Patent Application
20100014417
This application is a continuation of International Application No. PCT/CN2008/073124, filed on Nov. 20, 2008, which claims priority to Chinese Patent Application No. 200710168174.9, filed with the Chinese Patent Office on Nov. 28, 2006 and entitled “Method and Apparatus of ring network protection”. The contents of these applications are hereby incorporated herein by reference in their entirety.
The present disclosure relates to optical network technologies, and in particular, to a method and an apparatus of ring network protection.
In the traditional Synchronous Digital Hierarchy (SDH) Synchronous Optical Network (SONET) or Optical Transport Network (OTN) field, ring networking is an important networking mode if the effectiveness (utilization ratio of channels), reliability and cost-efficiency of the network are taken into consideration. In order to improve the robustness of the ring network, all the nodes on the ring need to provide an automatic protection switching mechanism.
The prior art provides a ring network protection mechanism in an SDH system. As shown in
As shown in
As analyzed above, with regards to switching of service 1 between node 1 and node 3, if node 1 and node 3 perform switching directly, the trail subsequent to switching is the optimum trail. However, the nodes that actually perform the switching operation are node 2 and node 3. Consequently, the transmission trail of the service is longer than necessary. When the distance between nodes is very long, such a switching mode affects the plotection switching time and the signal quality, and is rather inappropriate.
Likewise, when the working channel and the protection channel between node 4 and node 5 fail, node 4 and node 5 also perform switching to switch service 2 to the protection channel. If the foregoing two failures coexist, according to the rules of switching of the nodes between both sides of a failed section in the prior art, node 2, node 3, node 4, and node 5 perform switching concurrently. Consequently, misconnection occurs because service 1 and service 2 contend for the same protection channel.
As shown in
As shown in
As shown in
However, the ring network protection mechanism suffers from at least the following defects: One section of a ring network can have only one protected service. When multiple services on the physical section need to be protected, multiple protection groups need to be configured in different logical ring networks. When one section fails, multiple protection groups of multiple services on the section perform switching consecutively, and the interruption of the service which performs switching later tends to be rather long.
A method of ring network protection is disclosed in an embodiment of the present disclosure to shorten switching time of a service.
The ring network protection method includes:
determining, by a node in a ring network, a section that suffers a fault and determining a timeslot of a service impacted by the fault on the node according to section ring formation information and service route information preconfigured on the node when trigger conditions of protection switching are fulfilled; and
switching the service onto a protection channel that bypasses the section that suffers the fault, where the section ring formation information lists the information about the sections included in a ring consecutively in a preset direction, the information about the sections includes preassigned IDs of sections between adjacent nodes in the ring network, and the service route information lists the information about the trails that use the node as a source node and a sink node of the service in each specific timeslot.
A network element device is disclosed in an embodiment of the present disclosure. The network element device includes:
a section ring formation information module, adapted to: configure section ring formation information, where: the section ring formation information lists the information about the sections included in a ring consecutively in a preset direction; and the information about the sections includes preassigned IDs of sections between adjacent nodes in the ring network;
a service route information module, adapted to configure service route information, where the service route information lists the information about the trails that use the node as a source node and a sink node of the service in each specific timeslot;
a switching control module, adapted to: determine the section that suffers the fault and determine the timeslot of the service impacted by the fault on the node according to the section ring formation information configured in the section ring formation information module and the service route information configured in the service route information module when the trigger conditions of protection switching are fulfilled, and output the section ID and the service timeslot information; and
a switching module, adapted to switch the service in the timeslot indicated by the timeslot information to the protection channel that bypasses the section corresponding to the section ID according to the section ID and the timeslot information output by the switching control module.
Consistent with some embodiments of the present disclosure, a section ID is set, the service impacted by the fault is determined according to the section ID and the service route information carried in the fault information, and only the service impacted by the fault is switched, thus preventing the switched service from passing through more trails than necessary. Moreover, multiple protected services can exist on one section. Fault detection is performed on the service layer, and the service is switched on the channel layer in terms of the source node and the sink node of the service, thus shortening the switching time of the service and improving the signal quality after switching.
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate some embodiments and together with the description, serve to explain the principles of the embodiments disclosed herein.
Reference will now be made in detail to embodiments disclosed in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts.
1. A logical protection group is configured on each node of the ring network. The physical channels (optical interface or ODUk timeslot) corresponding to the eastbound and westbound working channel and protection channel are specified. The working channel and the protection channel may be in the same fiber (to form a 2-fiber ring) or in different fibers (to form a 4-fiber ring).
2. The eastbound channel and the westbound channel between two adjacent nodes are interconnected to form a closed ring network. In order to fulfill the 50 ms protection switching time, no more than 16 nodes can exist on the ring. Therefore, the number of sections on the ring is not greater than 16. Each section on the ring is numbered (from 0 to 15), and no duplicate serial number of the section is allowed among different sections.
