BGP Routing via AIGP Cost Control for Partial Link Failures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional BGP networks lack the capability to automatically redirect traffic in case of partial link bundle failures, leading to potential traffic losses, unlike IGP networks that can handle such failures through mechanisms like bandwidth-aware MPLS-TE.
Innovation Solution
Implement a mechanism that automatically detects capacity losses in BGP networks using real-time topology telemetry, adjusts Accumulated Interior Gateway Protocol (AIGP) costs to reroute traffic to alternate neighbors, and reverts the changes when the link capacity stabilizes, ensuring traffic is redirected without manual intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional BGP networks are used without automatic traffic redirection capability, then device complexity is reduced and ease of operation is improved, but traffic loss occurs during partial link bundle failures
Solution Approach 1:
The system enables BGP networks to automatically detect capacity losses and redirect traffic without manual intervention. The routing optimization mechanism self-manages the detection of link failures, calculation of alternate paths, and application of cost changes to reroute traffic, eliminating the need for manual CLI commands while preventing traffic loss during partial link bundle failures
Solution Approach 2:
The system implements continuous monitoring of link capacity and traffic flow through telemetry data collection. When capacity loss is detected, the system provides feedback by automatically adjusting AIGP costs to redirect traffic along alternate paths. This closed-loop feedback mechanism ensures reliable traffic redirection while maintaining manageable system complexity through automated decision-making
2Reliability
If manual cost changes are applied to BGP neighbors to redirect traffic, then traffic can be rerouted during link failures, but loss of time occurs due to manual intervention requirements
Solution Approach 1:
The system pre-configures alternate paths and monitors link capacity continuously before failures occur. When a partial link bundle failure is detected, the system immediately executes pre-planned traffic redirection by adjusting AIGP costs, eliminating the time delay associated with manual detection and configuration. The alternate routing paths are prepared in advance, enabling instant response to failures
Solution Approach 2:
The system automates the entire traffic redirection process by self-detecting capacity losses, self-calculating optimal alternate paths based on telemetry data, and self-applying cost changes to BGP neighbors. This eliminates manual intervention entirely, reducing the response time from minutes or hours of manual configuration to seconds of automated execution
3Reliability
If bandwidth-aware MPLS-TE is used in IGP networks for partial link failures, then traffic loss is avoided, but BGP networks lack this capability resulting in traffic losses
Solution Approach 1:
The system extends the bandwidth-aware traffic redirection capability from IGP/MPLS-TE networks to BGP networks. By implementing similar telemetry-based capacity monitoring and AIGP cost adjustment mechanisms in BGP, the system provides universal traffic loss avoidance across both IGP and BGP protocol domains, eliminating the capability gap while maintaining protocol-specific operational characteristics
Solution Approach 2:
The system changes the operational parameters of BGP by introducing dynamic AIGP cost adjustments based on real-time link capacity telemetry. This enables BGP to respond to partial link failures similarly to how IGP networks use MPLS-TE, achieving traffic loss avoidance through parameter-based routing optimization while preserving BGP's independent protocol functionality and compatibility
Data Source
AI summary
Aspects of the subject disclosure may include, for example, detecting whether a bandwidth capacity loss occurs on a first link bundle connecting one or more first core routers of a core Interior Gateway Protocol (IGP) network with one or more first provider routers, resulting in a detected capacity loss, wherein communication between the one or more first core routers of the core IGP network and the one or more first provider routers utilizes a Border Gateway Protocol (BGP), wherein a second link bundle connects one or more second core routers of the core IGP network with one or more second provider routers, and wherein communication between the one or more second core routers of the core IGP network and one or more second provider routers utilizes the BGP; responsive to the detected capacity loss, automatically increasing a cost associated with selected network traffic that was to be carried between the one or more first provider routers and the one or more first core routers of the core IGP network and that will instead be re-directed for communication via the second link bundle; and causing communication via the second link bundle of the selected network traffic, wherein the communication via the second link bundle of the selected network traffic is between the one or more second provider routers and the one or more second core routers of the core IGP network. Other embodiments are disclosed.