CN101449562A - Method and system for protecting a sub-domain within a broadcast domain - Google Patents

Abstract

The present invention provides a method and system for protecting a service available on a broadcast domain. A sub-domain is established within the broadcast domain. The sub-domain includes a group of nodes used to provide a communication path to the service. A primary sub-domain maintenance association and a back-up sub-domain maintenance association are monitored. The primary and sub-domain maintenance associations are a set of primary and back-up paths, respectively, representing connectivity between nodes acting as edge nodes in the sub-domain. A fault is detected within the primary sub-domain maintenance association and a switch to the back-up sub-domain maintenance association occurs.

Description

Be used to protect the method and system of the subdomain in the broadcast domain

Technical field

The present invention relates to network service, relate in particular to the method and system that is used for protecting the point-to-point and multiple spot connection that forms network sub-domain, network sub-domain is the part of broadcast domain, can find in the communication network of for example internet protocol-based (" IP ").

Background technology

A kind of environment has been created in popularizing of communication Network Based (for example use transmission control protocol/Internet protocol " TCP/IP " those), wherein shares physical resource by the service provider and has become general to hold different clients.For example, the service provider supplies Virtual Local Area Network (" VLAN ") service, and the logical layer connection separates for each client with communication in this service, even these clients share the communication of actual physics layer, for example Ethernet switches hardware, cable etc.

Broadcast domain is by transmitting the zone of the accessibility network of frame that just is being broadcasted.Like this, with regard to VLAN, the frame that is broadcasted, the frame that for example has the destination of unknown unicast address, broadcast or multicast, be sent to the device in VLAN (or LAN) and receive, but be not device on other VLAN or the LAN, even they are parts of same physical network by it.Correspondingly, LAN and multiple spot VLAN are the examples of " broadcast domain ".Broadcast domain can be the zone in the multiple spot Ethernet, and the frame that wherein has the destination of unknown unicast, broadcast or multicast is broadcasted.

Institute of Electrical and Electronic Engineers (" IEEE ") 802.1Q standard correction is set up the parameter that is used for packet-based trunk bridge coil as 802.1ad and 802.1ah standard.Although the operation and management responsibility of extensive service provider network can physically be demarcated to allow regional project management physical basis structure, the viewpoint of being disposed from service depending on circumstances is not like this.Like this, from service point, these standards can not set up a kind of method and arrive less than other any territory of broadcast domain level to be used to provide backup protection.The result is, than the broadcast domain rank of small grain size more, owing to monitor and the management service poor availability, caused inefficient reserve to provide.

For example; comprise division multilink relaying (" SMLT ") and aggregation of links although be used to provide the proposal of the big regular network of backup protection (such as huge Ethernet); but these propose also not satisfy service provider's needs; because they are not deterministic; be developed to the requirement of the original applications that satisfies them, i.e. enterprise network.

Desirable is that a kind of certainty arranges that wherein broadcast domain can be segmented based on multiple unique VLAN layout that the common service end points for example is provided.Here the service of mentioning both can have been made a comment or criticism in the user's who is supplied to provider's network peer-to-peer services, also can refer to the facility by provider's use of supply peer-to-peer services.What further wish is that described layout provides one of these unique VLAN layouts with the predominating path as peer-to-peer services data (being called " business " here), not too is suitable for providing under the situation of wishing to serve at predominating path having one or more unique VLAN layouts to be used for business.There is also a desire for a kind of layout, it is provided at quick switching between these VLAN with the overseas device transparent way of antithetical phrase under failure condition.

Summary of the invention

The present invention advantageously provides a kind of method and system and is used for protecting the service that broadcast domain can be used of striding.Main is based upon in the broadcast domain with at least one reserve subdomain, by foundation and the supervision of subdomain maintenance association (" SDMA "), in the visit of subdomain rank reserve to service.SDMA is the set in point-to-point connection/path, for example medium access control (" MAC ") layer destination, source, and the connectedness of representative between the fringe node of subdomain, and set up to be used for the main and reserve subdomain in the maintenance field.The fringe node of subdomain can be the fringe node or the core node of broadcast domain.Each subdomain protection group (" SDPG ") has main and reserve SDMA; and provide the logic handover mechanism; when occurring in the node in the link in the path in the main SDMA or path with convenient fault, envoy's grouping route of naming a person for a particular job switches to reserve SDMA from main SDMA.

