross-domain intermediate node of connecting portion branch in second territory are respectively as the newly-built source and destination that are connected; and consideration constraints: this connection and the cross-domain link road of having set up are set up cross-domain protection then and are connected by separating.When breaking down, can adopt the protection of one of following protected mode professional: traditional static protection mode (comprising 1+1 protection, protection in 1: 1), static virtual connections protected mode, dynamic heavy-route protected mode.
Described method; wherein, after the connection foundation of crossing over adjacent two territories finishes,, the connection that raw requests requires do not finish if also setting up; then continue to set up the cross-domain connection that comprises that protection connects between the territory, the whole connection of the connection source and destination that require up to raw requests is set up and is finished.
Described method wherein, can be set up the back that finishes (not comprising that protection connects between the territory) in the connection that raw requests requires, and the protection of setting up one by one automatically between adjacent domains connects again.
Described method, wherein, described method is characterized in that, recovery mechanism can adopt between described territory: traditional static protection mode, static state virtually connects protected mode, dynamic heavy-route protected mode;
Described traditional static protection mode refers to that it is real connection that protection connects, protection connect not domain of dependence fault and with the territory between normally is connected all the time existence jointly accordingly, and the traditional protection mode that business is protected;
Described static virtual connections protected mode refers to that it is virtual connections that protection connects, and protection connect do not rely on fault and with the territory between normally is connected all the time existence jointly accordingly, after normal connection fault, virtual connections is become actual physical to be connected, business switches in the protection connection then, thereby recovers the fault business;
Described dynamic heavy-route protected mode refers to not set up in advance protection and connects; after Gateway Network Element fault or inter-domain link fault occurring; dynamically foundation is connected by the protection that separates with former cross-domain normal link road, and business is switched in the protection connection, thereby recovers the fault business.
Described method, wherein, described method comprises that also several are recovered the territory constitutes a big territory, and each Control Node has the network topological information in whole big territory, can calculate for the connection that will set up in the big territory link road by.
Described method, wherein, when described method is determined to connect the domain-to-domain link of being crossed over when setting up cross-domain connection, in the time of promptly need determining Gateway Network Element that this link strides across, the Gateway Network Element of described selection satisfies condition: this Gateway Network Element of getting along well of the domain-to-domain link between other territory between interconnected Gateway Network Element intersects or risk is shared to the link of adjacent domains.
Described method; wherein; described method connects the full automation of setting up for guaranteeing all; the source and destination that the protection of adopting the mode of agreement in advance to specify to be used for fault recovery between the territory connects, select normal second jumping that is connected in normal the last two jump that connects and second territory first territory in respectively.
Described method; wherein, in above-mentioned establishment of connection process, can be further according to the priority of business; press priority from high to low, use following each recovery mechanism successively: traditional static protection mode, static virtual connections protected mode, dynamic heavy-route protected mode.
Described method, wherein, in the described step d) the Control Node of the outlet network element in the 1st territory to outbound the Control Node of transmission Gateway Network Element of next adjacent domains of process send and connect when setting up request, this request comprises outside the solicited message of above-mentioned normal connection request, also comprise last one of connection and jump out a mouthful positional information: outbound link position, adjacent two territories are to the public identities of cross-domain normal connection: public connection ID.
Described method, wherein, described route computing method comprises centralized or distributed.
A kind of method of crossing over the network disaster recovery of multiple domain connection provided by the present invention; compared with prior art; by organic cooperation of protection mechanism deterministic process between protection mechanism deterministic process, territory in cross-domain normal connection procedure, the territory; automatically having set up the static protection that is used for fault recovery between the territory connects; or the protection of fault connects between dynamically setting up at the territory when fault; thereby promptly guaranteed the service failure recovery in the territory, be to have provided when guaranteeing that fault takes place between the territory technical method that can recover business innovatively priorly.
