JP5485998B2 - Method and apparatus for protection - Google Patents

Description

  The present invention relates generally to the technical field of network communications, and in particular to protection for provider bridge traffic engineering.

  Provider Backbone Bridge Traffic Engineering (PBB-TE) uses end-to-end or end-to-end 1: 1 protection schemes, referred to in this source as “e2e protection”. This scheme provides a robust protection measure that can be used in any mesh type network connection. However, the physical connections assumed for many networks often use fiber rings, in which case protection of the ring topology is important. In the case of a cascade ring network as shown in FIG. 1, the communication links of logically adjacent rings are interconnected using a matched pair of nodes. The primary path between endpoints passes through the link between those belonging to a pair of nodes (ie, through the primary path node). A protection path passes through different links between related ones in a pair of nodes (ie, through a protection node). Each path in TBB-TE is associated with a VLAN-ID (VID). When a failure occurs in the link or node of the primary path, the traffic is switched to the protection path by the head end point changing the VID used for the data frame.

  The end-to-end 1: 1 protection scheme works, but has some drawbacks. For example, if different rings are associated with different geographic regions and are managed and operated by different organizations within the carrier, each of those organizations will not be able to work with other organizations, You may want to schedule (but it is difficult). Furthermore, in the case of long-distance communication, the frequency disturbance in the extended e2e path is so large that the probability that the second fault will occur in the protection path before the first fault is restored becomes unacceptably large. It is to end. Extended e2e protection schemes can be used to alleviate this problem, but the number of protection paths required can be quite large depending on the protection scope and the number of rings, e.g. multiple rings This will protect against multiple problems. For example, when separating ring pairs in a cascade system, four end-to-end paths are required. Because there may be four paths. In the case of three rings in cascade, six paths are required to protect against two independent failures in any ring pair, with complete protection from the simultaneous occurrence of a single failure on each ring. Will need 8 passes. It is difficult to maintain such a large number of paths. This is because the transfer state is directly associated with a large number of installed paths, and a CFM (Continuity Falt Management) session is added to a node that terminates any path.

  The object of the present invention is to address at least one of the above problems.

The method according to one embodiment is:
A method for protection in a provider bridge backbone network,
Depending on the failure, if possible, using section protection, where the session is defined between the entry and exit points and does not extend to the entire provider bridge backbone network. When,
Using the end-to-end protection if the failure could not be recovered by section protection.

Diagram showing end-to-end 1: 1 protection scheme. The figure for demonstrating the system of the section protection applied only to an end-to-end protection and a primary path. The figure which shows a mode that section protection is used about both a primary path | pass and a protection path | pass. The figure which shows an example of this invention used as a foundation explaining the case where it responds to the state of various malfunctions shown in FIG. 5 thru | or FIG. The figure which shows the malfunction of the upstream in a primary route. The figure which shows the malfunction of the downstream in a primary route. FIG. 7 is a diagram showing how the correspondence shown in FIGS. 5 and 6 potentially competes under the conditions of a single forwarding database for each bridge. The figure which shows the solution means of the problem shown in FIG. 7 based on VID swap. The figure which shows the sympathetic switch solution (sympathetic switch solution) with respect to the problem shown in FIG. The figure which shows the sympathetic switch solution (sympathetic switch solution) with respect to the problem shown in FIG.

  The method according to an embodiment of the present invention is a method for protection in a provider bridge backbone network, using section protection if possible depending on faults, faults, faults, abnormalities or faults. The section is defined between the entry point and the exit point and does not extend to the entire provider bridge backbone network and the failure could not be recovered by section protection In some cases, using end-to-end protection.

  An apparatus according to another embodiment of the present invention is an apparatus that performs communication between endpoints using a provider bridge backbone network, and includes an entry point node and an exit point node, and the entry point node and the exit point node. Both provide protection for the section of the network between the entry point node and the exit point node, the section does not extend to the entire provider bridge backbone network, and the provider bridge backbone network Accordingly, an apparatus that uses the protection of the section if possible and uses end-to-end protection between the endpoints if the failure could not be recovered by section protection.