According to an aspect, the invention provides a kind of service method available on broadcast domain that is used for protecting.Subdomain is based upon in the broadcast domain.Subdomain comprises a group node of the communication path that is used to provide described service.Main subdomain maintenance association and reserve subdomain maintenance association are monitored.Main and reserve subdomain maintenance association is respectively the set of main and back-up path, represents in subdomain as the connectedness between the node of fringe node.In main subdomain maintenance association, detect fault, and proceed to the switching of reserve subdomain maintenance association.

According to another aspect, the invention provides a kind of storage medium of storage computation machine program, when computer program is performed, finish a kind of service method available on broadcast domain that is used for protecting, wherein subdomain is based upon in the broadcast domain.Subdomain comprises a group node of the communication path that is used to provide described service.Main subdomain maintenance association and reserve subdomain maintenance association are monitored.Main and reserve subdomain maintenance association is respectively the set of main and back-up path, represents in subdomain as the connectedness between the node of fringe node.In main subdomain maintenance association, detect fault, and proceed to the switching of reserve subdomain maintenance association.

According to another aspect, the invention provides a kind of system that is used to provide service available on broadcast domain.A plurality of nodes are arranged to be provided to the subdomain of the communication path of described service.Each node has storage device and CPU.Memory device stores and main subdomain maintenance association and the corresponding data of reserve subdomain maintenance association.Main and reserve subdomain maintenance association is respectively the set of main and back-up path, represents in subdomain as the connectedness between the node of fringe node.The CPU operation is to detect the fault in main subdomain maintenance association and to switch to reserve subdomain maintenance association.

Description of drawings

Of the present inventionly understand more comprehensively and follow advantage and feature by being more readily understood when connection with figures is considered with reference to following detailed description, in the accompanying drawings:

Fig. 1 is the calcspar according to the system of principle construction of the present invention;

Fig. 2 is the calcspar according to the subdomain of principle construction of the present invention;

Fig. 3 is a chart, is illustrated in the relation in the subdomain maintenance association;

Fig. 4 is a chart, the subdomain maintenance association state machine of expression example; And

Fig. 5 is a chart, and expression is used for the subdomain maintenance association sight of the example of subdomain protection group.

Embodiment

Referring now to accompanying drawing, that like reference numerals indication like wherein, Fig. 1 are represented to build according to principle of the present invention and refer generally to be shown the calcspar of the system of " 10 ".System 10 comprises broadcast domain 12.Broadcast domain 12 comprises one or more subdomains, for example subdomain X14a, subdomain Y 14b, and subdomain Z 14c (being referred to as here as subdomain 14).Subdomain 14 each self-defined subdomain.

Subdomain is the subclass of node, and node is the part of broadcast domain.All nodes in the subdomain are the set of node that the transmission of a Service Instance or a plurality of Service Instances is provided by network (for example Ethernet).In other words, subdomain is the part (or all) of broadcast domain, and broadcast domain is based on the service of this part of using broadcast domain.As used herein, term " service " is applied to end-to-end connectedness, wherein connectedness can be point-to-point, multiple spot and point-to-multipoint, be supplied to the user of broadcast domain or facility (for example main line), be used in the broadcast domain to carry integrally or partly relevant business with end-to-end connectedness.

As used herein, term " territory " is the foundation structure with multiple spot connectedness, can be used to supply point-to-point, multiple spot and point-to-multipoint connectivity services, if need require like this based on system design.As one aspect of the present invention, subdomain can be the subclass of node, and it is the part of broadcast domain but to need not to be physics adjacent.In other words, between groups of nodes, logical relation can be arranged, thereby the visit of service is self-contained in the subdomain and does not consider physical connection.As another aspect of the present invention, subdomain can be the subclass of node, its be the part of broadcast domain and in handoff environment physics adjacent.In other words, between groups of nodes, physical relation can be arranged, thus needn't be self-contained in subdomain to the visit of service.

Each subdomain comprises a group node 16, and it is defined in the path between the fringe node in the subdomain 14.Meriting attention, according to the service of supporting between fringe node and to supporting the needs of different specific nodes 16 of serving, might node 16 be parts of a plurality of subdomains 14.For example, might node 16 can support to share shared end points or port but be associated with different subdomains and by two kinds of separate service of its protection.