On the basis of the networking mode of ring network protection and the mode of configuring a single-node protection group shown in
Section ring formation information and service route information need to be configured on each node of the ring network in the embodiment of the present disclosure. The section ring formation information can be illustrated in the form of figures or tables. For example, the section ring formation information is rendered as a section ring chart. On each node, the section ring formation information may be set up from west to east or from east to west so long as the section ring formation information is set up in the same sequence for all nodes on the ring. In this embodiment, it is assumed that the section ring formation information is set up from west to east. The service route information may also be rendered in a tabular form. For example, the section ring formation information and the service route information of node A in
The section ring formation information of node A lists the sections included in a ring that uses node A as a source node and a sink node from west to east. As shown in
The ring chart and the service route information on each node may be configured by the user manually, or set up by software automatically by means of a search protocol so long as all nodes on the ring set up their own section ring formation information and service route information in the same way.
When a node on the ring detects failure of a working channel on a section, or when other switching conditions are fulfilled (for example, forced switching, manual switching, exercise switching), the node needs to notify the information such as section state request of this section, ID of this section, long trail ID, and short trail ID to other nodes on the ring. Such information constitutes a protection signaling. The section state request includes switching causes, for example, Signal Failure (SF), Signal Deterioration (SD), forced switching, manual switching, or exercise switching. Generally, protection signaling is carried through overhead (either channel-associated overhead or not). Table 2 defines the format of a protection signaling:
The length of a protection signaling is 2 bytes. The first byte is APS 1, which denotes the section state request and indicates the switching causes and the switching mode. The second byte is APS2. The first 4 bits of APS2 denote a section ID, and the next bit denotes a flag of long trail and short trail, and the last 3 bits denote a bridge switching flag. The bridge switching flag is optional, and the remainder is mandatory. In the APS bytes in the prior art, one byte is required for carrying the source node ID and the sink node ID, another byte is required for the section state request, and another byte is required for the bridge switching flag, amounting to 4 bytes. By comparison, the APS bytes under the present disclosure require only 2 bytes, thus saving the signaling bandwidth effectively. The functions of the section state request of the APS1 are described in Table 3:
The contents of Table 3 and Table 4 are for the exemplary purpose only, and do not constitute limitations on the present disclosure. Moreover, any content item in APS1 or APS2 is omissible and serves as a default value.
The conditions of triggering protection switching of each node on the ring include three aspects:
1. This node detects the channel failure state (SF/SD).
2. This node receives an external switching command (forced switching, manual switching, and so on).
3. This node receives an APS request from other nodes.
The protection switching process in an embodiment of the present disclosure is described below through example 1. As shown in
Step 601: After detecting signal failure of westbound section 2, node B sends a failure message to other nodes on the ring through an APS request. That is, node B sends a short-trail APS request westward: SF|2|S|IDLE, and sends a long-trail APS request eastward: SF|2|L|IDLE. In this way, all the intermediate nodes on the ring receive the APS request indicating that section 2 on the ring fails.
Step 602: After receiving the APS request indicating that section 2 on the ring fails, node A determines that the service of its westbound timeslot 2 is interrupted according to the section ring formation information and the service route information recorded by node A. Likewise, node D determines that the service of its eastbound timeslot 2 is interrupted according to the section ring formation information and the service route information recorded by node D.
Step 603: Node A and node D bridge and switch the service of timeslot 2 to the reverse protection channel that bypasses section 2. Other nodes have no service that passes through section 2, namely, the fault of section 2 exerts no impact on the service of such nodes. Therefore, such nodes let the APS request and the switching service pass. Finally, the service between node A and node D is recovered, and the source-and-sink switching function for such a service is implemented.
In the switching process illustrated in example 1, the service is switched in terms of the source node and the sink node of the service, thus shortening the switching time of the service and improving the signal quality after switching.
The following example 2 illustrates how to prevent misconnection of services in the case that more than one instance of protection switching occurs in the protection switching process in an embodiment of the present disclosure. As shown in
Step 801: When section 2 and section 6 fail concurrently, node C detects the eastbound failure, and therefore, sends a short-trail APS request eastward (namely, in the direction where the failure occurs): SF|2|S|IDLE, and sends a long-trail APS request westward (namely, in the reverse direction): SF|2|L|IDLE. Likewise, node F detects the westbound failure, and therefore, sends a short-trail APS request westward: SF|6|S|IDLE, and sends a long-trail APS request eastward: SF|6|L|IDLE.
Step 802: Node D and node E on the ring receive the foregoing two long-trail APS requests in the east and west directions, indicating that section 2 and section 6 on the ring have failed. In the case that more than one section fails, the ring network is unable to connect into an entirety and is divided into multiple subrings: If two sections fail, the ring network is divided into two parts, namely, two subrings; if three sections fail, the ring network is divided into three parts, namely, three subrings, and so on. If the source node and the sink of the service belong to different subrings, it is impossible to set up a connection between the source node and the sink no matter how the service is switched. In this embodiment, node D and node E determine that section 2 and section 6 have failed according to the two APS requests received, and therefore, determine that the ring network is divided into two subrings.
