Cisco CCIE Service Provider Exam: Troubleshooting VPLS Performance Issues

Addressing VPLS Performance Issues: Techniques for Bandwidth and CPU Optimization

Question

A VPLS network consists of provider edge routers that have fully meshed pseudowires.

The network runs unicast and multicast traffic.

The network recently experienced bandwidth- and CPU-related performance issues.

While monitoring the VPLS statistics and interface statistics, you note that the network is experiencing an unusual amount of broadcasts and undesired multicast on the network.

Which two techniques should the operations team consider in addressing these problems?

Answers

Explanations

Click on the arrows to vote for the correct answer

A. B. C. D. E. F.

AD.

VPLS (Virtual Private LAN Service) is a type of Layer 2 VPN technology used to connect multiple sites in a single bridged domain. VPLS is implemented using pseudowires, which are point-to-point connections between provider edge (PE) routers that carry Layer 2 traffic.

When a VPLS network experiences performance issues related to bandwidth and CPU utilization, it is important to investigate the cause of the problem. In this scenario, the network is experiencing an unusual amount of broadcasts and undesired multicast on the network. This could be due to a number of factors, including a misconfiguration of the VPLS network or the presence of rogue multicast sources.

To address these problems, the operations team should consider the following techniques:

A. Implement IGMP snooping in the VPLS network: IGMP (Internet Group Management Protocol) snooping is a feature that allows a switch or router to listen in on the IGMP conversation between hosts and multicast routers. By doing so, the switch or router can determine which ports need to receive the multicast traffic and which ports can be pruned. In the context of a VPLS network, implementing IGMP snooping can help to reduce the amount of undesired multicast traffic on the network.

B. Enable split horizon between all the pseudowires in the VPLS network: Split horizon is a technique used to prevent loops in a network. In the context of a VPLS network, enabling split horizon between all the pseudowires can help to prevent loops from forming and reduce the amount of broadcast traffic on the network.

C. Set a limit on MAC learning in the VPLS network: Setting a limit on MAC (Media Access Control) learning in the VPLS network can help to reduce the amount of unnecessary broadcast traffic on the network. By limiting the number of MAC addresses that can be learned on a particular port or VLAN, the network can reduce the amount of broadcast traffic that is generated by unknown unicast traffic.

D. Implement Storm Control in the VPLS bridge domain: Storm control is a feature that allows a switch or router to monitor the amount of broadcast, multicast, or unicast traffic on a particular port or VLAN. When the traffic exceeds a predefined threshold, the switch or router can take action to limit or drop the traffic. In the context of a VPLS network, implementing storm control can help to reduce the amount of broadcast and multicast traffic on the network.

E. Implement MAC withdrawal on each EFP in the VPLS bridge domain: MAC withdrawal is a feature that allows a switch or router to withdraw MAC addresses from the forwarding table when they are no longer needed. In the context of a VPLS network, implementing MAC withdrawal on each EFP (Ethernet Flow Point) in the VPLS bridge domain can help to reduce the amount of unnecessary broadcast traffic on the network.

F. Change the network to run VPWS to manage the broadcast traffic efficiently: VPWS (Virtual Private Wire Service) is a Layer 2 VPN technology that provides a point-to-point connection between two sites. In the context of a VPLS network, changing the network to run VPWS can help to manage the broadcast traffic more efficiently by reducing the size of the bridged domain.

In conclusion, when a VPLS network experiences performance issues related to bandwidth and CPU utilization, it is important to investigate the cause of the problem. Implementing IGMP snooping, enabling split horizon, setting a limit on MAC learning, implementing storm control, implementing MAC withdrawal, or changing the network to run VPWS can help to address the problems caused by excessive broadcast and multicast traffic on the network.