Represent and describe example subdomain 14 with reference to Fig. 2.The subdomain 14 that Fig. 2 represents comprises node S1 16a, S2 16b, S3 16c, S4 16d and S5 16e.Node S1 16a, S2 16b, and S516e be fringe node, make the user arrive network interface (" UNI ") 18 and network service port P1 and P2, corresponding with the service in being self-contained in subdomain.Also be susceptible to, one or more nodes 16 can be the fringe nodes that the subdomain of network-network interface (" NNI ") is provided for same services example (not shown).Also be susceptible to, as another example, subdomain can comprise not having any UNI 18 interfaces and only provide the node 16 of network to the NNI interface for one or more Service Instance (not shown)s.The physical composition of node 16 can be network service switch or any other interconnection device that is suitable for implementing function described herein.Node 16 comprises suitable CPU, easy become estranged non-volatile memory memory and be arranged as the interface of finishing function described herein.

By customer set up (or customer network itself) is connected to fringe node through the UNI 18 identical with UNI on the broadcast domain, support peer-to-peer services.A service of subdomain protection or one group of Service Instance.As the node 16 of service end node, in subdomain also by " M " the preceding indication that stops.Fig. 2 represents main subdomain of being indicated by solid line connected node 16 and the reserve subdomain of being indicated by dotted line connected node 16.For example, the predominating path between node S1 16a and the S5 16e is through node S3 16c, and the back-up path between node S1 16a and the S5 16e is through node S4 16d.

Subdomain maintenance association (" SDMA ") is defined as the set of paths of representative connectedness between fringe node (for example node S1 16a and S5 16e) in subdomain 14.Arriving the state in the path of remote node in the subdomain is represented by remote maintenance associated endpoint (" RMEP ") state.This RMEP be ITU-T Y.1731 with the example more specifically of the defined MEP of IEEE 802.1ag, corresponding in logic not with to its SDMA just by the mutually juxtaposed MEP of device that popularizes.The state of SDMA is derived by collective's state of the RMEP that is associated with the SDMA of each node.Certainly, be appreciated that RMEP can be associated with a plurality of SDMA.Such situation occurring is that subdomain can be overlapping because as discussed above, promptly shares identical and/or end points.Be noted that also SDMA can comprise the subclass of the RMEP that is monitored by maintenance association (" MA ").

Defined the set of paths of representative connectedness between fringe node 16 subdomain 14 in, definable and explanation are used for providing for service available on the network protection and the grouping of backup protection.In order to protect the group that the visit of service is set up in subdomain to be defined within the subdomain protection group (" SDPG ").The node that comprises example SDPG is displayed on Fig. 2, and explains with reference to Fig. 3.Subdomain protection relation table 20 is the parts with the SDPG of main and reserve SDMA configuration.Yet service is associated with SDPG itself.For example, the Service Instance that is used for provider's backbone bridges network is service identifier (" SID ").When link or the node when taking place of fault in SDMA, SDPG is provided at the handover mechanism between main and the reserve SDMA.

SDPG can be expressed as the table such as table 20, and representative is about the protection group relation of node (for example node S1 16a).Other node has their table and data structure.In keeping territory 22, safeguard relevant 24 respectively with respect to main and reserve subdomain 26 and 28 and set up.Maintaining end point is meant the node 16 (" MEP ") of the end, path in the subdomain.With reference to Fig. 2, MEP M1, M2 and M5 rely on their positions as the end points in the describing example subdomain 14, indicated respectively and with node S1 16a, S2 16b, and S5 16e corresponding.Might node S3 16c as the maintaining end point of different and subdomain that do not describe.

Subdomain protection concerns that 20 are shown with respect to MEP M1.Be appreciated that other subdomain concerns that 20 can make up other MEP that is used for subdomain, for example be used for the subdomain protection relation of MEP M5.For main subdomain 26, the subdomain protection that is used for MEP M1 concerns that 20 comprise RMEPM2, RMEP M5 and RMEP M7.As Fig. 2 saw, RMEP M2 was corresponding with S216b, and RMEP M5 refers to node S516e.Correspondingly, can arrive and each RMEP of being associated with SDMA is provided in the subdomain protection relation table 20.Table 20 is stored in corresponding node (in this case, in node S116a).Merit attention, RMEP M7 is presented in main subdomain 26 and the reserve subdomain 28.RMEP M7 is the part of whole maintenance association 24, but is not defined as the part of the subdomain that Fig. 2 and 3 describes.The present node that RMEP and MEP definition refer to remote site respectively and be considered is as stating among ITU-tY.1731 and the IEEE 802.1ag.