  The advantages associated with the present invention include the ability to more efficiently utilize network resources and enhance protection functions. Each of the many paths required for comprehensive protection requires the forwarding state installed in the node to determine the path that the traffic uses, but is used by PBB-TE 1: Only data traffic propagates through one end-to-end path at any time in accordance with 1 “head-end switching”. This can overcome the problem of inefficient utilization of network resources in the 1: 1 form of traditional section protection, and the traditional method is an uneconomical form of section protection combined with end-to-end path protection. It was something like that. In addition, using section protection for the primary and protection end-to-end paths can simply double the amount of state information stored in a pair of nodes, but withstand one failure per ring. The failure includes the failure of one of the pair of nodes (in this case, the failure is caused to the two rings). However, it should be understood that the present invention is not limited to a ring architecture.

  These and other advantages of the present invention will become more apparent from the following detailed description and drawings.

  Data communication networks include a variety of computers, servers, nodes, routers, switches, bridges, hubs, proxies, and other network devices that are coupled to each other and configured to transmit data. These devices are referred to herein as “nodes”. For example, data such as Internet protocol packets, Ethernet frames, data cells, sections, other logically related data, etc. can be communicated between nodes by using one or more communication links between the nodes. Communication is performed by exchanging protocol data units via the data communication network. A particular protocol data unit is processed by multiple nodes and passes through multiple communication links as it propagates between the source and destination over the network.

  What has been described in the present application provides illustrative explanations of exemplary embodiments of the invention and is believed to be a useful and understandable description in terms of the principles and concepts of the invention. There is no intention to present a more detailed structural form of the invention than is necessary for a basic understanding of the invention.

  Figure 2 shows a protection scheme for both end-to-end protection and section protection from the perspective of two endpoint nodes 200, 202 connected via three cascade rings, where the three cascade rings are matched node pairs. They are interconnected by (matched pairs of nodes) (204, 206), (208, 210). The primary path 212 associated with the links 214, 216, 218 is protected by a protection path associated with the links 222, 224, 226 on an end-to-end basis or end-to-end basis. One or more sections of the primary path are also protected by local-basis, where a “section” is a link, ring, trunk, or part of another network. For example, link 214 is protected by path 228 by link 222, link 216 is protected by path 230 by link 224, and link 218 is protected by path 232 by link 226. The failure status of each path through the section is monitored independently by the CFM session between the section endpoints. In response to detecting a fault condition, section protection is used if possible to address the fault, and end-to-end protection is used if not. For example, the timeout period before a failure is signaled by an end-to-end CFM session is set longer than the timeout period of the corresponding section so that unnecessary end-to-end protection switches are not triggered. Note that activating section protection does not indicate switching from the primary e2e path to the protection e2e path. Rather, the primary e2e path is rerouted or rerouted only for the affected section. Thus, at least some of the problems with end-to-end 1: 1 protection schemes can be solved. Although Figure 2 specifically shows a ring, the section protection scheme described in this application has one constraint that the two paths that form a protected section must not cross between the beginning and end of protection. And may be used in any topology or form.

  Each of the many paths required for comprehensive protection requires a forwarding state installed in the node to determine which path to use for traffic, but is used by PBB-TE 1: It should be understood that the “head-end switching” model means that traffic flows on one end-to-end path at any given time. This can overcome the problem of inefficient utilization of network resources in the 1: 1 form of traditional section protection, and the traditional method is not suitable when deploying section protection combined with end-to-end path protection. It was economical.

  The protection for a given section is closed at the corresponding node or one matched node at the entry and exit points of the section. For example, the primary link 214 and the corresponding protection path 228 by the link 222 are closed at the node 204, and the node 204 is paired with the node 206. Thus, the failure in link 214 can be overcome by using section protection link 228 without switching from node 204 or requiring active involvement by node 206 in the switching process. In the rare case of a failure of node 204, as a result of a failure of the primary end-to-end CFM session for path 212, end-to-end protection is triggered and traffic is switched to the link and node for protection path 220, which Using node 206 instead of node 204. This advantageously reduces the problem with protection synchronization between matched nodes.