In addition, the inventive method has also adopted virtual connections mechanism to carry out the business recovery of fault between the territory, has used Internet resources thereby optimize; Because it is relatively rare to transmit Internet resources between the territory, therefore, adopt the sharing functionality of virtual connections mechanism just to seem that the meaning of bigger resource rational utilization is arranged.
Description of drawings
Shown in Figure 1 is the restoration methods of the traffic failure of prior art;
What Fig. 2 represented is the system schematic that realizes the inventive method;
Fig. 3 is that the virtual connections of the inventive method is used schematic diagram;
Fig. 4 is the schematic flow sheet that the inventive method is used;
Fig. 5 is the transmission network application schematic diagram of the inventive method.
Embodiment
To the enforcement of technical solution of the present invention be described in further detail below in conjunction with accompanying drawing:
Be convenient narration and understanding, at first virtual connections notion of the present invention be explained and illustrated:
Virtual connections: be responsible for setting up the connection of across a network at control plane.At control plane, it is definite to form the link connection and the network element cross reference that connect, but transmitting the plane network element, does not have corresponding network element interconnection, and the link that constitute to connect is connected and transmits the plane and can be assigned with or be mounted, and also can not distribute or be mounted.With respect to virtual connections, the common described connection of the present invention is real connection.The real connection is to transmit the in esse physical connection in plane.If do not add specified otherwise, below still use " connections " expression " reality connection ".
The method of the network disaster recovery that leap multiple domain of the present invention connects, its hardware system consists of the following components, as shown in Figure 2:
(1) transmit planar system TP:
A plurality of network element devices connect and compose a network by link, and these transmission network elements are used for distributing or set up link and connect, and set up the subnet connection; These network element devices can be SDH/SONET equipment, WDM equipment, ethernet device, router device, atm device of transport service etc., and their transmission signal particle degree can be: Packet, TDM, optical fiber, wavelength etc.The corresponding Control Node of each network element device.Have communication interface between Control Node and the corresponding net element node, by this interface, Control Node can obtain the link circuit resource information of network element device and may command network element device and set up link and connect and be connected with subnet.Control Node can be actual equipment, this equipment can with a coexist ground or separate of original its control network element device.Control Node also can be a pure logic software system, and its function is embedded in the network element device.
According to network size, this network some territories have been divided into.Between the territory, there is devoid of risk shared link between a plurality of territories that connect the gateway joint.The link connection here is meant the resource that can independently support given traffic streams (as the Business Stream of specific bandwidth requirement) transmission that is included in the link.A link can contain one or more links and connect, and these link connection status can have three kinds: (1) is actual to be mounted and can to utilize; (2) can be mounted at any time and be used; (3) can not utilize.Network element is consistent with the notion of node, claims also sometimes in this specification that network element is a node.Other has a network management system equipment and transmits between network element and establish a communications link, and network management system equipment is used for managing and transmits network element.
(2) CP of control plane system:
Be made up of a plurality of Control Node, each transmits the unique corresponding Control Node of network element, can communicate by communication port between Control Node.Each Control Node has link-state information between this node and adjacent node, the link circuit resource information that link-state information comprises idle capacity, is used.Set up communication between the Control Node of adjacent transmission network element, each Control Node is known the Control Node mailing address of the adjacent transmission network element correspondence of the transmission network element that its is controlled.The major function of control plane is to set up, discharge, safeguard end-to-end connection.In the present invention, end-to-end connection is connected by the link that transmits the plane and is connected with subnet that (or the interconnection of title network element) form, and each is end-to-end to be connected with specific resource bandwidth requirement.Because when connecting must transmit network element through one the time, if determined that inbound link connects the position that is connected with outbound, then subnet connects also definitely simultaneously, so in the methods of the invention, establishment of connection is primarily aimed at link-attached definite and foundation.
At described control plane, its regards each territory of transmitting network as a mesh net, and it is carried out link road by calculating and carrying out various controls.