Step 803: Node D also determines that the source node and the sink node of the service in its eastbound timeslot 2 belong to two different subrings as a result of division, and therefore, does not switch the service in eastbound timeslot 2, thus avoiding misconnection of the service. Likewise, node E knows that the source node and the sink node of the service in its westbound timeslot 2 belong to two different subrings as a result of division. In this case, the service cannot be recovered no matter how the service is switched. Therefore, node E does not switch the service in westbound timeslot 2.
a section ring formation information module 910, adapted to configure section ring formation information, where the section ring formation information lists the rings that use the section ring formation information module as a source node and a sink node and lists the sections included by the rings consecutively in a specific direction, and the specific direction is a direction from west to east or a direction from east to west and keeps consistent for all the nodes on the ring;
a service route information module 920, adapted to configure service route information, where the service route information lists the layout of the source node and the sink node of the westbound service and the eastbound service in each timeslot;
a switching control module 930, adapted to: determine the section that suffers the fault and determine the timeslot of the service impacted by the fault on the node according to the section ring formation information configured in the section ring formation information module 910 and the service route information configured in the service route information module 920 when the trigger conditions of protection switching are fulfilled, and output the section ID and the timeslot ID; and
a switching module 940, adapted to switch the service in the timeslot represented by the timeslot ID to the protection channel that bypasses the section corresponding to the section ID according to the section ID and the timeslot ID output by the switching control module.
Each section of a ring network may include multiple protected services. When a section fails, all the services on the section are protected concurrently only if the whole ring network performs switching in a single attempt. Therefore, the switching time of all services is short enough.
The switching control module 930 includes:
a detecting unit 931, adapted to detect whether signal failure or signal deterioration occurs on a section in the ring network, wherein: if such is the case, the trigger conditions of protection switching are fulfilled; and the detecting unit 931 generally detects only two sections adjacent to the switching control module;
a fault determining unit 932, adapted to: determine the ID of the section that suffers a fault and determine the ID of the timeslot of the service impacted by the fault on the switching control module according to the section ring formation information configured in the section ring formation information module 910 and the service route information configured in the service route information module 920 when the trigger conditions of protection switching are fulfilled; and
an outputting unit 934, adapted to output the section ID and the timeslot ID determined by the fault determining unit 932 to the switching module 940.
The switching control module 930 further includes an APS transceiver module 950. When the detecting module 931 detects signal failure or signal deterioration on a section in the ring network, the detecting unit notifies the section ID of the section that suffers from the signal failure or signal deterioration to the APS request transceiver module 950.
Therefore, the APS request transceiver module 950 generates an APS request inclusive of the section ID according to the section ID received from the detecting unit 931, and sends the APS request to other nodes in the ring network.
The APS request transceiver unit 950 is further adapted to receive the APS requests from other nodes in the ring network. When the APS request transceiver module receives an APS request, the trigger conditions of protection switching are fulfilled.
The switching control module 930 further includes a misconnection preventing unit 933, which is adapted to: judge whether the source node and the sink node of the service can be connected through the node in the ring network after the fault determining unit 932 determines the section that suffers the fault and the service impacted by the fault on the switching control module; and, if the source node and the sink node of the service can be connected, trigger the outputting unit 934 to output the section ID of the section and the timeslot ID of the service.
Embodiments consistent with the present disclosure may provide the following benefits:
The service is switched in terms of the source node and the sink node of the service, thus shortening the switching time of the service and improving the signal quality after switching;
through the source-and-sink switching for the service, the service misconnection is prevented;
each section of a ring network may include multiple protected services. When a section fails, all the services on the section are protected concurrently only if the whole ring network performs switching in a single attempt. Therefore, the switching time of all services is short enough;
the protection group is separated from the protected service. In this way, the protection group can be configured more flexibly, and the intelligent features of the optical network such as dynamic configuration of services are supported; and
as against the 4-byte APS signaling in the prior art 1, the embodiments of the present disclosure require only 2-byte APS signaling, thus saving signaling bandwidth.
Although the disclosure has been described through some exemplary embodiments, the disclosure is not limited to such embodiments. It is apparent that those skilled in the art can make various modifications and variations to the disclosed embodiments without departing from the spirit and scope of the disclosure. The present disclosure is intended to cover the modifications and variations provided that they fall in the scope of protection defined by the following claims or their equivalents.
| Number | Date | Country | Kind |
|---|---|---|---|
| 200710168174.9 | Nov 2007 | CN | national |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/CN2008/073124 | Nov 2008 | US |
| Child | 12547702 | US |
