Multicast Vector Stack Routing Across Autonomous Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Multicast Virtual Private Networks (MVPNs) face challenges in spanning multiple Autonomous Systems (AS) without requiring service providers to make their Provider Edge (PE) routers globally addressable via unicast, due to security concerns and the need to redistribute PE information across competitor domains.
Innovation Solution
The implementation of a Multicast Vector Stack (MVS) mechanism, where a vector stack is added to multicast messages to guide intermediate routers across AS boundaries, allowing the construction of multicast trees without relying on unicast routing, using BGP updates and Route Distinguishers to determine the path through ASBRs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If PE routers are made globally addressable via unicast to enable multicast across multiple ASs, then multicast delivery capability is improved, but security is worsened and information leakage to competitor domains occurs
Solution Approach 1:
The patent introduces ASBRs (Autonomous System Boundary Routers) as intermediary devices that mediate between internal PE routers and external multicast traffic. The ASBRs receive multicast packets from external sources, perform RPF checks using BGP-learned routes, and forward packets to appropriate PE routers without requiring PE routers to be globally addressable. This mediator approach enables multicast delivery while maintaining security boundaries between autonomous systems.
Solution Approach 2:
The patent segments the multicast routing function into two distinct parts: (1) BGP routing for control plane information exchange between ASBRs, and (2) Multicast forwarding data plane operations within each AS. This segmentation allows PE routers to remain invisible to external systems while still participating in multicast delivery through their connected ASBRs, thus maintaining security while enabling multicast capability.
2Ease of operation
If PE information is redistributed across competitor domains to enable multicast routing, then routing capability is improved, but information security is worsened
Solution Approach 1:
ASBRs serve as intermediaries that exchange only necessary routing information with external systems through BGP, while keeping detailed PE router information within the autonomous system. The ASBRs perform route lookups and forwarding decisions without exposing internal network topology or PE router identities to competitor domains, thus enabling routing capability while preventing information leakage.
Solution Approach 2:
The patent implements local quality by allowing each autonomous system to maintain its own routing information and security policies locally. BGP updates are exchanged between ASBRs at the boundary, but detailed routing information remains localized within each AS. This enables routing capability through controlled boundary interactions while maintaining information security through localized information management.
3Measurement precision
If RPF checks are performed using BGP-learned routes through ASBRs, then routing accuracy is improved, but processing complexity is worsened
Solution Approach 1:
The patent implements preliminary action by pre-establishing BGP routing relationships and learning routes between ASBRs before multicast traffic flows. ASBRs perform RPF checks using these pre-learned BGP routes, which eliminates the need for complex real-time route calculations during multicast forwarding. This preliminary routing setup improves routing accuracy while reducing processing complexity during actual multicast operations.
Data Source
AI summary
A method and apparatus for providing multicast messages across a data communication network, the method comprising receiving a multicast message and adding to the multicast message a vector stack including at least one address of a router to which the multicast message is to be sent. The multicast message and the vector stack are then forwarded. At the first router indicated by the vector stack, the next address to which the multicast message is to be sent is read. This is repeated as necessary until the multicast message is received by the final address in the vector stack. The multicast message is then routed to the address indicated in the original multicast message.