Route is calculated to divide between the territory with in the territory and is connected two kinds of situations considerations, the boundary node in territory is that gateway node uses Border Gateway Protocol, by using this agreement, the borde gateway node is known the connection accessibility of other domain node, and with the Gateway Network Element of these reachability territory interior nodes and adjacent domains.Between the territory, adopt the intradomain gateway agreement to obtain whole network topology, this network topology comprises the link circuit resource state information between all adjacent network elements on the entire domain.Route in the territory is calculated can adopt two kinds: concentrate or distribution.Concentrated routing algorithm is meant: have a master control node to know whole network topology in the territory, calculate for the connection of setting up in the territory link road that needs by.The distribution routing algorithm is meant: each master control node is known the whole network topology in the territory, can calculate link road for the connection that will set up in the territory by being called the source routing mode.The network topology here is meant and transmits the topology that network element constitutes.
Each Control Node can and its network element device and network management system equipment of interrelating communicate, the control network element device is to the use of link circuit resource, and can obtain the state information of network element device about link circuit resource, the fault such as whether.When the end-to-end connection of setting up supporting business, the Control Node of whole network cooperatively interacts, determine to form the some links connections that connect and be connected with subnet, also claim the network element interconnection, the corresponding network element device of each Control Node is finished the foundation that link connects and network element intersects.Control Node can be actual equipment, can coexist with the network element device of original its a control ground or separate of this equipment.Control Node also can be an independently logic software system, and its function is embedded in the network element device.
The embodiment of particular hardware part as shown in Figure 2, this hardware components is formed by transmitting plane and control plane, their are formation as follows respectively:
(1) transmit the plane: a plurality of SDH network element devices connect and compose a network by link, and this network is divided into two territories, and each territory is an abundant mesh net, between the territory by two parallel links.Other has a network management system equipment and transmits between network element and establishes a communications link, and network management system equipment is used for management and transmits network element, and the network management system functions of the equipments realize on a PC.
(2) control plane: be made up of a plurality of Control Node, each transmits the unique corresponding Control Node of network element.Each Control Node is an independent PC system.Between Control Node, and Control Node and transmission network element device and network management system equipment room, adopt the ethernet communication mode to realize the exchange of control information.
The concrete implementation step of the method for the invention technical scheme is as follows:
Step 1: described Control Node receives the normal connection request (also claiming initial connection request) that the fault recovery requirement is arranged from other Control Node or network management system equipment one, therefore the source transmission node of the connection that this connection request indicates and purpose transmission node claim the cross-domain connection that is connected to of being set up not on same territory.To source transmission node and purpose transmission node, it can be Gateway Network Element or the territory interior nodes of removing Gateway Network Element.The territory at transmission node place, source is called as the territory, source, and the territory at destination node place is called as the purpose territory.The Control Node of receiving connection request is the Control Node of source transmission node correspondence, this Control Node transmits destination node and the bandwidth resources requirement that is connected according to connecting, in the territory at source node place, determine normally to be connected the outlet gateway location in this territory: borde gateway network element, the outbound that leaves this territory are connected with outbound.
Step 2: adopt centralized or distributed route computing method calculate from the source transmission node to the territory that normally is connected inner outlet position, territory, source in the normal link road that connects by, the transmission network element corresponding control nodes of normal connection in the territory and the required process link connection resource and the network element interconnection resource of distributing then, and order the transmission plane to be transmitted and set up the actual physical connection, claim that this is the normal connection in the link field.
Step 3: to normally connecting in the territory of setting up, the Control Node of normal Control Node that connects the head-end network element and distal point network element is responsible for adopting traditional recovery technology method etc. to set up territory internal fault Restoration Mechanism, when application domain internal fault Restoration Mechanism when network element or link occurs fault in the territory that connects process in the territory.Traditional recovery technology method in the territory comprises: static 1+1 protection technology, service layer's resist technology, dynamic rerouting recovering technology etc.