To Fig. 3, during point-to-point path in fault occurs in SDMA, SDPG is provided at the handover mechanism between main and the reserve SDMA.As Fig. 3 showed, main SDMA30 and reserve SDMA 32 (being associated with RMEP M2, RMEP M5 and RMEP M7 separately) were associated with subdomain protection group 34.The subdomain protection is organized 34 autoprotections and is provided serving the visit of A36 and B38.Be used between main subdomain 26 and reserve subdomain 28, switching and monitor and switch therebetween to provide mechanism to be discussed in more detail below to the visit of serving A36 and B38.Merit attention,, be appreciated that in SDPG and can support any amount of service although have only two kinds of services to be presented at Fig. 3.Similarly, be subjected to the processing and the storage restriction of node 16, any amount of RMEP also can be associated with specific sub-domains protection group.

Advantageously, according to one embodiment of present invention, do not need new MEP for subdomain protection with respect to the MEP that defines in the existing standard.Such situation occurring is because subdomain MEP is used for the whole subclass that monitors the needed territory MEP of foundation structure facility in broadcast domain.The selection of the respective subset of SDMA and territory MEP is based on the needs that protection is provided for special services subclass (its be on the foundation structure facility of the broadcast domain in service provider's the network in carrying and the whole set of service supported).As shown in Figure 3, the MEP that is associated with SDMA is disposed in the same endpoints of foundation structure facility, node S1 16a for example, and wherein related service and their respective communication enter, emit.

According to another embodiment of the invention, the new MEP identical with the MEP that defines in the existing standard is created and is used for the subdomain protection.Such situation occurring is because subdomain MEP is used to be independent of the whole mode of the needed territory MEP of foundation structure facility in broadcast domain that monitors.SDMA MEP is placed on the fringe node of subdomain, thinks that special services subclass (it is in the whole set of service of carrying and supporting of the foundation structure facility in the broadcast domain in service provider's network) provides protection.Some or all of these SDMA MEP can shared domain MEP same endpoints, when fringe node 16 was supported UNI 18, wherein related service and their correspondence were communicated by letter and are entered, emit.Do not support UNI18 but when supporting the fringe node 16 of NNI to locate SDMA MEP, end points is not shared with territory MEP when striding across.According to this embodiment of the invention, under than the high speed of the speed of using the wide maintenance association of territory MEP supervision field width, carry out SDMA by SDMA MEP and monitor.

As discussed in more detail below, the fault in the subdomain 14 monitors that by using loop supervision message (as loop verification message or " CCM ") particular remote MEP is detected by having " M " preceding node that stops at the MEP place of Fig. 3 indication.CCM is not explained here in detail by International Telecommunications Union (" ITU ") and IEEE definition.Notice that CCM is the particular instance of loop supervision message, the use here is intended to and broad terms " loop supervision message " synonym more.Merit attention, MEP can describe the loss of communicating by letter with the RMEP that uses clean culture/multicast CCM.Yet, MEP can not detect by use multicast CCM may be just at the specific RMEP of detection failure.Such situation occurring is because described specific RMEP (it contributes fault) is not transmitted in the remote defect identification (" RDI ") that receives, and has detected fault and only transmit RMEP.Yet,, might determine whether RMEP is experiencing the problem of communicating by letter with local MEP if use clean culture CCM.

About monitoring main and reserve SDMA, for example with main subdomain 26 and reserve subdomain 28 corresponding SDMA.Go through the actual SDMA state of contact the present invention definition below.Usually, when detecting the fault in main SDMA, make and switch decision corresponding with service is switched to the reserve connection of subdomain.Switch the state that reserve SDMA is also depended in decision, because if relevant for the problem of reserve, as network or node interruption etc., what meaning then switches to reserve SDMA does not have.Certainly, be susceptible to also and can use recovery scheme, thereby because the fault of main SDMA proceeds to the protection of reserve SDMA when switching, when main SDMA where applicable once more, the main connection is resumed.Yet such recovery scheme and can be used any available recovery scheme outside scope of the present invention.

In order to realize switching, must the node in subdomain keep the information of RMEP and SDMA state from main subdomain to the reserve subdomain.The most at the beginning, node (for example node S116a) is arranged to have the MEP of establishment to send periodically clean culture CCM.In operation, periodically clean culture CCM is sent to remote node from each node in subdomain.For example, with regard to node S116a, this node periodically clean culture CCM sends to M2 and M5 (being respectively node S2 16b and S5 16e).Situation also is like this with regard to VLAN.If remote node is come a plurality of subdomains on particular source node, then send single CCM message for all SDMA that are associated with remote node.

