Multicast Replication Offloading to Physical Switches
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing multicast routing architecture in software-defined networks (SDNs) faces scalability issues due to centralized T0 edge nodes handling multicast replication for multiple T1 edge nodes, leading to degraded throughput and increased latency, making it difficult to scale for additional tenant networks.
Innovation Solution
Offloading multicast replication from tiered edge routers to physical switches using underlay multicast group IPs and leveraging a SDN controller for synchronized VTEP lists, which reduces the number of replications at the source edge and improves routing throughput by allowing more parallel flows and enabling more tenants to participate in multicast routing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If centralized T0 edge nodes handle multicast replication for multiple T1 edge nodes, then multicast routing can be managed centrally, but scalability is limited and throughput degrades with increased latency
Solution Approach 1:
The patent segments the centralized multicast replication function into distributed components at each T1 edge node. Instead of one central T0 node handling all replications, each T1 edge node performs local replications for its own tenants, distributing the replication load across multiple nodes and eliminating the bottleneck at the central T0 node.
Solution Approach 2:
The patent introduces a new dimensional approach by implementing multicast replication at the T1 edge node level (a different hierarchical dimension) rather than only at the central T0 level. This multi-level replication approach allows simultaneous central coordination and distributed execution, improving scalability without sacrificing throughput.
2Adaptability or versatility
If centralized T0 edge nodes handle multicast replication for multiple T1 edge nodes, then centralized control is maintained, but latency increases with more tenants
Solution Approach 1:
The patent segments the multicast replication process into local operations at each T1 edge node, reducing the distance and hops packets must traverse. By performing replications locally at the T1 node rather than routing through the central T0 node, packet latency is reduced while maintaining scalability.
Solution Approach 2:
The patent implements preliminary action by pre-establishing multicast group memberships and replication rules at each T1 edge node. This allows packets to be replicated immediately upon arrival at the T1 node without requiring real-time coordination with the central T0 node, reducing latency while maintaining centralized control over the overall multicast architecture.
3Device complexity
If centralized T0 edge nodes handle multicast replication, then routing can be simplified, but the load on T0 edges increases making it difficult to scale
Solution Approach 1:
The patent segments the replication load from the central T0 edge nodes and assigns it to individual T1 edge nodes. This segmentation reduces the processing burden on T0 nodes while distributing the workload across the network, enabling the system to scale to more tenants without overloading the central nodes.
Solution Approach 2:
The patent implements self-service by enabling each T1 edge node to autonomously perform multicast replication for its own tenants. Each T1 node independently handles replication operations without requiring centralized processing, allowing the network to scale organically as new T1 nodes and tenants are added.
Data Source
AI summary
A method for offloading multicast replication from multiple tiers of edge nodes implemented by multiple host machines to a physical switch is provided. Each of the multiple host machines implements a provider edge node and a tenant edge node. One host machine among the multiple host machines receives a packet having an overlay multicast group identifier. The host machine maps the overlay multicast group identifier to an underlay multicast group identifier. The host machine encapsulates the packet with an encapsulation header that includes the underlay multicast group identifier to create an encapsulated packet. The host machine forwards the encapsulated packet to a physical switch of the network segment. The physical switch forwards copies of the encapsulated packet to tenant edge nodes at one or more ports that are determined to be interested in the underlay multicast group identifier.