Step 4: the Control Node of the above-mentioned outlet network element that is connected the territory, source to outbound the Control Node of transmission Gateway Network Element of next adjacent domains (below be called the 2nd territory) of process send and connect the request of foundation, this request is except the solicited message that comprises above-mentioned normal connection request, also comprise last one of connection and jump out a mouthful positional information: outbound link position, adjacent two territories are to the public identities of cross-domain normal connection: public connection ID.Control Node is after receiving this connection request, if the purpose of initial connection request transmits network element not on the 2nd territory, then by state step 1,2 identical methods are set up in the territory of striding the 2nd territory and are normally connected; Otherwise, calculate from connect through the 2nd territory transmit Gateway Network Element transmit to initial connection request purpose in the territory of network element normal link road by, link connection resource and the network element interconnection resource that the normal corresponding Control Node that connects the transmission network element of required process is distributed in the territory then, and order the transmission plane to set up the actual physical connection, thereby set up normal connection the in the territory.
Step 5: to normally connecting in the 2nd territory of setting up, the method identical with step 3 set up territory internal fault Restoration Mechanism, when application domain internal fault Restoration Mechanism when network element or link occurs fault in the territory that connects process in the territory.
Step 6: this step is finished and is protected establishment of connection between the territory; select normally to be connected the connecting portion branch is crossed over first territory in a intermediate node and the cross-domain intermediate node of connecting portion branch in second territory (intermediate node is meant two end nodes not considering the coupling part territory in) respectively as the source and destination of newly-built connection the (be called and protect connection between the territory); and consideration constraints: this connection and cross-domain connection of having set up (the normal connection of using public connection ID identification to set up) route separate (except that the end points that cross-domain protection connects; these two connect no public transmission network element and transmit link); normally is connected identical method and sets up cross-domain protection and connect with setting up by above-mentioned then, this newly-built connection with the relation that the original cross-domain part that normally is connected between described two intermediate nodes is connected is, and: the 1+1 that traditional route is separated or protect at 1: 1 concerns etc.When domain-to-domain link or gateway break down and influence cross-domain connection, can recover traffic failure like this by the traditional protection method.Because when the network planning, each territory can be abundant mesh net, connect number of routes N between promptly any two nodes greater than 1.N is big more, claims that then the adequacy of mesh net is strong more.For guaranteeing that all connect the full automation of setting up; the source and destination that the protection that can adopt the mode of prior agreement to specify to be used for fault recovery between the territory connects such as second jumping of selecting respectively normally to be connected in normal the last two jump that connects and second territory in first territory.In order to optimize the use resource; general select the normal jumping that connects near the 2nd territory in first territory, particularly select the last two jump, and the jumping in close first territory of normal connection in second territory; particularly second jump, respectively the source and destination that connect as protection between the territory.
This application diagram has comprised two mesh territories as shown in Figure 5, has two two the parallel links that connect different Gateway Network Element respectively between these two territories.The cross-domain normal connection of having set up is arranged among the figure, further,, set up respectively to protect between protection connection and territory in the territory in and be connected in order to consider fault recovery.Connection is by having subnet to connect (be meant network element enter the mouth the intersection of network element outlet) here and link connects to form, and the connection layout has here dispensed the subnet coupling part, but the not explanation of fault recovery principle between the domain of influence.
Step 7: the connection foundation of crossing over adjacent two territories finished after (comprising that protection connects between the territory); if also setting up, the connection that raw requests requires do not finish; then continue to adopt above-mentioned identical mode; foundation comprises between the territory the cross-domain connection that protection connects, and the whole connection of the connection source and destination that require up to raw requests is set up and finished.In addition, set up the protection connected mode, also can not consider to protect between the territory establishment of connection earlier, but after the connection foundation that raw requests requires finished, the protection of setting up one by one automatically between adjacent domains connected again about static state.