Person of skill in the art will appreciate that, the invention is not restricted to here more than specifically expression and the content described.In addition, unless more than carry out opposite mentioning, should note all nots to scale (NTS) of institute's drawings attached.In view of the various modifications and changes of above instruction all are possible, and do not break away from the scope and spirit of the present invention that only limit by following claims.

Determine the state of each RMEP.The state of RMEP is determined by receiving the CCM that sends from other node on each node.If the CCM of predetermined quantity is not received, think that then RMEP shuts down and move to malfunction in specified time period.If detect the RMEP fault, then remote defect sign (" RDI ") message is sent to the remote node that is associated with fault RMEP in the unicast messages of appointment, with the notification signal fault detect, guarantee that thus two end points in the path in subdomain all detect unidirectional fault and other fault.

Collective's state of the RMEP that the representative of SDMA state is associated with the SDMA of intranodal.For example, with reference to Fig. 3, node S1 16a safeguards the state of RMEP M2, RMEP M5 and RMEPM7.State about the maintenance of main subdomain 26 relevant 24 is maintained among the node S1 16a of this intranodal.Like this, if the fault of detecting, whether the fault or the RMEP M5 that can not communicate by letter at least of the form of then storing in S1 16a indication RMEP M5 move to determining of reserve subdomain with communication thereby can make.

The present invention defines a plurality of SDMA states." IS " state means that SDMA can use managerially in service and to other node 16 in the subdomain, promptly RMEP can provide full service." IS-ANR " state means SDMA managerially in service, but can not provide full service to some paths (being RMEP) of other node in subdomain.In other words, the one or more RMEP in SDMA are outside service (" OOS ").Y.1731, this can detect with IEEE 802.1ag agreement by using ITU-T.

" OOS-AU " state means SDMA managerially in service, but the path (being RMEP) of other node can not provide full service in the subdomain.In other words, all RMEP in SDMA outside service, can detect as using IEEE 802.1ag." OOS-MA " state means SDMA managerially outside service, and all paths to other node can both provide full service in subdomain.In other words, all RMEP are in service, but SDMA is managerially outside service." OOS-MAANR " state means SDMA managerially outside service, and some paths of other node can not provide full service in the subdomain but have only.In other words, the one or more RMEP in the SDMA outside service, as by ITU-T Y.1731 with IEEE 802.1ag protocol detection.At last, " OOS-AUMA " state means SDMA managerially outside service, and all can not provide full service to all paths of interior other node of subdomain.In other words, Y.1731 all RMEP in the SDMA outside service, detect with IEEE 802.1ag agreement as using ITU-T.

Use these states, SDMA state one by one moves.For example, if all RMEP detect to being out of order, then the SDMA under " IS " state is movable to " OOS-AU " state.Similarly, all RMEP have been out of order but the situation recovered can make the SDMA state move to " IS " from " OOS-AU ".Correspondingly, can create the state table of the state of expression subdomain, example is shown as the state machine 40 of Fig. 4.

Whether the state of RMEP state and definite RMEP has changed employed information can be finished by monitoring from RMEP reception CCM, and the enforcement of can programming in corresponding node 16.For example, expiring of predetermined time interval can be used to trigger the indication that RMEP has been out of order and has not received CCM.Similarly, shorter threshold time section can be used to indicate the degeneration of RMEP communication performance, has perhaps indicated fault.For example, predetermined amount of time can be established, can indication fault thereby in three time intervals, fail to receive CCM, and the reception of the CCM between two and three time intervals can be used to refer to the degeneration communication performance about RMEP.Based on the detection of RMEP event of failure, if fault necessitates state variation, then the state of SDMA state machine is renewable.CCM sends on inherent each the destination end points of broadcast domain (it can be defined by VLAN).

As the another kind selection that is used for safeguarding RMEP and SDMA state, multicast CCM and clean culture CCM can identify (" RDI ") by means of remote defect and be used to refer to the defective form.In this case, periodically multicast CCM is used for all other MEP receptions from each node transmission.Select as clean culture CCM discussed above, multicast CCM sends by each VLAN, thereby if remote node is shared (BTAG) for a plurality of subdomains of sharing VLAN, has only a CCM periodically to be sent to VLAN.Select as clean culture CCM, the RMEP state is determined by receiving the CCM that sends from other node.If detect the RMEP fault, the remote node of then indicating the clean culture CCM of RDI also periodically to be sent to be associated with RMEP guarantees thus that with the notice fault detect two end points in path all detect unidirectional fault and other fault in subdomain.For this pattern, CCM sends on each source MEP, and is multicasted to all RMEP in the broadcast domain.Broadcast domain generally can be defined by VLAN.In other words, multicast CCM is sent by each MEP.Be out of order if suspect a RMEP, the MEP that then detects fault also will indicate the clean culture CCM of RDI to send to specific suspicion RMEP.

