Distributed Multicast Logical Router for Network Traffic Load Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current logical routers processing multicast data messages create unnecessary traffic load by replicating packets between edge nodes and logical networks, necessitating a distributed processing solution for multicast data messages originating and terminating within a logical network.
Innovation Solution
A managed network with centralized and distributed components implements a distributed multicast logical router, where managed forwarding elements (MFEs) send multicast group queries to data compute nodes, receive reports, and generate summarized reports to populate output lists for replicating multicast data messages, utilizing a central controller to aggregate and distribute configuration information across MFEs and logical switches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If logical routers process and replicate multicast data messages at edge nodes, then multicast messages can be delivered to logical networks, but unnecessary traffic load is created on the system
Solution Approach 1:
The patent segments the multicast routing function by introducing a distributed multicast router component at each edge node that works independently with the central multicast router. This segmentation allows local multicast messages to be handled locally without being replicated unnecessarily across the entire system, thus reducing traffic load while maintaining delivery capability.
Solution Approach 2:
The patent implements local quality by enabling edge nodes to perform local multicast routing decisions for messages originating and terminating within the logical network. Each edge node maintains local multicast state information, allowing it to handle local multicast traffic independently without involving other edge nodes, thereby reducing unnecessary system-wide traffic load.
2Loss of energy
If a distributed multicast logical router is implemented, then traffic load is reduced, but system complexity increases due to multiple components coordinating
Solution Approach 1:
The patent merges the central multicast router and distributed multicast routers into a unified distributed multicast routing system. The central controller maintains global multicast state while distributed routers at edge nodes maintain local state, and they work together through standardized protocols. This merging approach manages complexity by creating a coordinated system where each component has a defined role rather than creating entirely separate systems.
Solution Approach 2:
The patent introduces an intermediary mechanism through the standardized communication protocols between the central multicast router and distributed multicast routers. This intermediary layer manages the coordination and information exchange, abstracting the complexity of inter-component communication and allowing each component to operate independently while maintaining system-wide consistency.
3Productivity
If multicast messages are replicated at edge nodes, then all DCNs receive messages, but DCNs not in the multicast group still receive traffic
Solution Approach 1:
The patent applies preliminary action by having edge nodes pre-establish multicast group membership information before multicast messages arrive. The distributed multicast router at each edge node maintains local multicast state information about which DCNs belong to which multicast groups, allowing it to make immediate local routing decisions without replicating messages to DCNs that are not members of the multicast group.
Solution Approach 2:
The patent implements feedback mechanisms where distributed multicast routers at edge nodes report local multicast group membership information to the central multicast router. This feedback loop allows the system to maintain accurate global and local multicast state information, enabling precise replication only to DCNs that are actual members of multicast groups, thus eliminating unnecessary traffic to non-participating DCNs.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
For a managed network implementing at least one logical router having centralized and distributed components, some embodiments provide a method for processing multicast data messages at a first managed forwarding element (MFE) executing on a first host machine that implements a distributed multicast logical router and multiple logical switches logically connected to the logical router in conjunction with a set of additional MFEs executing on additional host machines. The method replicates multicast data messages received from a source data compute node (DCN), operating on the first host machine, that logically connects to a first logical switch of the multiple logical switches. The method replicates the multicast data message to a set of DCNs in the multicast group in the logical network without routing through a centralized local multicast router.