Step 8: the protection of described step 6 connects to set up and also can change the dynamic protection mode into: the Control Node of two promptly above-mentioned intermediate transport node correspondences can be responsible between the territory network failure taking place and when influencing cross-domain normal connection; dynamically set up with former cross-domain normal link road and be connected by the protection that separates; and business is switched to protection connect, thereby recover the fault business.If adopt dynamic reset mode, then need to stride in the process of normal connection of adjacent domains in foundation, specify two intermediate nodes in two territories to be respectively respectively: when normal connection fault, to carry out the source and destination of dynamic rerouting recovering.
Step 9: the static protection that described step 3, step 5, step 6 are set up connects also can adopt the virtual connections mode: it is a virtual connections that the protection of foundation connects, and the virtual connections of the cross-domain normal coupling part correspondence that wherein different devoids of risk is shared can be shared the transmission resource.When network failure takes place between the territory or in the territory and when influencing cross-domain connections, notice transmits the plane and sets up actual physical and connect, then business is switched to protection to connect, thus recovery fault business.Make advantage in this way be: to share occupation mode by Internet resources, make the Internet resources utilization that is optimized.The normal connection of this paper is meant: should connect the request of setting up, and the connection of transport service signal when fault-free of being set up, it is to protect between protection connection and territory in the above-mentioned relatively territory to be connected.Here, two shared implications of connection risk are: in case fault can influence two connections simultaneously.
As shown in Figure 3, network of forming by A, B, C, D, E, a F6 network element of example shown.At this normal connection c1 and c2 that two business are arranged respectively, the virtual connections of these two business is arranged respectively also simultaneously.SNP is to the on-position of link-attached endpoint location or service signal among the figure.With to be connected c1 the same with c2, virtual connections c1 ' and c2 ' also have network element to intersect respectively and link connects to form, it is empty that the network element of virtual connections that different is intersects, and promptly transmits network element device and does not set up this interconnection, or set up exchange and transmit and distribute respective resources.Constituting that the link virtually connect connects can be to exist transmitting the plane, though or do not exist transmitting the plane, required resource free time and Be Controlled plane are determined.Shown also among this figure that virtual connections c1 ' and virtual connections c2 ' can share a link and be connected, this be because they normal connection c1 and c2 not in same risk shared group.
Step 10: several recover territories and constitute big territories, and each Control Node has the network topological information in whole big territory, can calculate for the connection that will set up in the big territory link road by.Then request is set up in the initial connection in the above-mentioned steps 1, set up the cross-domain connection of the connection source and destination of no any protection earlier, the business recovery when use guarantees the territory internal fault with above-mentioned identical step method then.About protecting establishment of connection between the territory, remove routing mode and adopt source routing mode in the big completely territory, and do not need in conjunction with the route accessibility of borde gateway routing protocol mechanism release carry out pathfinding with above-mentioned different outside, other method is all identical.Looser when normal connection setup time is required, and service failure recovery time requirement can be adopted the method when relatively harsher.
Step 11: in above-mentioned steps, when determining to connect the domain-to-domain link of being crossed over when setting up cross-domain connection, in the time of also promptly need determining Gateway Network Element that this link strides across, the Gateway Network Element of selection preferably satisfies condition and intersects or risk is shared to the link of adjacent domains for: the domain-to-domain link between other territory between interconnected Gateway Network Element this Gateway Network Element of getting along well.Like this, foundation is connected by the protection that separates with former cross-domain normal link road.Obviously, when the parallel link (promptly non-intersect each other) of individual connection Gateway Network Element of N (N〉1) and non-shared risk is arranged between the territory, optionally select a Gateway Network Element, just can satisfy aforementioned condition.
Step 12: in the process that above-mentioned connection is set up; if used mechanism: in the territory, source in the reciprocal the 2nd jump with the 2nd territory in the 2nd jumping as protecting the source and destination that are connected between the territory; then can set up in the 2nd territory through after the connection in this territory; this territory connect first jump Control Node upstream (i.e. the connection source side that indicates towards raw requests to) Control Node send cross-domain normal connection and set up the notification message that finishes; this message comprises normal connection ID; the Control Node of intermediate NE is transmitted this notification message; and after this solicited message is received in the 2nd jumping reciprocal in the territory, source; automatically produce a protection connection request: this connection request contains a source (the reciprocal the 2nd jumps in the territory, source in) and a destination (the 2nd jumps in the 2nd the territory in), comprises request attribute: protect connection and normally the cross-domain normal link road that indicates of connection ID by separating.