As another selection, RMEP and SDMA state can safeguard that its use has the multicast CCM of RMEP faulty indication, by the use of multicast CCM and RDI and the maintenance of the long-range MEP tabulation of fault.In this case, as preceding two kinds were described, MEP was created to send periodic multicast CCM message.Similarly, multicast CCM sends on each VLAN rank.The state of RMEP is determined by receiving the CCM that sends from other node on each node.If the message of predetermined quantity does not receive, then RMEP is moved to the state of being out of order in specified time period.If the RMEP fault is detected, then multicast CCM message comprises RDI and is the tabulation of out of order RMEP after testing.This information can be used for upgrading their state table by other remote node.

Certainly, CCM upgrades and the purpose of state variation is to allow the part (being subdomain) of broadcast domain to switch to the reserve subdomain from main subdomain, and vice versa, to keep service and visit running and operation to serving.Fig. 5 represents to be used for the example sight of provider's backbone network, has SDMA that is used for main subdomain " broadcast domain 1 " and the 2nd SDMA that is used for reserve subdomain " broadcast domain 2 ".Three RMEP of example supposition that Fig. 5 represents.Like this, in the sight 1 in the example of expression, main and reserve SDMA is in service, so the use of SDPG forward-facing state display of broadcast territory 1 (being main subdomain).Sight 2 is illustrated in the example of RMEP (being RMEP 2) outside service on the reserve subdomain.Correspondingly, the state of reserve subdomain is set to " IS-ANR ", and forward-facing state remains main subdomain.On the contrary, sight 3 is illustrated in RMEP 3 in the main subdomain outside service status, thereby the state of main subdomain is set to " IS-ANR ".In this case, the SDPG forward-facing state is configured to use the reserve subdomain, because RMEP3 is in the service of using the reserve subdomain.

Main and all out of order situation of reserve SDMA of sight 4 expressions.In this case, the SDPG forward-facing state remains broadcast domain 1, because no matter use which SDMA that fault is all arranged.Yet, also imagine the SDPG forward-facing state and can be configured to use SDMA with minimum number fault.Under the situation of sight 4, this means and use reserve SDMA, because it has only single fault, i.e. the fault of RMEP3.

RMEP among the main SDMA of sight 6 expressions is outside service status.In this case, the SDPG forward-facing state is configured to use reserve SDMA.Certainly, the sight of representing at Fig. 5 only is an example, because the quantity of RMEP and possible failure scenario are much larger than the example of describing.

Use above the explanation, clearly, switching is based on interested subdomain.For example, as discussed above, might specific node 16 can participate in subdomain 14 more than one.Correspondingly, the fault of this node or may mean and force variation more than the reserve subdomain of a subdomain to the fault of the link of this node.This may influence the availability more than a kind of service again.Similarly, fault that might specific node 16 or may not influence service in the subdomain to the fault of the link of node 16.Correspondingly, if a certain section in the subdomain has been detected as fault, then only carry out switching from main to reserve SDMA.This can be by explaining with reference to Fig. 2.

Although do not show, suppose that node S4 16d supports the different service of being supported through UNI 18 with node S1 16a, S2 16b and S516e of service.Fault on the link between node S1 16a and the S4 16d can not influence the service that can use through UNI 18, if subdomain is used as its main link with the link between node S116a and the S4 16d, then may influence service and visit.Under such a case, the subdomain that is supported in the service on the S4 16d can be seen the state variation of main SDMA, and need switch to reserve SDMA, perhaps uses the route through node S3 16c and S5 16e.Under such a case, the service on SDMA is unaffected, and uses other available service of other SDMA to be affected.Advantageously, owing to monitor and switch and carry out in the subdomain rank according to the present invention, so service impacting variation can be by grain refined, and the influence that causes minimizes in the best domain of broadcast domain.

According to another aspect of the present invention; when SDMA MEP being arranged in the fringe node 16 of supporting the NNI (not shown); protection from the predominating path to the back-up path switch may relate to enter professional VLAN (its can be with subdomain in the corresponding VLAN of predominating path) to the switching of the corresponding reserve VLAN of back-up path, when detecting that main SDMA shuts down and when needing switching to reserve SDMA.Similarly,, can carry out similar switching and return to its original value outside subdomain when subdomain strides across the fringe node 16 of supporting NNI and sends in business with value with VLAN.This allows the subdomain protection is transparent for the outer entity of subdomain.Switch in the business that enters at fringe node 16 on UNI 18 interfaces, main and back-up path in subdomain keeps identical, stride UNI 18 interfaces and encapsulate in the same manner because usually enter traffic frame at the main or back-up path of fringe node 16, and go out traffic frame stride UNI 18 interfaces at fringe node 16 from main or back-up path decapsulation in the same manner.

