US 8,619,553 B2Granted

Methods and systems for mesh restoration based on associated hop designated transit lists

AssigneeCiena Corporation

InventorsVagish Madrahalli, Alessandro Fabbri, John Oltman

Life of this patent

Estimating the term…

Loading patent…

Abstract

The present disclosure provides systems and methods for mesh restoration based on Associated Hop Designated Transit Lists (DTLs). The Associated Hop DTLs are calculated through a global restoration calculation (GRC) algorithm that can be run by a central controller in an offline manner. The restoration paths calculated by the GRC algorithm can then be pushed down to the originating nodes for each connection to utilize at the time of a given failure scenario. This GRC algorithm can be performed for each possible bundle failure in the network, where a bundle failure is determined by the set of all links which may fail together due to common shared risk, such as a common conduit or DWDM fiber. The globally calculated, contention free, restoration paths are then pushed down to each node in the network.

Claims12

Dependent claims

A mesh restoration method comprising:

selecting a target link from a plurality of links in an optical network, the optical network utilizing a signaling and routing protocol;
modeling failures on the target link and all partner links of the plurality of links with a common protection bundle identification, wherein the failures affect a plurality of sub-network connections traversing the target link and the partner links, each of the sub-network connections comprises call-initiated allocations of dedicated network resources in the optical network;
determining pre-computed restoration paths based on the failures for each of the affected plurality of sub-network connections;
providing the pre-computed restoration paths to an originating node in form of Associated Hop Designated Transit Lists for each of the pre-computed restoration paths;
routing, using the signaling and routing protocol, data of the affected plurality of sub-network connections along the pre-computed restoration paths on a per sub-network connection basis in accordance with the Associated Hop Designated Transit Lists;
determining a plurality of sectors each comprising a portion of a total number of the plurality of links;
repeating said selecting, said modeling, said determining, and said providing for each of the plurality of links by using a parallel computation based on the plurality of sectors;
and
assigning, to a plurality of machines, the parallel computation on the portion of the total number of the plurality of links in the plurality of sectors;
the plurality of machines having identical copies of a network model of the optical network;
the parallel computation comprising:
sharing, among the plurality of the machines, the Associated Hop Designated Transit Lists for links of the plurality of sectors having a common protection bundle;
communicating, among the plurality of machines, an indication indicating a generation of the Associated Hop Designated Transit Lists and a required arbitration;
and sorting a set of links handled, by each of the plurality of machines for parallel computation, in a manner to minimize computation collisions.

11 more elements

An optical network, comprising:

one or more originating nodes;
one or more terminating nodes interconnected to the one or more originating nodes through a plurality of optical links, the optical network utilizing a signaling and routing protocol;
and
a controller in communication with each of the one or more originating nodes and the one or more terminating nodes;
wherein the controller is configured to:
select a target link from the plurality of links;
model failures on the target link and all partner links of the plurality of links with a common protection bundle identification, wherein the failures affect a plurality of sub-network connections traversing the target link and the partner links, each of the sub-network connections comprises call-initiated allocations of dedicated network resources in the optical network;
determine pre-computed restoration paths based on the failures for each of the affected plurality of sub-network connections;
provide the pre-computed restoration paths to the one or more originating nodes in form of Associated Hop Designated Transit Lists for each of the pre-computed restoration paths;
route, using the signaling and routing protocol, data of the affected plurality of sub-network connections along the pre-computed restoration paths on a per-sub-network connection basis in accordance with the Associated Hop Designated Transit Lists;
determine a plurality of sectors each comprising a portion of a total number of the plurality of links;
repeat said selecting, said modeling, said determining, and said providing for each of the plurality of links by using a parallel computation based on the plurality of sectors;
and
assign, to a plurality of machines, the parallel computation on the portion of the total number of the plurality of links in the plurality of sectors;
the plurality of machines having identical copies of a network model of the optical network;
the parallel computation comprising:
sharing, among the plurality of the machines, the Associated Hop Designated Transit Lists for links of the plurality of sectors having a common protection bundle;
communicating, among the plurality of machines, an indication indicating a generation of the Associated Hop Designated Transit Lists and a required arbitration;
and sorting a set of links handled, by each of the plurality of machines for parallel computation, in a manner to minimize computation collisions.

16 more elements

A management system for a mesh network, comprising:

a data store;
a memory;
a processor;
a network interface communicatively coupled to one or more nodes in the mesh network;
and
a local interface communicatively coupling each of the data store, the memory, the processor, and the network interface;
wherein the processor is configured to:
select a target link from a plurality of links in an optical network, the optical network utilizing a signaling and routing protocol;
model failures on the target link and all partner links of the plurality of links with a common protection bundle, wherein the failures affect a plurality of sub-network connections traversing the target link and at least one link in the common protection bundle, each of the sub-network connections comprises call-initiated allocations of dedicated network resources in the optical network;
determine pre-computed restoration paths based on the failures for each of the affected plurality of sub-network connections;
provide the pre-computed restoration paths to the one or more nodes in form of Associated Hop Designated Transit Lists for each of the pre-computed restoration paths;
route, using the signaling and routing protocol, data of the affected plurality of sub-network connections along the pre-computed restoration paths on a per-sub-network connection basis in accordance with the Associated Hop Designated Transit Lists;
determine a plurality of sectors each comprising a portion of a total number of the plurality of links;
repeat the selecting, the modeling, the determining, and the providing for each of the plurality of links by using a parallel computation based on the plurality of sectors;
and
assign, to a plurality of machines, the parallel computation on the portion of the total number of the plurality of links in the plurality of sectors;
the plurality of machines having identical copies of a network model of the optical network;
the parallel computation comprising:
sharing, among the plurality of the machines, the Associated Hop Designated Transit Lists for links of the plurality of sectors having a common protection bundle;
communicating, among the plurality of machines, an indication indicating a generation of the Associated Hop Designated Transit Lists and a required arbitration;
and sorting a set of links handled, by each of the plurality of machines for parallel computation, in a manner to minimize computation collisions.

18 more elements

Dependencies are inferred from the claim text.