Step 13: in above-mentioned establishment of connection process, can press priority from high to low, use following recovery mechanism successively further according to the priority of business:
Static protection connection, static protection virtual connections, dynamic heavy-route.This is because by order from left to right, the time of the service failure recovery that these Restoration Mechanism are finished increases successively gradually.
The specific embodiment of software section of the present invention is described as follows, and is as follows as Fig. 4 and treatment step shown in Figure 5:
(1) one Control Node CP_A receives the connection request that the fault recovery requirement is arranged from network management system equipment one, this connection request is from network management system equipment, requet0 has indicated source node T_A and destination node T_Z, and wherein the Control Node of source node T_A correspondence is CP_A.According to connecting destination node and the bandwidth resources requirement that is connected, in the territory at source node place, the route that forms according to Border Gateway Protocol can reach information, determines to be connected the exit position in this territory: borde gateway network element T_A2, the outbound L1 of normal connection that leaves this territory and the position that the link in the link L1 connects LinkConnection1.
(2) described Control Node CP_A calculate in the territory from source T_A to destination T_BG1 normal link road by.According to normal link road in the territory that calculates by, normal corresponding Control Node distribute links connection resource and the definite network element interconnection that connects the transmission network element (T_A, T_A1, T_A2) of required process in the territory, order transmit the plane and set up actual physical connection connection1.
(3) in the territory; adopt traditional protection technologies; the protection of foundation and normal physical connection connection1 correspondence is connected Pconnection1 (this order of connection is through network element T_A, T_A4, T_A2), and connection1 and Pconnection1 constitute the 1+1 protection relation.
(4) the Control Node CP_A2 that transmits network element T_A2 sends connection request request1 to the Control Node CP_Z6 that transmits network element T_Z6, this request is except comprising the requet0 content, also comprise last one of connection and jump out a mouthful positional information: outbound link position, adjacent two territories are to the public identities of cross-domain normal connection: connection ID.CP_Z6; receive request request1; earlier the link of setting up through node T_A2 and T_Z6 according to " connects last and jump out a mouthful positional information " is connected LinkConnection1, adopt then and above-mentioned steps 2 and 3 in identical method set up the cross-domain normal connection2 of being connected (this order of connection process network element T_Z6, T_Z3, T_Z) and the protection that concerns of formation 1+1 protection accordingly is connected Pconnection2 (this order of connection process network element T_Z6, T_Z4, T_Z).Arrive this, the connection of cross-domain two adjacent domains is established, and it comprises: connect in two sections territories with one section territory between normal link be connected.Connect in the territory and comprise that normal connection is connected with protection.CP_Z6 sends a notification message inform then, can upstream notify normal connection connection1 the Control Node of transmission network element of process: normal connect to set up finish, and carry information: public connection ID, normally connect connection2 and partly show routing iinformation (indicating this connections jumps and second jumping network element: T_Z6, the T_Z3 through first of process successively).
(5) after the last two jump T_A1 of connection1 receives inform, produce one automatically and connect foundation request requet2.This connection request has indicated essential information: 1) source node T_A1, destination T_Z3; 2) this new link road by be connected link road that public connection ID indicates by separation; 3) this new connection is the protection connection that public connection ID indicates connection.Then according to above-mentioned connection1, connection2, the step that LinkConnection1 is identical set up; and further consider the route separation requirement; set up cross-domain protection and connect, this connection is with original cross-domain normally to be connected the relation that node T_A1 is connected with part between T_Z3 be that the 1+1 protection that traditional route is separated concerns.
(6) when break down in territory A or inside, territory (non-gateway network element fault), utilize the 1+1 protection relation of setting up in the territory that the fault business is in time recovered.When between the territory link or gateway fails taking place, utilize the 1+1 protection relation of setting up between the territory that the fault business is in time recovered.
The inventive method compared with prior art, by above implementation process, it can be set up the static protection that is used for fault between protected field automatically and connects, thus when occur link failure or gateway fails between the territory, equal quick-recovery business soon.
Should be understood that the above-mentioned description at embodiment of the present invention is too concrete, can not therefore be interpreted as the restriction to scope of patent protection of the present invention, its scope of asking for protection should be as the criterion with claims.