Provide protection to share the ability of the multiple service of the common points in the huge broadcast domain according to subdomain protection of the present invention.This subdomain protection is set to require to use the service of multiple spot layout that the protection solution is provided.Collective's state in the point-to-point path between the inherent node of subdomain is determined the state of described subdomain.According to the present invention, main and reserve subdomain is used for providing protection mechanism for the service in the subdomain.State-driven protection main and the reserve subdomain is switched to be used for service main and that the reserve subdomain is transmitted.As above going through, the invention provides subdomain protection group, main and reserve subdomain is associated with it and follows the tracks of.

Advantageously, each subdomain does not require privacy protection message resource, i.e. CCM.Subdomain maintenance association group comprises the RMEP resource of the state that is used for determining subdomain.RMEP can be associated with a plurality of SDMA, and MEP and RMEP resource can be expanded and the protection mechanism uncoupling of enforceable scheme with providing.

The present invention can realize with the combination of hardware, software or hardware and software.The enforcement of method and system of the present invention can stride across the distribution mode realization that several interconnection computing systems disperse with centralized system or with different elements in a computing system.The computing system of any kind of or be suitable for realizing the miscellaneous equipment of method described herein is suitable for finishing function described herein.

The typical combination of hardware and software can be to have one or more treatment elements and the special use or the general-purpose computing system of the computer program stored on storage medium, when computer program loads and execution, control computer system and/or the element in computer system, thus computer system realizes method described herein.The present invention also can be embedded in the computer program, and this product comprises all features of the enforcement that can realize method described herein, and this product can be realized these methods in being carried in computing system the time.Storage medium is meant any easy mistake or Nonvolatile memory device.

Computer program herein or application program are meant any expression of any language, code or the note of instruction set, and instruction has directly system or finish the information processing capability of concrete function after following one or two: the conversion of a) arriving another kind of language, code or note; B) with the reproduction of different materials form.In addition, unless more than carry out opposite mentioning, should be noted that all nots to scale (NTS) of institute's drawings attached.In particular, the present invention can and not break away from its spirit and base attribute with other particular form enforcement, and correspondingly, when indication scope of the present invention, should be necessary with reference to following claims, rather than above specification.

Claims (26)

1. method that is used for protecting available service on the broadcast domain, described method comprises:

Set up subdomain in broadcast domain, subdomain comprises a group node of the communication path that is used to be provided to described service;

Monitor main subdomain maintenance association and reserve subdomain maintenance association, main and reserve subdomain maintenance association is respectively the set of main and back-up path, and representative is as the connectedness between the node of the fringe node in the described subdomain;

Detect the fault in the described main subdomain maintenance association; And

Switch to described reserve subdomain maintenance association.

2. the described method of claim 1 is wherein set up described subdomain based on the physical relation between the described groups of nodes.

3. the described method of claim 1 is wherein set up described subdomain based on the logical relation between the described groups of nodes, thereby is self-contained in the described subdomain to the visit of service.

4. the described method of claim 1, also comprise the node in the path in fault appears at main subdomain maintenance association and link one of at least the time, the route of will dividing into groups is from switching to and the corresponding subdomain of described reserve subdomain maintenance association with the corresponding main subdomain of described main subdomain maintenance association.

5. the described method of claim 4 comprises that also the service that will be managed is associated with subdomain protection group, wherein uses described subdomain protection to organize and manages described switching.

6. the described method of claim 1, comprise also one or more remote node end points (" RMEP ") are associated with described main and reserve subdomain maintenance association that wherein the state of communicating by letter with RMEP is monitored to detect the fault in the described main subdomain maintenance association.

7. the described method of claim 6 wherein uses clean culture continuity verification message to monitor the state of communicating by letter with described one or more RMEP.

8. the described method of claim 6, wherein use the multicast and the clean culture continuity verification message of indication remote defect sign (" RDI "), monitor the state of communicating by letter with described one or more RMEP, the described unicast messages of indication RDI is sent to has the RMEP that detects communication failure.

