SMLT Hold-Down Timer for IP Multicast Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current SMLT techniques fail to achieve sub-second failover efficiency for IP multicast traffic, leading to significant traffic loss and prolonged downtime due to the time required for multicast routes to converge after a failover event.
Innovation Solution
Implementing an IP multicast hold-down timer mechanism where a network device starts a timer upon returning to an active state, sends messages to its peer indicating SMLT VLAN IDs, and performs Layer 2 forwarding of multicast traffic during the timer period, allowing the peer to handle traffic until the timer expires and multicast routing is enabled, thereby minimizing traffic loss and enabling sub-second failover.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multicast routing is enabled immediately after failover, then routing convergence can proceed, but significant traffic loss occurs during the convergence period
Solution Approach 1:
The patent implements a hold-down timer that is started immediately upon failover detection, before multicast routing can converge. This preliminary action prevents traffic from being forwarded to the failed node during the convergence period, avoiding packet loss while routes are being re-established
Solution Approach 2:
The patent introduces an intermediary mechanism (the hold-down timer) that mediates between the failover event and multicast routing convergence. The timer acts as a buffer, controlling when traffic is allowed to flow again and ensuring that routing tables are fully converged before resuming normal traffic forwarding
2Reliability
If a hold-down timer is implemented to prevent traffic loss, then traffic forwarding reliability improves, but the failover time increases
Solution Approach 1:
The patent makes the hold-down timer dynamic by setting it to expire automatically after a predetermined period. This allows the system to initially prioritize reliability by blocking traffic during convergence, then automatically transition to normal operation once routes are established, minimizing total downtime
Solution Approach 2:
The patent implements periodic checking of route convergence status through the timer mechanism. The hold-down timer creates a structured time-based approach where traffic forwarding is periodically reassessed - initially blocked, then allowed once the timer expires and convergence is confirmed
Data Source
AI summary
A method, apparatus and computer program product for providing mulitcast failover and recovery which minimizes lost packets is presented. A first network device returns to a fully active state and starts a multicast hold-down timer. A first message is sent to a second network device, the message indicating at least one Split Multi Link Trunk (SMLT) Virtual Local Area Network (VLAN) Identifier (ID) correlating to the multicast hold-down timer. The first network device receives an acknowledgment of said first message and Layer 2 forwards multicast data traffic to said second network device until the timer expires, whereupon the first device is capable of handling its own multicast data traffic, and then has Layer 2 bridging turned off and enables IP multicast routing. During the period of multicast hold-down timer, the second device will perform IP multicast forwarding on behalf of the first device.