Patent Application
20080031619
The present invention relates to technologies for recovering from a link failure, and in particular to a method and device for recovering from an inter-domain link failure.
Traditionally, traffic scheduling in a conventional optical network is implemented through a static configuration by a network administrator, but isn't supported by dynamic provision. Typically, the traditional optical network is of linear or ring networking, and its protective recovery is implemented by Multiplex Section Protection and Sub-Network Connection Protection (SNCP), which is substantially a static recovery method. However, as data services and dedicated services develop rapidly, requirements for network bandwidth increases gradually, and requirements for dynamic allocation of network bandwidth becomes increasingly pressing. The network has been expected to provide the dynamic provision and to support a meshed network architecture, and also expected to be capable of flexible extension and fast protective recovery.
The Automatically Switched Optical Network (ASON) has effectively solved the above problem. The ASON uses the Generalized Multi-Protocol Label Switching (GMPLS) protocol within its control plane, which has become a core technology during the development of optical networks. Two connection types are newly provided for the ASON, one of which is a soft permanent connection type, and the other of which is a switched connection type. At present, the International Telecommunications Union—Telecommunication Standardization Sector (ITU-T) has substantially accomplished the definition for ASON architecture and requirements. The Internet Engineering Task Force (IETF) has accomplished the protocol extension and definition for signaling, automatic discovery and routing within a single domain.
As the ASON develops, it is required for the control plane to manage a large-scale network. At present, both the ITU-T and the Optical Internetworking Forum (OIF) adopt a hierarchical network model, where a control domain at a lower layer is represented as a proxy node at an upper layer. The proxy node can distribute an abstract topology, an inter-domain link, a reachable addresses, etc. indicative of a domain, and thus, a hierarchical network can be formed upward in a layer-by-layer way. As illustrated in
Within such a multi-domain network, a rate of recovering from a connection failure becomes a critical bottle neck, since in the case of multiple domains, a connection usually spans multiple domains, and passes far more nodes than in the case of a single domain. Therefore, the rate has become an urgent challenge to be addressed for the extended network. Generally, a current cross-domain connection is recovered locally by means of an internal tunnel within a domain in the case of a link failure within a single domain and recovered by means of an end-to-end connection recovery in the case of an inter-domain link failure. As shown in
The present invention provides a method and a device for recovering from a link failure. Therefore, the time required for a recovery from an inter-domain link failure can be reduced.
A method for recovering from a link failure according to the present invention may be applicable to recovery from an inter-domain link failure in a hierarchical automatic-switching optical network. The method includes:
initiating a local recovery by an upstream node of a failed inter-domain link;
performing the local recovery with the upstream node of the failed inter-domain link as a source node and a downstream node of the failed inter-domain link as a destination node.
The local recovery is performed if the upstream node detects the inter-domain link failure or if the downstream node detects and notifies the upstream node.
The local recovery is performed in unit of connection, or the local recovery is performed on all traffics routed through the failed inter-domain link in unit of link.
The performing the local recovery includes:
inquiring to obtain a recovery path to the downstream node by a connection controller at the upstream node;
initiating establishment of the recovery path from the upstream node to the downstream node by a protocol controller;
upon the establishment of the recovery path, bridging traffic on the failed inter-domain link onto the recovery path.
Optionally, the inquiring for a recovery path to the downstream node includes:
inquiring a routing controller for the recovery path, with the upstream node of the failed inter-domain link as a source node and the downstream node of the failed inter-domain link as a destination node for connection recovery, and calculating and returning the recovery path to the connection controller by the routing controller.
Optionally, the inquiring for a recovery path to the downstream node includes:
inquiring, by the connection controller, to obtain the recovery path which configured in advance and is stored locally, and using spare resource at each node on the recovery path.
Optionally, the inquiring for a recovery path to the downstream node includes: inquiring, by the connection controller, to obtain the recovery path which is configured in advance and stored locally, and using reserved resource at each node on the recovery path.