9. the described method of claim 6, wherein use multicast continuity verification message to monitor the state of communicating by letter with described one or more RMEP, at least a portion indication remote defect sign (" RDI ") of described multicast continuity verification message, described multicast message are indicated RDI and are comprised the tabulation with the RMEP that detects communication failure.

10. the described method of claim 1 also comprises monitoring described territory maintenance association, wherein monitors described territory maintenance association and monitors that described main and reserve subdomain maintenance association undertaken by same group of MEP.

11. the described method of claim 1 also comprises monitoring described territory maintenance association, wherein monitors described territory maintenance association and monitors that described main and reserve subdomain maintenance association undertaken by first group and second group of MEP respectively.

12. the described method of claim 11, the supervision of wherein said territory maintenance association is carried out under first rate, and described supervision main and reserve subdomain maintenance association is carried out under second speed, and the second speed ratio first rate is fast.

13. the described method of claim 1, wherein the switching to reserve subdomain maintenance association comprises, switch to the active vlan path values and, between predominating path and back-up path, stride subdomain NNI interface service switchover by entering VLAN when subdomain sends, recovering the VLAN value.

14. a system that is used to provide available service on the broadcast domain, described system comprises:

A plurality of nodes, described a plurality of nodes are arranged to the subdomain of the communication path of the service that is provided to, and each node comprises:

Storage device, be arranged to storage and main subdomain maintenance association and the corresponding data of reserve subdomain maintenance association, described main and reserve subdomain maintenance association is respectively the set of main and back-up path, represents as the connectedness between the node of the fringe node in the subdomain; With

CPU, the operation of described CPU so that:

Detect the fault in the described main subdomain maintenance association; With

Switch to described reserve subdomain maintenance association.

15. the described system of claim 14, wherein said subdomain is based on the physical relation between the described groups of nodes.

16. the described system of claim 14, wherein said subdomain is based on the logical relation between the described groups of nodes, thereby is self-contained in the described subdomain to the visit of service.

17. the described system of claim 14, wherein the node in the path in fault appears at main subdomain maintenance association and link one of at least the time, CPU also will be divided into groups route from switching to and the corresponding subdomain of described reserve subdomain maintenance association with the corresponding main subdomain of described main subdomain maintenance association.

18. the described system of claim 17, the service and the subdomain maintenance association of being managed are associated with subdomain protection group, wherein manage described switching based on the state of described subdomain maintenance association.

19. the described system of claim 14, also comprise one or more remote node end points (" RMEP ") are associated with described main and reserve subdomain maintenance association, wherein the state of communicating by letter with RMEP by the CPU supervision to detect the fault in the described main subdomain maintenance association.

20. the described system of claim 19 wherein uses clean culture continuity verification message to monitor the state of communicating by letter with described one or more RMEP.

21. the described system of claim 19, wherein use the multicast of indication remote defect sign (" RDI ") to monitor the state of communicating by letter with described one or more RMEP with clean culture continuity verification message, the described unicast messages of indication RDI is sent to has the RMEP that detects communication failure.

22. the described system of claim 19, wherein use multicast continuity verification message to monitor the state of communicating by letter with described one or more RMEP, at least a portion indication remote defect sign (" RDI ") of described multicast continuity verification message, described multicast message are indicated RDI and are comprised the tabulation with the RMEP that detects communication failure.

23. the storage medium of a storage computation machine program is finished a kind of service method available on broadcast domain that is used to protect when computer program is performed, described method comprises:

Set up subdomain in broadcast domain, subdomain comprises a group node of the communication path that is used to the service that is provided to;

Monitor main subdomain maintenance association and reserve subdomain maintenance association, described main and reserve subdomain maintenance association is respectively the set of main and back-up path, represents as the connectedness between the node of the fringe node in the subdomain;

Detect the fault in the described main subdomain maintenance association; And

Switch to described reserve subdomain maintenance association.

24. the described method of claim 23, when also comprising the going up one of at least of node on the link in the path in fault appears at described main subdomain maintenance association and link, the route of will dividing into groups is from switching to and the corresponding subdomain of described reserve subdomain maintenance association with the corresponding main subdomain of described main subdomain maintenance association.

25. the described method of claim 24 comprises that also the service that will be managed is associated with subdomain protection group, wherein uses described subdomain protection to organize and manages described switching.

26. the described method of claim 23, comprise also one or more remote node end points (" RMEP ") are associated with described main and reserve subdomain maintenance association that wherein the state of communicating by letter with RMEP is monitored to detect the fault in the described main subdomain maintenance association.

Images