Cross-layer OSPF-TE Backup Path Reconfiguration for Multiple-link Failures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current reconfiguration techniques in backbone networks are inadequate for handling multiple-link failures, particularly in large carrier-class networks, as they lack a simple distributed protocol for cross-layer reconfiguration of impacted backup paths using Open Shortest Path First (OSPF) - Traffic Engineering (TE) and Resource Reservation Protocol (RSVP).
Innovation Solution
A method is introduced where a link state advertisement (LSA) is used to identify impacted links, and a new shortest path is computed by ignoring the failed link and common links to the backup path, using multiprotocol label switching (MPLS) fast reroute, with the option to set up the new path using RSVP messages without source routing, ensuring lower congestion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional reconfiguration techniques are used for link failures, then single-link failure recovery is achieved, but multiple-link failure survivability is insufficient
Solution Approach 1:
The patent segments the reconfiguration process into two distinct layers: (1) link-level reconfiguration using MPLS Fast Reroute for immediate local backup path activation, and (2) cross-layer reconfiguration using OSPF-TE and RSVP for coordinated backup path adjustment across multiple nodes. This segmentation allows each layer to handle specific aspects of failure recovery independently, improving overall reliability without requiring a monolithic complex protocol.
Solution Approach 2:
The patent introduces Link State Advertisements (LSAs) as intermediaries that carry capacity information about backup paths between nodes. When a link fails, the head-end node generates an LSA containing backup path capacity data, which is distributed to other head-end nodes. These LSAs enable coordinated reconfiguration decisions without requiring direct complex communication between all nodes, thus improving survivability while managing protocol complexity.
2Reliability
If backup paths are pre-computed for all possible link failures, then failure recovery capability is improved, but network resource overhead increases
Solution Approach 1:
The patent implements preliminary action by pre-computing and establishing backup paths for each link using MPLS Fast Reroute before failures occur. These backup paths are ready for immediate activation upon failure detection. Additionally, the system continuously maintains capacity information about these pre-established backup paths through OSPF-TE, enabling rapid cross-layer reconfiguration without requiring extensive real-time computation during actual failures.
Solution Approach 2:
The patent dynamically changes the cost parameters in OSPF-TE based on the capacity availability of backup paths. When a link fails and its backup path is activated, the system updates the cost parameters to reflect the reduced capacity on shared links. This parameter change mechanism allows the routing protocol to adapt to changing network conditions and make optimal reconfiguration decisions without requiring complete re-computation of all paths.
3Reliability
If cross-layer reconfiguration is implemented using OSPF-TE and RSVP, then multiple-link failure survivability is improved, but protocol implementation complexity increases
Solution Approach 1:
The patent leverages the multi-functionality of existing standard protocols to achieve cross-layer reconfiguration. OSPF-TE is used for both traditional routing and for distributing backup path capacity information through extended LSAs. RSVP is used for both resource reservation and for setting up reconfigured backup paths. This universal use of existing protocols reduces implementation complexity compared to designing and deploying entirely new specialized protocols for each function.
Solution Approach 2:
The patent implements self-service mechanisms where each head-end node independently determines whether it needs to reconfigure its backup paths by evaluating received LSAs containing backup path capacity information. Each node autonomously computes new backup paths using OSPF-TE cost updates and sets them up using RSVP without requiring centralized control or complex coordination protocols. This distributed self-service approach improves survivability while managing implementation complexity.
Data Source
AI summary
A second link in a network is reconfigured after failure of a first link, so that multiple-link failures may be survived. After receipt of an OSPF link state advertisement (LSA) indicating use of a backup path bl after the failure of link l, new backup paths are determined for each impacted link (x,y) that has a backup path sharing one or more links with the backup path bl. The new backup paths are computed while ignoring the failed link l, the impacted link (x,y), and links common to the backup paths b(x,y) and bl. The LSA protocol is modified to accommodate information used in determining whether a link is an impacted link.