New Training: Multicast Routing Foundations and Design

In this 16-video skill, CBT Nuggets trainer Knox Hutchinson explains how the fundamentals of multicast networking work Watch this new networking training.

Learn networking with one of these courses:

• Cisco Network Services

• Designing Cisco Enterprise Networks

This training includes:

• 16 videos

• 1.3 hours of training

You’ll learn these topics in this skill:

• Introducing Multicast Fundamentals

• The Point of Multicast

• IGMP and PIM

• Source Trees

• Shared Trees

• Rendezvous Points

• Reverse Path Forwarding

• Summarizing Multicast Fundamentals

• Introducing Multicast Routing Design

• Understanding IGMP

• PIM Dense-Mode (PIM-DM)

• PIM Sparse-Mode (PIM-SM)

• Bidir-PIM

• Source Specific Multicast (SSM)

• MSDP

• Summarizing Multicast Design

Preserve Your Network’s Resources with Reverse Path Forwarding

If you want to make sure that your traffic reliably navigates particularly risky sections of your topology, you might set up a redundant multicast topology. To do so, you tell one of your routers that it should forward all of Interface A’s packets to Interface B and all of Interface B’s packets to Interface A. While this can go a long way to ensuring redundancy, it can also lead to routing loops.

Routing loops aren’t the end of the world. After all, TTLs exist for that very reason. But if they happen often enough, they can sap network resources quickly. Routing loops most often occur in redundant multicast topologies.

A solution to this is Reverse Path Forwarding (RPF). Put simply, RPF gets combined with a multicast routing protocol and looks backward — rather than forward — to a packet’s source address. The source address is looked up in a routing table, and the decision of whether and where to multicast is based on that, and those routing tables are built to allow for redundancy but prevent loops. In addition to helping provide loop-free multicast forwarding, RPF can also be useful for preventing IP address spoofing in unicast.