  Referring to FIG. 3, in one embodiment, section protection is provided in both the primary path and the protection path. More specifically, in addition to the paths and links described with respect to FIG. 2, link 222 is protected by path 300 by link 214, link 224 is protected by path 302 by link 216, and link 226 is path 304 by link 218. It is protected by The protection of each section starts and ends independently in both the primary path and the protection path. A CFM Connectivity Check session is performed on the primary path and protection path between endpoints 240, 242 to continuously determine the status or state of the two end-to-end paths. Two CFM sessions 306 (one clockwise and the other counterclockwise) determine the status of the two paths between nodes 200 and 204 using the VID used by the primary e2e path, Another CFM session 307 determines the status of the two paths between nodes 200 and 206 using the VID being used by the protection e2e path. Two CFM sessions 308 use the VID used by the primary e2e path to determine the status of the two paths between nodes 204 and 208, and two other CFM sessions 309 are used by the protection e2e path The status of the two paths between the nodes 206 and 210 is determined using the VID that has been set. Individual section protection CFM sessions do not need to be created for each end-to-end path. For example, two sessions 308 may be used for all routes in and out of the middle ring (middle ring) via nodes 204 and 208. Using section protection on the primary and protection paths simply doubles the amount of state information stored in matched nodes, while allowing one fault per ring to survive, Includes the case of one failure of a matched node (in which case it will cause failure on the two rings).

  FIG. 4 shows an example of the present invention used as a basis for explaining the case of responding to various fault states shown in FIGS. Without loss of generality, only the case where traffic flows from the left to the right is shown for all cases, and the terms “upstream” and “downstream” are used with respect to the flow direction. As shown in FIG. 4, backbone edge bridges (BEE) 400, 402 communicate over the ring. These rings are relayed in a matched pair 404a, 404b, 406a, 406b of a backbone core bridge (BCB). Each ECB includes four provider network ports (Provider Network Ports: PNP) such as 408a, 408b, 408c, 408d. Each of the BEBs includes a customer backbone port (CBP) 410 and two PNPs 412a, 412b. The primary path 420 between BEBs is protected by an e2e protection path 424 and a section protection path 422 between BCBs 404a and 406b. For simplicity of illustration, section protection of the e2e protection path is not shown. To detect faults and trigger protection switching, CFM sessions are maintained between the following pairs: (BEB400, BCB404b); (BEB400, BCB404a); (BEB404b, BCB406b); (BEB404a, BCB406a); BEB406b, BCB402); and (BEB406a, BCB402).

  FIG. 5 shows a problem in the primary path 420 that is upstream of the BCB 404a. Depending on the failure detected by the CFM session anomaly, the BEB 400 reroutes traffic along the section protection path for 420 along the first section, from PNP 412a to BCB 404b and back to BCB 404a (newly Route is set). BCB 404a returns the traffic rerouted at PNP 408d to primary path 420.

  FIG. 6 shows a failure of the primary route downstream of BCB 404a. In response to this fault condition, traffic is re-routed with section protection for the second section. In particular, BCB 404a reroutes traffic received at PNP 408c, and traffic proceeds to BCB 404b and BCB 406 via PNP 408c. BCB 406b forwards traffic to BCB 406a, which returns the rerouted traffic back to the primary path to PNP 408d. The rerouted frames are not changed in any way at the protection start point BCB 404a and they can use the same VID value (related to the primary e2e path).

  FIG. 7 shows why the countermeasures shown in FIGS. 5 and 6 require that the BCB matched node 406b has two forwarding paths for section protection of the primary e2e path, where the primary e2e path is It is used in various situations. Since two paths cannot be active at the same time, BCB404b does not provide a direct means of determining which path is used when, and maintaining this without active involvement This is advantageous from the viewpoint of state and complexity. Installing both of these transfer paths simultaneously indicates a potential conflict. In particular, different routes in BCB 404b are associated with the same address (which is of endpoint BEB 402). This is a problem because BCB typically has only one forwarding database (FDB). One direct solution to this problem is to have a different forwarding database in each PNP. However, this is contrary to the Ethernet bridge standard specifications IEEE802.1Q and IEEE802.1ah. In the standard specifications, FDB can be applied to all bridge ports and conform to these standard specifications. If this is desired, it becomes a problem.