Optionally, the obtaining of the recovery path by the connection controller includes: determining a switching through a preset cycle composed of inter-domain links and a node.
Optionally, the recovery path is a path composed of failure-independent inter-domain links or a path composed of links passing through one or more other domains.
Optionally, the method may further include:
in the case of a return-type recovery from a failure, returning to the original path upon elimination of the inter-domain link failure.
Optionally, the method may further include:
initiating automatically an end-to-end recovery in the case that the local recovery from the inter-domain link failure fails.
A device for recovering from an inter-domain link failure according to the present invention includes:
a link resource manager, adapted to detect an inter-domain link failure;
a connection controller, adapted to inquire for a recovery path to a downstream node, and notify a protocol controller to establish the recovery path;
the protocol controller, adapted to establish the recovery path.
Optionally, the device further includes a routing controller, for calculating the recovery path dynamically with an upstream node of the failed inter-domain link as a source node and a downstream node of the failed inter-domain link as a destination node in response to the inquiry from the connection controller.
Optionally, the connection controller inquiries to obtain the recovery path which is configured in advance and stored locally with a spare or reserved resource.
In the present invention, the upstream node of the failed inter-link domain link can initiate the local recovery from the link failure, and the local recovery from the link failure can be performed with the upstream node as the source node and the downstream node as the destination node. Since it is only required to perform the local recovery from the link failure between the upstream node and the downstream node of the failed link, a recovery path can be calculated dynamically or configured in advance, and resource can be reserved for sharing of a recovery bandwidth. It is possible to improve greatly the rate of recovering from the inter-domain link failure and to save the bandwidth.
According to the present invention, a fast local recovery is performed in the case of an inter-domain link failure. A recovery path can be calculated dynamically or be preset. Reservation of resources can be performed for sharing recovery bandwidth. Besides, the recovery path may be another inter-domain link or a link passing through another domain. Therefore, it is possible to improve greatly a rate of recovering from an inter-domain link failure and to save the bandwidth. Moreover, a coordination mechanism is provided for the fast local recovery and an end-to-end recovery, which will be described in detail hereinafter.
In step 11, in the case of an inter-domain link failure, it is determined, at a source end of a link, which recovery method to be used, if the recovery method is the fast local recovery, going to step 12, otherwise going to step 13.
In step 12, the fast local recovery method is used to determine an upstream node and a downstream node of the failed inter-domain link, and to initiate a recovery from the link failure with the upstream node and the downstream node as a source node and a destination node. If the fast local recovery method fails to recover from the failure, the process goes to step 13, where the recovery from the failure is performed by means of the end-to-end recovery method.
In step 13, the end-to-end recovery method is performed, that is, the head node of an LSP is notified to perform an end-to-end recovery.
Hereinafter, the method of a fast local recovery from a link failure according to the present invention will be described by way of an embodiment thereof.
According to the embodiment of the present invention, the upstream node (referred to as a failure recovery point) of the link can initiate the fast recovery from the inter-domain link failure. Particularly, upon detection of a failure, the upstream node of the link initiates a recovery from the inter-domain link failure, or upon detection of a failure, the downstream node of the link notifies the upstream node of the link to initiate a recovery from the inter-domain failure.
It should be noted that the user can initiate the inter-domain fast recovery in unit of connection or link. In the former case, if there is an inter-domain link failure, the fast local recovery is performed for a connection for which the fast recovery has been initiated, and the end-to-end recovery is performed for other connections, and in the latter case, the fast local recovery can be performed for all traffics routed through this inter-domain link.
Various solutions, such as a recovery method of dynamic calculation or a method of presetting a recovery path, can be supported at the source end of the inter-domain link. For the recovery method of dynamic calculation, a route is calculated with two ends of this inter-domain link as a source node and a destination node for connection recovery, so as to obtain a recovery path. Then, establishment of the recovery path and traffic switching can be initiated through a signaling in accordance with the recovery path.
