Route Forwarding Table for Fast Service Convergence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies for fast service convergence in networks, such as MPLS, TE FRR, and IGP, are inadequate for achieving end-to-end service convergence within 50 ms when a link failure occurs between Autonomous System Border Routers, resulting in slow convergence times that fail to meet the requirements for voice and video services.
Innovation Solution
A method and network device that utilize a route forwarding table with pre-set preferred and secondly-preferred routes, where the status of the preferred route is set to unavailable upon detection of a failure, allowing service packets to be rapidly switched to the secondly-preferred route, ensuring convergence within 50 ms, independent of the number of service routes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If routing protocol convergence is used for fast service convergence upon link failure between ASs, then service convergence can be achieved, but the convergence time is too slow (1s-10s) to meet voice and video service requirements (<50ms)
Solution Approach 1:
The patent pre-calculates and stores multiple forwarding paths (preferred path and alternative paths) in the forwarding table before failure occurs. When a link failure is detected, the system immediately switches to a pre-computed alternative path without performing route re-computation, thereby achieving fast service convergence within 50ms for voice and video services.
Solution Approach 2:
The patent segments the routing function into path selection (performed in advance) and path execution (performed on-demand). The control plane pre-computes multiple forwarding paths and stores them in the forwarding table, while the data plane simply selects from pre-computed paths during failure, separating the complex computation from the time-critical switching operation.
2Reliability
If routing protocol convergence is used, then service convergence is achieved, but convergence time increases significantly as the number of service routes increases
Solution Approach 1:
The patent pre-calculates and stores multiple forwarding paths (preferred path and alternative paths) in the forwarding table before failure occurs. When a link failure is detected, the system immediately switches to a pre-computed alternative path without performing route re-computation, thereby achieving fast service convergence within 50ms for voice and video services.
Solution Approach 2:
The patent creates and stores multiple copies of forwarding paths (preferred and alternative) in the forwarding table for each service route. These pre-computed path copies are readily available for immediate switching without requiring real-time computation, thus maintaining fast convergence even as the number of service routes increases.
Data Source
AI summary
A method and network device for fast service convergence, used for communications between Autonomous Systems, includes: presetting a route forwarding table, each forwarding table item in the forwarding table includes information of a preferred route and a secondly-preferred route; setting the status of the preferred route as unavailable in the forwarding table item after a failure is detected in the preferred route; after receiving a service packet, querying in the route forwarding table the forwarding table item corresponding to the service packet, and determining whether the status of the preferred route is available, and if yes, the service packet is forwarded through the preferred route, otherwise through the secondly-preferred route. Because the failure detection time is not longer than 30 ms, the switch time of the service packet from the preferred route to the secondly-preferred route may be within 50 ms, thereby enabling fast service convergence.