What is the difference between the ISM and SSM multicast models?
With the ISM model, the routers in the network maintain source information. However, with SSM the receivers are responsible for specifying the desired source. The receiver application obtains knowledge of the source unicast IP address or URL through methods external to the multicast protocol, such as the user selecting the source from a list of available sources supplied by a web page.
How many possible globally scoped multicast addresses are available for ISM? For SSM?
The IANA allocates multicast addresses within the range 126.96.36.199 188.8.131.52 to SSM.
For the entire multicast range 184.108.40.206 220.127.116.11, there are 16 * 224 = 16 * 16,777,216 = 268,435,456 possible addresses available. Therefore, to calculate the available addresses for ISM applications, subtract the locally scoped addresses (that is, 28 = 256 addresses for the range 18.104.22.168 22.214.171.124) and the SSM addresses from this number. The result is 268,435,456 16,777,216 256 = 251,657,984. You have to divide this number by 32 because of the 32-to-1 mapping of addresses to Ethernet MAC addresses. The result is 251,657,984 / 32 = 7,864,312 possible addresses available for ISM and 16,777,216 / 32 = 524,288 for SSM. Because multicast applications randomly assign multicast addresses, the chances of having frames with duplicate MAC addresses on an Ethernet LAN are very low.
What is the difference between source and shared distribution trees?
With source trees, the source of the multicast stream is at the root of the tree, providing the shortest possible path between the source and receivers of the content. Rendezvous Points are the root of shared trees and often cause the multicast stream to take a suboptimal path to the receivers.
How does dense-mode multicast differ from sparse-mode multicast?
Dense-mode routers periodically flood the multicast stream to the entire network, even to last-hop routers that do not have active receivers for the group. Sparse-mode routers send the multicast stream only to last-hop routers that contain multicast receivers for the stream.
How do multicast routers route multicast packets downstream?
Multicast routers use reverse path forwarding (RPF) to forward frames to last-hop routers.
How do multicast routers route multicast packets upstream?
For forwarding PIM Join messages up the distribution, instead of checking to see if the source of the packet (the last-hop router, in this case) is reachable from the RP interface using RPF, the router checks its outgoing interface (OIF) list. If the router receives the packet from an interface in the OIF list, then the router forwards the packet out its RPF interface.
For forwarding multicast stream data up the distribution tree, you need to configure Bidir-PIM.
What is the difference between Auto-RP and BSR?
Auto-RP distributes the RP candidate information using dense-mode multicast, forcing you to enable PIM sparse-dense mode on all Auto-RP routers. BSR uses a hop-by-hop method, allowing you to configure only PIM sparse mode on all BSR routers.
What are the differences between Auto-RP, BSR, and Anycast RP?
The Auto-RP and BSR protocols do not allow you to configure load-balancing of streams within the same group between multiple active RPs. Anycast RP allows per-group load balancing.
What are the differences between CGMP and IGMP snooping?
Both CGMP and IGMP snooping provide you with the ability to restrict multicast traffic in a Layer 2 switched network to only those ports with active receivers attached. However, you must configure CGMP on the router to signal to the switch which ports to forward the traffic to. You do not need to configure IGMP snooping on the router because the switch inspects the IGMP messages directly.