MPLS Congestion Management via Traffic Drop Count Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional MPLS and GMPLS networks do not consider traffic queue congestion as a traffic engineering link metric, leading to severe traffic loss due to unaddressed congestion in data plane traffic queues, especially during transient failures or high data volumes exceeding transport capacity.
Innovation Solution
Monitoring traffic drop counts at network elements to convert them into normalized congestion indicators, which are used to optimize label switched paths and trigger rerouting or protection switchover when congestion exceeds thresholds, employing protocols like RSVP-TE or GMPLS-RSVP-TE for route optimization and protection switching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional MPLS and GMPLS networks use traditional traffic engineering link metrics (link bandwidth, service expenses, delay/jitter) without considering traffic queue congestion, then the network operation is simple and conventional, but severe traffic loss occurs due to unaddressed congestion in data plane traffic queues
Solution Approach 1:
The patent applies preliminary action by monitoring traffic queue congestion status in advance and converting it to congestion indicators before severe congestion occurs. The system proactively identifies congested links and triggers rerouting operations before traffic loss becomes severe, rather than reacting after congestion has already caused problems.
Solution Approach 2:
The patent implements feedback by continuously monitoring traffic drop counts in data plane queues, converting these to congestion indicators, and using this information to dynamically adjust routing decisions. The congestion status feeds back into the traffic engineering metrics, creating a closed-loop system that adapts to real-time network conditions.
2Reliability
If the network monitors traffic drop counts and implements dynamic rerouting based on congestion indicators, then traffic loss is reduced and reliability improves, but the complexity of network operations and processing increases
Solution Approach 1:
The patent applies self-service by enabling the network to automatically monitor its own congestion status, convert drop counts to congestion indicators, and trigger rerouting operations without external intervention. The system autonomously manages congestion by using its own operational data (traffic drop counts) to make routing decisions, reducing the need for manual network operations.
Solution Approach 2:
The patent changes the parameter set used for routing decisions by introducing congestion indicators derived from traffic drop counts as new traffic engineering metrics. This parameter expansion allows the routing system to consider congestion status alongside traditional metrics like bandwidth and delay, enabling more informed routing decisions.
3Productivity
If congestion status is considered as a traffic engineering link metric for constrained path calculation, then LSP rerouting occurs on congested links and traffic loss is minimized, but the computational overhead and processing time increase
Solution Approach 1:
The patent applies preliminary action by continuously maintaining congestion indicators based on monitored traffic drop counts before congestion becomes severe. This pre-computed congestion information is readily available when path calculation is needed, eliminating the need for time-consuming congestion analysis during critical rerouting decisions.
Solution Approach 2:
The patent substitutes complex real-time congestion analysis with a simplified mechanism that uses pre-converted congestion indicators. Instead of performing heavy computational analysis of queue dynamics during path calculation, the system uses the already-processed congestion indicator values, reducing computational overhead and response time.
Data Source
AI summary
Methods and apparatus for mitigating congestion by using advanced traffic engineering services in a multiprotocol label switching (MPLS) network and/or a generalized multiprotocol label switching (GMPLS) network are provided. In an example, provided is a method for mitigating congestion in an MPLS network. The method includes monitoring, at a network element, a traffic drop count due to traffic queue overflow on at least one of a traffic queue coupled to a link and a network interface coupled to the link. Then the traffic drop count is converted to a normalized congestion indicator for the at least one of the traffic queue and the network interface. Further, based on the normalized congestion indicator, optimize a route of a label switched path in the MPLS network so the route avoids the link. The provided methods and apparatus advantageously improve throughput, reliability, and availability of traffic engineered MPLS and GMPLS transport services.