Claims (10)

1. method of crossing over the network disaster recovery that multiple domain connects, it is applied to a system, described system comprises that one transmits a plane and a control plane, described transmission plane comprises a network that is connected and composed by link by a plurality of network element devices, this transmission network element is used for distributing or sets up link and connects, and set up the subnet connection, this network is divided into a plurality of territories; Described control plane comprises a plurality of Control Node, the unique corresponding Control Node of network element on described each transmission plane; It is characterized in that described method is further comprising the steps of:
A) described Control Node receives the normal connection request that has fault recovery to require from other Control Node or network management system equipment, the source transmission node of the connection that this connection request indicates and purpose transmission node be not on same net territory, this Control Node determines normally to be connected the outlet gateway location in this territory in the 1st territory at described source node place: borde gateway network element, the outbound that leaves this territory are connected with outbound;
B) in calculating from the source transmission node to the territory that normally is connected inner outlet position, described the 1st territory normal link road by, normal corresponding Control Node distribute links connection resource and the network element interconnection resource that connects the transmission network element of required process in the territory, and order then transmits the plane and sets up the actual physical connection;
C) to normally connecting in the territory of setting up, normally connect the Control Node of head-end network element and the Control Node of distal point network element and be responsible for setting up territory internal fault Restoration Mechanism;
D) Control Node of the above-mentioned outlet network element that is connected the 1st territory to outbound process next adjacent domains promptly the Control Node of the transmission Gateway Network Element in the 2nd territory send and connect the request of foundation, after Control Node received this connection request, the outlet network element in the 1st territory that connects earlier was connected with the normal of transmission Gateway Network Element in the 2nd territory; If the purpose of initial connection request transmits network element not on the 2nd territory, then set up normally connection in the territory of striding the 2nd territory by above-mentioned steps; Otherwise, calculate from connect through the 2nd territory transmit Gateway Network Element transmit to initial connection request purpose in the territory of network element normal link road by, the corresponding Control Node distribute links connection resource and the network element interconnection resource that normally connect the transmission network element of required process then in the territory, and order the transmission plane to set up the actual physical connection, thereby set up normal connection the in the territory;
E) to normally connecting in the 2nd territory of setting up, the method identical with described step c) set up territory internal fault Restoration Mechanism, when application domain internal fault Restoration Mechanism when network element or link occurs fault in the territory that connects process in the territory;
F) finish and protect establishment of connection between the territory: intermediate node selecting normally to be connected connecting portion branch leap first territory in and the cross-domain intermediate node of connecting portion branch in second territory are respectively as the newly-built source and destination that are connected; and consideration constraints: this connection and the cross-domain link road of having set up are set up cross-domain protection then and are connected by separating.
2. method according to claim 1; it is characterized in that; after the connection foundation of crossing over adjacent two territories finishes; if also setting up, the connection that raw requests requires do not finish; then continue to set up the cross-domain connection that comprises that protection connects between the territory, the whole connection of the connection source and destination that require up to raw requests is set up and is finished.
3. method according to claim 1 is characterized in that, recovery mechanism adopts between described territory: traditional static protection mode, static state virtually connects protected mode or dynamic heavy-route protected mode;
Described traditional static protection mode refers to that it is real connection that protection connects, protection connect not domain of dependence fault and with the territory between normally is connected all the time existence jointly accordingly, and the traditional protection mode that business is protected;
Described static virtual connections protected mode refers to that it is virtual connections that protection connects, and protection connect do not rely on fault and with the territory between normally is connected all the time existence jointly accordingly, after normal connection fault, virtual connections is become actual physical to be connected, business switches in the protection connection then, thereby recovers the fault business;
Described dynamic heavy-route protected mode refers to not set up in advance protection and connects; after Gateway Network Element fault or inter-domain link fault occurring; dynamically foundation is connected by the protection that separates with former cross-domain normal link road, and business is switched in the protection connection, thereby recovers the fault business.
4. method according to claim 1 is characterized in that, described method is after the connection foundation that raw requests requires finishes, and the protection of setting up one by one automatically between adjacent domains connects again.
5. method according to claim 1 is characterized in that, described method comprises that also several are recovered the territory constitutes a big territory, and each Control Node has the network topological information in whole big territory, can calculate for the connection that will set up in the big territory link road by.
6. method according to claim 1, it is characterized in that, when described method is determined to connect the domain-to-domain link of being crossed over when setting up cross-domain connection, in the time of promptly need determining Gateway Network Element that this link strides across, the Gateway Network Element of described selection satisfies condition: this Gateway Network Element of getting along well of the domain-to-domain link between other territory between interconnected Gateway Network Element intersects or risk is shared to the link of adjacent domains.
7. method according to claim 1; it is characterized in that; described method connects the full automation of setting up for guaranteeing all; the source and destination that the protection of adopting the mode of agreement in advance to specify to be used for fault recovery between the territory connects, select normal second jumping that is connected in normal the last two jump that connects and second territory first territory in respectively.
8. method according to claim 1; it is characterized in that; in above-mentioned establishment of connection process; further according to the priority of business; press priority from high to low, use following each recovery mechanism successively: static protection connection, static protection virtual connections, dynamic heavy-route protected mode.
9. method according to claim 1, it is characterized in that, in the described step d) the Control Node of the outlet network element in the 1st territory to outbound the Control Node of transmission Gateway Network Element of next adjacent domains of process send and connect when setting up request, this request comprises outside the solicited message of above-mentioned normal connection request, also comprise last one of connection and jump out a mouthful positional information: outbound link position, adjacent two territories are to the public identities of cross-domain normal connection: public connection ID.
10. method according to claim 1 is characterized in that described route computing method comprises centralized or distributed.
CNB2004100918310A 2004-12-24 2004-12-24 Method of network fault recovery crossing over connections in multiple domains CN100373866C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2004100918310A CN100373866C (en) 2004-12-24 2004-12-24 Method of network fault recovery crossing over connections in multiple domains