For the presetting of a recovery path for recovering from a failure, two approaches can be used, one of which is to preset a recovery path but reserve no resource, i.e. two ends of this inter-domain link are taken as a source node and a destination node for connection recovery. Thus, establishment of the recovery path and traffic switching can be initiated through a signaling in accordance with the preset recovery path, and a spare resource can be used on the recovery path.
The other approach is to preset a recovery path and reserve a resource thereof, i.e. two ends of this inter-domain link are taken as a source node and a destination node for connection recovery. Thus, establishment of the recovery path and traffic switching can be initiated through a signaling in accordance with the preset recovery path, and the reserved resource can be used on the recovery path. The reserved resource can be shared by a failure-independent connection.
It should be noted that the recovery path can be obtained in other approaches. For example, a p-cycle (preset cycle) may be composed of inter-domain links and a node, which is well known in the prior art and thus will not be described here.
Further, the local recovery according to the present invention can be of a return or non-return type. In the case of the return type, the connection will return to the original path (via the link) upon elimination of the inter-domain link failure.
The recovery path for fast recovering from an inter-domain failure may be a link between two domains, which does not belong to an unassociated failure SRLG (Shared Risk Link Group). Alternatively, the recovery can be accomplished through another domain, and in the case that the fast inter-domain recovery fails, the end-to-end recovery can be initiated to maximize traffic survivability.
Now the description is given with specific examples. Referring to the network as shown in
Hereinafter, interactions between elements in an ASON according to the present invention will be described in detail with reference to the example in
Referring to
1) The recovery path is obtained by dynamic calculation: the CC inquiries a Routing Controller (RC) for the recovery path, with two ends of the inter-domain link as a source node and a destination node for connection recovery, and the RC calculates the recovery path dynamically.
2) If the recovery path is preset at the CC by the user without a resource reserved on the recovery path, the CC can inquire directly to obtain the locally stored recovery path and use a spare resource.
3) If the recovery path is preset at the CC by the user with a resource reserved on the recovery path, the CC can inquire directly to obtain the locally stored recovery path and use the reserved resource.
Upon obtaining the recovery path, the CC notifies the PC (through the RSVP-TE protocol or otherwise) to establish the recovery path, and bridges the traffic onto the recovery path. The recovery process is thus completed substantially.
Moreover, in the case of a recovery from a cross-domain failure, referring to the network as shown in
An embodiment of a device for recovering from an inter-domain link failure according to the present invention includes a link resource manager (LRM), a connection controller (CC), a protocol controller (PC), and a routing controller (RC).
Particularly, upon detection of a failure, the LRM at a node notifies the CC which in turn initiates an inter-domain fast recovery, and in this case, it is unnecessary for the PC to notify the head node of the traffic. In the embodiment, the CC can function as the following.
1) The recovery path is obtained by dynamic calculation: the CC inquiries the RC for the recovery path, with two ends of the inter-domain link as a source node and a destination node for connection recovery, and the RC calculates the recovery path dynamically.
2) If the recovery path is preset at the CC by the user without a resource reserved on the recovery path, the CC can inquire directly to obtain the locally stored recovery path and use a spare resource.
3) If the recovery path is preset at the CC by the user with a resource reserved on the recovery path, the CC can inquire directly to obtain the locally stored recovery path and use the reserved resource.
Upon obtaining the recovery path, the CC notifies the PC to establish the recovery path and bridges the traffic onto the recovery path. The recovery process is thus completed substantially.
While the present invention has been described and illustrated with reference to the embodiments thereof and the appended drawings, it shall be recognized by those skilled in the art that those embodiments and drawings are merely illustrative and not restrictive, that the present invention shall be not limited thereto, and that various modifications and variations can be made thereto in light of the descriptions and the drawings without departing from the sprit and scope of the present invention as defined by the accompanying claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| CN200510035896.8 | Jul 2005 | CN | national |
The present application is a continuation application of PCT application No. PCT/CN2006/001680, filed Jul. 14, 2006, which claims a priority of Chinese application No. 200510035896.8.
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/CN06/01680 | Jul 2006 | US |
| Child | 11881804 | Jul 2007 | US |