  FIG. 8 shows another solution to the problem shown in FIG. The solution shown is based on VID swap processing. In this solution, VID swap is performed by the BCB switch from the upstream node to the buddy node or from the buddy node to the downstream node in the section protection path. In the specifically illustrated example, a VID swap is performed at the PNP 408a of the BCB 404b and again at the PNP 408d so that the traffic leaving the PNP 408d has the same VID as the traffic entering the PNP 408a, but within the BCB 404b There is no competition. This is because different VIDs are associated with these paths. Therefore, one FDB per VLAN can be maintained according to Ethernet standard specifications.

  FIGS. 9 and 10 show another protection switching solution for downstream and upstream failures, respectively, and an extra state is installed in the matched node depending on the type of CFM session, but it is still 404a and 404b. No clear synchronization is required between matched node pairs such as In this solution, an additional pair of CCM sessions (clockwise and counterclockwise sessions) match a match node that starts session protection in the primary path e2e path and a session protection in the protection e2e path. Maintained between the nodes. A “crossed or diagonal” session between these matched nodes is used by upstream matched nodes in the section protection path, and the downstream portion of the section protection path is used for the section protection end or end point. Be able to determine directly whether or not it is possible. For example, a diagonal CCM session 900 is maintained between BCB 406a and BCB 404b. The diagonal session allows the network to directly determine whether the BCB 404b can reach the section end point or end point in the PNP 408d of the BCB 406a. If a failure 902 occurs in the primary path, the BCB 404b forwards traffic to the BCB 406a along the protection path in a clockwise direction. This is because the buddy matched node 404a can directly detect the failure from the failure of the primary CCM session with the BCB 406a and switch the traffic in the section protection path. The upstream defect 1000 is not detected by the CCM session 900, so traffic will be routed by the BCB 404b according to the default (ie setting when the CFM session 900 is operating properly) as if the primary path of the upstream ring Traffic is forwarded as if a failure occurred, ie, failure 1000 is assumed. The operation of BCB404b does not depend on the form of failure 1001 in the link between matched nodes that destroys both paths, because both rings have failed and the subsequent in the primary e2e path through both sections. Note that the bug can only be recovered with the e2e protection switch.

  Although the present invention has been described with reference to the above embodiments, those skilled in the art will appreciate that variations and modifications of the described embodiments can be obtained without departing from the described inventive concept. Will. Moreover, while the preferred embodiment has been described in connection with various illustrated structures, those skilled in the art will recognize that the system may be implemented using various specific structures. The present invention is not to be defined except by the purpose and spirit of the appended claims.

  This application benefits from the priority of US Provisional Application No. 61 / 096,011 entitled “SECTION PROTECTION FOR PROVIDER BACKBONE BRIDGE TRAFFIC ENGINEERING” filed on September 11, 2008.

Claims (18)