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2004100918310A CN100373866C (en) 2004-12-24 2004-12-24 Method of network fault recovery crossing over connections in multiple domains

Publications (2)

Publication Number Publication Date
CN1798051A CN1798051A (en) 2006-07-05
CN100373866C true CN100373866C (en) 2008-03-05

Family

ID=36818847

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2004100918310A CN100373866C (en) 2004-12-24 2004-12-24 Method of network fault recovery crossing over connections in multiple domains

Country Status (1)

Country Link
CN (1) CN100373866C (en)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100490390C (en) * 2006-01-10 2009-05-20 华为技术有限公司 Service failure recovery method
CN100433661C (en) * 2006-12-22 2008-11-12 中国联合通信有限公司 A soft switch network and its networking method
CN101325542B (en) * 2007-06-15 2011-09-14 华为技术有限公司 Method for acquiring PCE capability information between fields, PCE and capability-acquiring apparatus
CN101715149B (en) * 2009-07-21 2013-01-23 北京邮电大学 Method and device for restoring parallel cross-domain fault of multilayer and multi-domain distributed optical network
US10225178B2 (en) 2010-09-15 2019-03-05 Telefonaktiebolaget Lm Ericsson (Publ) Methods and apparatus for relocating and restoring connections through a failed serving gateway and traffic offloading
US8848516B2 (en) * 2010-09-15 2014-09-30 Telefonaktiebolaget L M Ericsson (Publ) Methods and apparatus for relocating and restoring connections through a failed serving gateway and traffic offloading
CN103493434B (en) * 2011-02-22 2018-02-23 瑞典爱立信有限公司 Fault protecting method and fail-safe system in multiple-domain network
CN102546389B (en) * 2011-11-08 2015-01-14 杭州华三通信技术有限公司 Method and device for flow trusteeship of cross-data center
EP3001612B1 (en) * 2013-06-29 2017-08-09 Huawei Technologies Co., Ltd. Multi-domain network protection method and node
CN103428031B (en) * 2013-08-05 2016-04-13 浙江大学 A kind of inter-domain link fast failure recovery method based on software defined network
US10009287B2 (en) * 2013-12-26 2018-06-26 Huawei Technologies Co., Ltd. Hierarchical software-defined network traffic engineering controller

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1189269A (en) * 1995-06-26 1998-07-29 艾利森电话股份有限公司 Self-healing network
CN1292193A (en) * 1998-03-11 2001-04-18 西门子公司 Local network with redundancy properties, especially ethernet network and redundancy
US20020131409A1 (en) * 2001-03-13 2002-09-19 Frank David L. Self-healing multi-level telecommunications network
CN1371201A (en) * 2001-02-22 2002-09-25 杨震 Asynchronous transferring virtual channel self-healing ring
CN1440159A (en) * 2003-03-31 2003-09-03 中国科学院计算机网络信息中心 Control method for layered switching network node region
CN1512716A (en) * 2002-12-30 2004-07-14 北京邮电大学 Method for multilayer network to recover from

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1189269A (en) * 1995-06-26 1998-07-29 艾利森电话股份有限公司 Self-healing network
CN1292193A (en) * 1998-03-11 2001-04-18 西门子公司 Local network with redundancy properties, especially ethernet network and redundancy
CN1371201A (en) * 2001-02-22 2002-09-25 杨震 Asynchronous transferring virtual channel self-healing ring
US20020131409A1 (en) * 2001-03-13 2002-09-19 Frank David L. Self-healing multi-level telecommunications network
CN1512716A (en) * 2002-12-30 2004-07-14 北京邮电大学 Method for multilayer network to recover from
CN1440159A (en) * 2003-03-31 2003-09-03 中国科学院计算机网络信息中心 Control method for layered switching network node region

Also Published As

Publication number Publication date
CN1798051A (en) 2006-07-05

Similar Documents

Publication Publication Date Title
US10250459B2 (en) Bandwidth on-demand services in multiple layer networks
US10148349B2 (en) Joint IP/optical layer restoration after a router failure
Kodian et al. Failure-independent path-protecting p-cycles: Efficient and simple fully preconnected optical-path protection
Fumagalli et al. IP restoration vs. WDM protection: Is there an optimal choice?
CN104158747B (en) Method for discovering network topology and system
KR100462408B1 (en) A Fast Re-route Method Using GMPLS in Optical Transport Networks
US7558194B2 (en) Virtual private network fault tolerance
Ye et al. A simple dynamic integrated provisioning/protection scheme in IP over WDM networks
US7274869B1 (en) System and method for providing destination-to-source protection switch setup in optical network topologies
JP4145025B2 (en) Backup path setting method and apparatus
US7046619B2 (en) Method and system for bi-directional path switched network
Kiaei et al. A survey on the p-cycle protection method
EP1491000B1 (en) Network management system
DE60319215T2 (en) Restoration of a network
JP3714329B2 (en) Variable capacity link device and variable capacity link setting method
US9848049B2 (en) Service preemption selection systems and methods in networks
JP3762749B2 (en) Restoration protection method and apparatus
EP1391068B1 (en) Restoration protection in communication networks
CN101371150B (en) Dynamic protection against failure of a head-end node of one or more TE-LSPs
EP1470679B1 (en) Method and apparatus for multi-layer network in sonet /sdh
JP5180972B2 (en) Method and apparatus for network tree management
US7372806B2 (en) Fault recovery system and method for a communications network
CN101710877B (en) Method, device and system for processing traffic flow based on pseudowire
US7043250B1 (en) Systems and methods for forming and operating a communications network
JP4661892B2 (en) COMMUNICATION NETWORK SYSTEM, COMMUNICATION DEVICE, ROUTE DESIGN DEVICE, AND FAILURE RECOVERY METHOD

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant