Multicast

This guide explains the concept of multicast and shows how to configure multiple different variations.

Comments

Introduction

Multicast Addresses

Multicast Sparse Mode

Multicast Sparse & Dense Mode

IGMP Filter

IGMP Snooping

Auto RP

Introduction

Multicast is a communication method in computer networking that allows for the efficient distribution of data to multiple recipients simultaneously. In contrast to unicast, where a message is sent to a single recipient, and broadcast, where a message is sent to all devices on a network, multicast is a one-to-many communication method that only sends the message to the devices that have expressed an interest in receiving it. Multicast works by using a single source device to send a message to a special multicast address that is recognized by all devices on the network that have subscribed to that address. Devices that are not interested in receiving the message will not process it, which makes multicast a very efficient method for distributing information to a large number of devices without overwhelming the network with traffic. Multicast is commonly used in applications such as video streaming, online gaming, and audio conferencing, where large amounts of data need to be sent to multiple recipients simultaneously. It is also used in routing protocols, where network routers use multicast to exchange information about network topology and route updates.

Description

Multicast allows for the efficient distribution of data to multiple recipients simultaneously

Multicast is a one-to-many communication method that only sends the message to the devices that have expressed an interest in receiving it

Multicast works by using a single source device to send a message to a special multicast address

Devices on the network can subscribe to that address

Multicast is an efficient method for distributing information to a large number of devices without overwhelming the network with traffic

It is commonly used in applications such as video streaming, online gaming, and audio conferencing, where large amounts of data need to be sent to multiple recipients simultaneously

It is also used in routing protocols, where network routers use multicast to exchange information about network topology and route updates

The multicast traffic is spread across the network from source to many destinations using multicast routers

Multicast routers need to know who is wants to receive the traffic

Multicast configured routers replicate data packets received at one input interface, and send the copied data out on multiple interfaces

Multicast is widely used by many routing protocols such as RIPv2, EIGRP, OSPF

Class D addresses are used for multicast groups (224.0.0.0 - 239.255.255.255)

Protocol Independent Multicast (PIM) is used to route multicast traffic between routers

Internet Group Management Protocol (IGMP) is used by a host to request multicast traffic from a multicast router

IGMP-Snooping on a switch listens to packets being exchanged by the host and the multicast router, and learns on what interfaces multicast traffic should be forwarded

Multicast is not enabled by default on a routing device

The application & the hosts need to support multicast

A multicast IP address is always used as the destination while the device that generates the traffic is the source

Multicast supports only UDP traffic

The internet uses mostly unicast, which consumes bandwidth depending on the number of receivers

A disadvantage for multicast is that all network devices need to be configured

Multicast Addresses

Multicast IP addresses are used for one-to-many communication, where data packets are sent from one sender to multiple receivers. Multicast addresses are defined within the IPv4 and IPv6 address spaces. In IPv4, multicast addresses are defined within the range of 224.0.0.0 to 239.255.255.255. The range from 224.0.0.0 to 224.0.0.255 is reserved for local network control purposes, such as multicast routing protocols like OSPF. Addresses from 224.0.1.0 to 238.255.255.255 are globally routable and can be used for general multicast communication within networks. The remaining range, from 239.0.0.0 to 239.255.255.255, is reserved for administratively scoped multicast addresses, which are typically used within private networks or domains. In IPv6, multicast addresses are defined within the range of ff00::/8

Class

CIDR

Address Range

Multicast

224.0.0.0/4

224.0.0.0 - 239.255.255.255

Reserved Link Local

224.0.0.0/24

224.0.0.0 - 224.0.0.255

Globally Scoped

224.0.1.0 - 238.255.255.255

Source Specific

232.0.0.0/8

232.0.0.0 - 232.255.255.255

GLOP

233.0.0.0/8

233.0.0.0 - 233.255.255.255

Limited Scope

239.0.0.0/8

239.0.0.0 - 239.255.255.255

Control Blocks

Local Network Control Block (224.0.0.0 /24) - Multicast IP packets in this range cannot be forwarded between networks or subnets

Address

Used for

224.0.0.1

All hosts

224.0.0.2

All multicast routers

224.0.0.5

OSPF

224.0.0.6

OSPF

224.0.0.9

RIPv2

224.0.0.10

EIGRP

224.0.0.18

VRRP

Internetwork Control Block (224.0.1.0 /24) - Multicast IP packets in this range can be forwarded between networks or subnets

Address

Used for

224.0.1.39

Cisco-RP-Announce

224.0.1.40

Cisco-RP-Discovery

Local network Block (239.0.0.0 /8) – Multicast IP packets in this range can be used in local networks

Multicast MAC Addresses

In the context of multicast communication, MAC addresses play a crucial role in delivering data packets to multiple recipients efficiently.

Description

Multicast MAC Addresses: In Ethernet networking, multicast MAC addresses are used to identify groups of devices that are interested in receiving multicast traffic. These MAC addresses start with a specific prefix that indicates multicast communication, followed by a 28-bit multicast group identifier. Devices interested in receiving multicast traffic join specific multicast groups by subscribing to the corresponding MAC addresses.

Destination MAC Address: When a device sends a multicast packet onto the network, it needs to specify the appropriate destination MAC address in the Ethernet frame header. This destination MAC address is derived from the multicast IP address being used. A mapping mechanism called Address Resolution Protocol (ARP) or Multicast Address Resolution Protocol (MARP) is used to translate the multicast IP address into the corresponding multicast MAC address.

Ethernet Switches and Multicast Forwarding: Ethernet switches play a crucial role in multicast communication by intelligently forwarding multicast traffic only to those ports where devices interested in receiving the multicast packets are connected. When a multicast packet enters a switch, the switch examines the destination multicast MAC address and forwards the packet only to the ports where devices interested in that multicast group are located.

Efficiency and Optimization: Multicast MAC addresses enable efficient distribution of data to multiple recipients, reducing network bandwidth consumption compared to sending separate unicast packets to each recipient. By using multicast MAC addresses, network resources are utilized more effectively, and network performance is optimized.

MAC addresses in multicast start with the following 24 bits: 01:00:5E:XX:XX:XX

IGMP & Multicast

Description

The Internet Gateway Management Protocol (IGMP) is used by hosts and adjacent routers on IPv4 networks to establish multicast group memberships.

IGMP messages are used primarily by multicast hosts to signal their local multicast router when they want to join a specific multicast group and start receiving group traffic.

IGMP exists in version 1, 2 & 3 - version 2 is the default version on many devices.

IGMP Version 1

Description

Hosts announce to a router that they want to receive multicast traffic from a specifig group.

Two messages are used: Membership Query & Membership Report.

When a host wants to join a multicast group, it sends a Membership Report message to the multicast group address.

The router receives the message, and starts to forward the traffic to the interface it received the Membership Report.

Routers use the Membership Query periodicall to all hosts by using the destination group 224.0.0.1.

Every host is by default part of the multicast group 224.0.0.1, and multicast routers are part of multicast group 224.0.0.2.

Hosts receive the query message from the router and can reply with a report message to continue receiving multicast traffic.

The router resets the expiry timer (60sec) whenever a report message is received.

IGMP Version 2

Description

Version 2 is the default version on many devices.

Hosts can send a leave-group message to stop receiving multicast traffic.

It is possible to send membership queries to either all hosts (224.0.0.1) or directly to a specific multicast group.

A querier election process decides which of multiple routers in a segment becomes the "Acive Querier".

The router with the lowest IP in the segment is elected to become the "Active Querier".

IGMP Version 3

Description

Source-Specific Multicast (SSM) Support: IGMPv3 adds support for source-specific multicast (SSM), which allows hosts to specify particular sources from which they wish to receive multicast traffic. This feature enhances security and reduces unnecessary traffic by allowing hosts to filter out unwanted multicast sources.

Group Management: Like previous versions, IGMPv3 allows hosts to join and leave multicast groups dynamically. Hosts use IGMP messages to communicate their group membership status to the local multicast router.

IGMPv3 Membership Reports: IGMPv3 introduces new membership report messages that include source-specific information. Hosts can specify the multicast group they are interested in and indicate specific sources from which they want to receive multicast traffic. This granularity enables more efficient use of network resources.

Compatibility with IGMPv2 and IGMPv1: IGMPv3 is designed to maintain backward compatibility with earlier versions of IGMP. This ensures that IGMPv3 routers can still communicate with hosts running IGMPv2 or IGMPv1. However, hosts using IGMPv3 can take advantage of its enhanced features only when communicating with IGMPv3-enabled routers.

Querying Mechanism: IGMPv3 routers periodically send IGMP membership queries to discover which multicast groups hosts are interested in. These queries prompt hosts to send membership reports indicating their group memberships and source-specific information.

Robustness and Reliability: IGMPv3 includes mechanisms for robust operation in multicast environments, such as handling packet loss, retransmissions, and duplicate messages, to ensure reliable communication between hosts and routers.

PIM - Protocol Independent Multicast

PIM stands for Protocol Independent Multicast, which is a family of multicast routing protocols used in computer networks to efficiently route multicast traffic. PIM is designed to work independently of the underlying unicast routing protocol, hence the term "Protocol Independent." PIM operates by building multicast distribution trees within the network, connecting multicast sources to multicast receivers. These trees ensure that multicast traffic is delivered only to the network segments where there are interested receivers, thus conserving network bandwidth.

Multicast Modes

Sparse mode (SM) and Dense mode (DM) are two different modes of operation for multicast routing protocols, primarily used in Protocol Independent Multicast (PIM). These modes dictate how multicast traffic is forwarded within a network. Sparse-Dense mode, on the other hand, is a hybrid approach that combines features of both sparse and dense modes.

Dense Mode (DM)

In Dense mode, multicast traffic is initially forwarded to all network segments within the multicast domain (unless pruned by routers) and is pruned back as necessary based on receivers' interest.

Dense mode assumes that multicast traffic is dense within the network, meaning there are many receivers interested in receiving the multicast traffic.

When a router receives multicast traffic, it floods the traffic to all its interfaces except the one it received the multicast traffic on. Pruning messages are then sent to the upstream router if there are no interested receivers on a particular branch of the multicast distribution tree.

Flooding: Initially, multicast traffic is flooded throughout the network. When a router receives multicast traffic on an interface where it hasn't yet heard of the multicast group, it forwards the traffic out of all other interfaces except the one it received the traffic on.

Pruning: After flooding, routers start to prune unnecessary links. As routers learn about the distribution of receivers for a particular multicast group, they prune back interfaces that do not have interested receivers. Pruning helps to reduce unnecessary traffic by eliminating the forwarding of multicast packets on segments of the network where there are no receivers for that multicast group.

The process of Flooding & Pruning is repeated frequently (3 minutes by default). It is not recommended to use Dense Mode because of this behaviour.

Your description here…

Sparse Mode (SM)

In Sparse mode, multicast traffic is only forwarded to network segments where there are interested receivers (members of a multicast group). This mode assumes that multicast traffic is relatively sparse within the network.

The router initially forwards multicast traffic only on request from receivers. When a router receives an IGMP (Internet Group Management Protocol) membership report indicating interest in a particular multicast group, it starts forwarding multicast traffic for that group.

Sparse mode uses a join mechanism where routers send join messages towards the root of the distribution tree to request traffic for specific multicast groups.

In Sparse Mode multicast routing, the Rendezvous Point (RP) is a designated router responsible for managing multicast traffic distribution within the network. It acts as a focal point where sources send their multicast traffic and where receivers can request to join multicast groups.

The RP is crucial in Sparse Mode because it helps manage multicast group membership information efficiently. When a router receives a multicast packet for a particular group but doesn't have information about where receivers for that group are located, it sends a (*, G) join message to the RP. This message indicates a request to join the multicast group G to the RP, regardless of the specific source (denoted by the asterisk).

The RP maintains a mapping of multicast group addresses to the sources that are transmitting data for those groups. It then forwards multicast traffic to the routers that have explicitly requested it, based on this mapping and the join messages it receives.

Tunnel interfaces are used to transport multicast traffic between different parts of the network, particularly between routers and the Rendezvous Point (RP). These tunnel interfaces encapsulate multicast packets and send them across the network, ensuring that multicast traffic reaches the appropriate destinations efficiently.

RP Tree: In Sparse Mode, routers establish a shared tree rooted at the Rendezvous Point (RP) for each multicast group. This shared tree is used to forward multicast traffic from sources to receivers.

Shortest Path Tree (SPT): Once a receiver joins a multicast group, routers can switch from using the shared tree to using a Shortest Path Tree (SPT) directly from the source to the receiver. This optimizes multicast traffic delivery and reduces unnecessary traffic across the network.

Tunneling: When a router needs to forward multicast traffic to the RP or to receivers located on different parts of the network, it encapsulates the multicast packets in IP packets and sends them over tunnel interfaces towards the destination. These tunnel interfaces effectively create logical paths through the network for multicast traffic.

Encapsulation: Multicast packets are encapsulated with a multicast tunneling protocol such as Protocol Independent Multicast (PIM) encapsulation, Generic Routing Encapsulation (GRE), or IPsec. This encapsulation ensures that multicast packets can traverse the network and reach their intended destinations without being affected by the underlying network topology.

Sparse-Dense Mode (SDM)

Sparse-Dense mode is a hybrid mode that combines features of both Sparse mode and Dense mode.

SDM uses Sparse mode behavior by default but can also switch to Dense mode behavior in specific scenarios. For example, if a router receives multicast traffic for which it has no explicit join, it behaves in Dense mode by forwarding the traffic to all interfaces.

SDM is particularly useful in networks where multicast traffic patterns are variable or difficult to predict. It allows for efficient delivery of multicast traffic while still supporting dense mode flooding when necessary.

The “S,G” value in multicast routing refers to a specific source (“S”) and a specific group (“G”) within a multicast session. It is used to identify a particular multicast stream originating from a specific source and intended for a specific multicast group. This designation helps routers and switches determine how to forward multicast traffic efficiently within the network.

224.0.1.39 (All PIM Routers): This multicast group address is used for Protocol Independent Multicast (PIM) routing protocol. It is designated for communication between PIM routers, allowing them to exchange control messages related to multicast routing. PIM is a multicast routing protocol used to establish and maintain multicast distribution trees within a network.

224.0.1.40 (All PIM Routers with Border Gateway Protocol (BGP) peering): Similar to 224.0.1.39, this multicast group address is also used for communication between PIM routers. However, it specifically indicates routers that have BGP peering enabled. BGP is a path-vector routing protocol used in the Internet to route traffic between different autonomous systems (ASes). When BGP is used in conjunction with PIM for multicast routing, routers may use this multicast group address to exchange routing information.

Premium Access

0 Comments

Submit a Comment Cancel reply

You must be logged in to post a comment.