A method for protection in a provider bridge backbone network including a first endpoint node , wherein the first endpoint node communicates with a second endpoint node via a plurality of sections, the method comprising: ,
If defects, a step of using the protection section,該Se click Deployment is defined between the entry point and exit point and from the first endpoint node to the second endpoint nodes A step that does not extend to the entire provider bridge backbone network; and
Using the entire end-to-end protection from the first endpoint node to the second endpoint node if the failure could not be recovered by section protection. Protection of d Ndotuendo comprises primary path and protection path comprises a protection for at least one section of the flop Rotekushon said primary path section, starting the method, the protection section independently for each section of the primary path and a step of ending, the process of claim 1. Using the management session connection failure, all in and out of the section of the provider bridge backbone network via first and second node sections traffic on all paths through the sections that can take advantage the method of claim 1, further comprising the step of determining a defect state. Protection section, and Oite terminated to one match Tonneau de at the end of each section, the method comprising the without synchronizing the single matched nodes to the corresponding matched nodes, utilizing protection section The method of claim 1, further comprising a step. The method comprising the one matched node in the primary path of the end-to-end, by using a management session defect connection with another matched node in end-to-end protection path, and protection paths section, the method of the corresponding further comprising the step of determining whether or not the connection is between the end point of the protection of the primary path downstream sections, according to claim 1, wherein. By maintaining multiple forwarding database for at least one node, at least one node, without the need to synchronize with other nodes, including the step of to participate in section of the at least two protection claim The method according to 1. The method comprising the first connection defect management session, failure to determine whether or not recovered by protection of end-to-end, by a second connection defect management session, the malfunction can be recovered by protection section comprising the step of determining whether Taka not, the first connection failure timeout period management sessions is longer than the timeout period of the second connection defect management session, the method of claim 1, wherein. The method comprising at the start of each of the protection section, without changing the VLAN identifier, based on the failure of the state and end-to-end VLAN identifier of each pass, the frame to a given destination, the sections The method of claim 1, comprising flowing to the primary path through the section or to a protection path through the section. The method comprising the nodes involved in the protection of both sections of the upstream and downstream, based on the state of the section of the end-to-end VLAN identifier and the downstream traffic, the end of the protection section of the upstream 6. A method according to claim 5, comprising the step of providing to a path leading to a point or to a path leading to an end point of protection of the downstream section. An apparatus for communicating between endpoints using a provider bridge backbone network including a first endpoint node , wherein the first endpoint node communicates with a second endpoint node via a plurality of sections The device is
An entry point node and an exit point node , the entry point node and the exit point node operating together to perform protection functions for a section of the network between the entry point node and the exit point node;
Said section, said not extend to the whole of the first endpoint node from said second endpoint node to the provider bridges backbone network, the provider bridges backbone network can, if failure, the protection of the previous SL section Use the end-to-end protection across the provider bridge backbone network from the first endpoint node to the second endpoint node if the failure could not be recovered by section protection. use, equipment. Wherein the end-to-end protection is the primary path and the protection path, sections flop Rotekushon includes protection for at least one section of the primary path, protection section for each section is independently opened Hajime及 beauty End It is that device of claim 10,. Using the management session connection failure, all in and out of the section of the provider bridge backbone network via first and second node sections traffic on all paths through the sections that can take advantage determining the defect state, the apparatus of claim 10. Protection section, and Oite terminated to one match Tonneau de at the end of each section, the protection section, are utilized without synchronizing the single matched nodes to the corresponding matched nodes, claim 10. The apparatus according to 10. And one matched node in the end-to-end of the primary path, using the management session defect connection with another matched node in end-to-end protection path, and protection paths section, the corresponding section primary It determines whether or not the connection is between the end point of the protection paths downstream apparatus of claim 10. By maintaining multiple forwarding database for at least one node, at least one node, without the need to synchronize with other nodes, you to participate in at least two protection sections, according to claim 10, wherein apparatus. Using a management session of the first connection failure, the failure to determine whether or not recovered by protection of end-to-end, using the management session of the second connection failure, the failure sections Protection The apparatus according to claim 10 , wherein it is determined whether or not recovery is possible, and a timeout period of the management session for the first connection failure is longer than a timeout period of the management session for the second connection failure. At the start of each of the protection section, without changing the VLAN identifier, based on the failure of the state and end-to-end VLAN identifier of each path, a frame to a given destination, the primary path through the section 11. The apparatus of claim 10, wherein the apparatus is provided in a protection path through or through the section. In node responsible for protection of both the sections of the upstream and downstream, based on the state of the section of the end-to-end VLAN identifier and the downstream traffic, path to the end point of the protection section of the upstream , or given to a path leading to the end point of the protection section of the downstream apparatus of claim 14, wherein.

Images