ECMP Failover via BFD and Pre-configured Backup Routes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data communication networks face inefficiencies in traffic distribution due to endpoint failures, which lead to radical rebalancing of data flows, decreasing metrics that define efficiency and effectiveness within the autonomous system.
Innovation Solution
Implementing a system that uses Border Gateway Protocol (BGP) for deterministic backend failover in equal cost multi-path (ECMP) routing, ensuring consistent load balancing by persistently directing data flows through the same backend node and employing bidirectional forwarding detection (BFD) for fast failover without propagating failure metrics upstream.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If ECMP routing is used to load balance traffic over multiple paths, then bandwidth utilization is improved, but system stability deteriorates when endpoint failures occur due to radical rebalancing of data flows
Solution Approach 1:
The system pre-establishes backup routes alongside primary ECMP paths before failures occur. When an endpoint fails, the backup routes are immediately activated without requiring radical rebalancing, thus maintaining routing stability while preserving bandwidth utilization benefits of ECMP.
Solution Approach 2:
The system implements monitoring of endpoint status and dynamically adjusts routing decisions based on real-time feedback. When endpoint failures are detected, the system responds by activating pre-configured backup routes, preventing radical rebalancing and maintaining stable traffic distribution across the network.
2Reliability
If deterministic backend failover is implemented to maintain routing stability, then system reliability is improved, but device complexity increases due to additional failover mechanisms
Solution Approach 1:
Backup routes are pre-configured and staged before failures occur, allowing immediate failover without complex real-time decision-making algorithms. This approach enhances reliability while minimizing the complexity of failover mechanisms by relying on pre-computed routing paths.
Solution Approach 2:
The system introduces a routing controller that acts as an intermediary between the ECMP routing system and backup routes. This controller manages failover logic centrally, simplifying the overall system architecture while maintaining high reliability through coordinated failover actions across multiple network nodes.
3Loss of time
If fast failover is implemented using BFD to detect endpoint failures, then response time is improved, but loss of information increases due to potential metric propagation issues
Solution Approach 1:
The system extracts and isolates failure detection functionality into a separate BFD mechanism that operates independently from the main routing protocol. This allows fast failure detection without propagating potentially harmful failure metrics upstream, as the BFD system can trigger failover actions locally without affecting global routing tables.
Solution Approach 2:
The routing controller serves as an intermediary that receives failure notifications from BFD and translates them into appropriate routing actions. This mediation prevents direct propagation of failure metrics upstream while maintaining fast response times, as the controller can immediately activate backup routes without waiting for upstream routing protocol convergence.
Data Source
AI summary
A method may include, with a controller of an AS, routing a data flow from a source device, through at least one front-end node to a plurality of back-end nodes, and balancing, by the controller, the data flow to the back-end nodes equally based at least in part on ECMP routing. A number of routes from the back-end nodes to endpoint devices may be determined based at least in part on a preference for a primary route from the back-end nodes to a corresponding one of the endpoint devices, and backup routes from the back-end nodes to the corresponding one of the endpoint devices. An indication of a failure of a first endpoint device is received, and the back-end nodes utilize a first backup route that is associated with a second endpoint device to rebalance the data flow from the first endpoint device to the second endpoint device.