Node Realignment in Label Switched Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for realigning nodes in label switched networks, especially during multiple restarts in rapidly changing networks, face challenges with incomplete state information and outdated backups, leading to failed route resurrection.
Innovation Solution
The method employs new Notify messages (Notify_Add, Notify_Upgrade, and Notify_Delete) and the concept of LSP Degree of Reliability to synchronize and validate path information across nodes, ensuring accurate and reliable state recovery by adjusting the reliability threshold based on network hops and configurable values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If periodic backups of node state are used for realignment, then node state recovery is possible after restart, but the backup becomes outdated and incomplete in networks with rapid route changes
Solution Approach 1:
The patent applies preliminary action by having nodes proactively store backup path status records in non-volatile memory before restarts occur. Each record captures the complete set of path status information at a specific point in time, ensuring that when a restart happens, the node can immediately restore its state without waiting for post-restart reconstruction. This preliminary backup mechanism ensures reliability of state recovery while minimizing information loss by capturing the most recent state before any changes occur.
Solution Approach 2:
The patent implements feedback through a realignment validation process where, after restoring from backup, the node exchanges path status information with neighboring nodes to verify the accuracy and completeness of the restored state. If discrepancies are detected, the node can identify which paths need to be re-established or corrected. This feedback loop ensures that even if the backup is slightly outdated, the node can quickly reconcile differences and achieve complete realignment.
2Loss of information
If graceful restart with neighboring node communication is used, then real-time state information can be obtained, but the method fails when multiple nodes restart simultaneously
Solution Approach 1:
The patent stores backup path status records in non-volatile memory at each node before restarts occur. This preliminary action ensures that even when multiple nodes restart simultaneously, each node has its own independent copy of the path status information that does not depend on neighboring nodes being operational. The backup records serve as a reliable foundation for state recovery regardless of the restart status of other nodes in the network.
Solution Approach 2:
The patent uses the backup path status record stored in non-volatile memory as an intermediary that mediates the state recovery process. Instead of directly relying on communication with neighboring nodes (which may fail during simultaneous restarts), the node first restores its state from the backup record and then uses this restored state as a basis for subsequent communication and validation with neighbors. This intermediary backup mechanism breaks the circular dependency that causes graceful restart to fail during simultaneous multi-node restarts.
3Reliability
If manual reconciliation of circuit status is performed after restart, then network status can be synchronized, but significant manual effort is required
Solution Approach 1:
The patent enables self-service realignment by automatically restoring path status information from backup records stored in non-volatile memory. When a node restarts, it autonomously retrieves its backup record, restores the path status information, and validates the restored state against current network conditions. This automated self-service mechanism eliminates the need for manual reconciliation operations, significantly reducing operational complexity while maintaining reliable network status synchronization.
Solution Approach 2:
The patent implements automated feedback mechanisms where the node, after restoring from backup, automatically exchanges path status information with neighboring nodes to verify synchronization. If any discrepancies are detected during this automated validation process, the system can identify and correct the specific paths that need attention. This automated feedback loop replaces manual reconciliation with an intelligent self-correcting process, reducing operational effort while ensuring network status synchronization.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method of realigning a node in a label switched network comprising a plurality of nodes and a node with processor, memory and a network interface, for carrying out the method. The method includes periodically maintaining backup path status information for the node restarting the node and re-establishing label switched paths with the other nodes using the backup status information. Communication with adjacent nodes is carried out in order to reconcile the path status information with respective path status information in the adjacent nodes in order to establish node realignment preferably judged against a threshold value for path reliability.