Menu

network collision

Collision Domain and Broadcast Domains

Collision Domain and Broadcast Domains

/ /

Collision Domain

The “collision domain” describes a network where packet collisions can occur when two devices on a shared network medium send packets simultaneously. The colliding packets are discarded and must be sent again, which reduces network speed and efficiency.

Collisions usually occur in a hub environment because each port on a hub is in the same collision domain. So, all devices connected to the hub are in the same collision domain, and only one device can transmit at a time. All other devices must listen to the network to avoid collisions. Total network bandwidth is shared among all devices.

In contrast to hubs, each port on a bridge, switch, or router has a different collision domain, which reduces and eliminates the possibility of collisions and enables the devices to use full-duplex communication. The full-duplex communication effectively doubles the speed of data capacity. To understand the collision domains, examine the following figure:

You can see that there are eight collision domains marked in the above topology. The hub is in a single collision domain, or all ports of the hub are in a single collision domain, but each port of the router, bridge, and switch is in a separate collision domain.

Broadcast Domain

All the devices in the broadcast domain can be reached via broadcast at the data link layer. A Broadcast Domain can receive any broadcast packet originating from any device within the network segment. All hub and switch ports belong to the same broadcast domain, but all router ports belong to a different broadcast domain.

All ports of the hub and switch are in the same broadcast domain. Hubs and Switches send broadcasts out to all interfaces except the interface on which they were received. Routers do not transmit broadcasts because when a router receives a broadcast, it does not forward it to other interfaces.

Each router interface belongs to a different broadcast domain, and each broadcast is only propagated within its specific domain. Routers separate the boundaries of the broadcast domains. Now, examine the same figure for the broadcast domain.

Collision Domain

In the figure above, you can see four broadcast domains marked. All ports on a hub, bridge, and switch are in the same broadcast domain, and all interfaces of the router are in a different broadcast domain.

Layer 2 devices send broadcasts known as ARP to a known IPv4 address on the local network to discover the MAC address. The host can get IP address configuration from the DHCP server using the Dynamic Host Configuration Protocol (DHCP). A large broadcast domain can connect many hosts. A problem with a large broadcast domain is that it generates excessive broadcasts and negatively affects the network.

A large number of Broadcasts also decreases the bandwidth of the network for normal traffic because the broadcast traffic is forwarded to all the devices in the domain. It also decreases the processing power of computers and network devices. Because the computers and network devices need to process all the broadcast packets received, a part of the CPU power is spent on processing the broadcast packets.

So we need to decrease broadcast. To decrease broadcast, we need to enter a router into the network. The router is an expensive device, So it is not possible to add a router to many networks. Another solution for decreasing is subnetting.

Subnetting is the process which decreases network traffic, improves network performance, and makes management very easy. We can isolate network segments very quickly using subnetting. We can also apply security policies, such as which subnets are allowed or not to communicate together. There are different ways of using subnets. Network administrators can also set services into subnets such as:

  • Different floors in a building
  • Different Organizations
  • A different section of the organizations
  • Types of Different Devices, such as servers, printers, and hosts
  • Any other division that makes sense for the network.
Carrier Sense Multiple Access (CSMA)

Carrier Sense Multiple Access (CSMA)

/ /

CSMA is the abbreviation for Carrier Sense Multiple Access. This networking protocol listens to network signals on the carrier/medium before transmitting data on the media. CSMA is implemented in Ethernet networks with more than one computer or network device attached to it. It is part of the Media Access Control (MAC) protocol. The protocol decreases the chances of collisions when two or more stations start sending their signals over the data-link layer. It requires that each station first check the state of the medium before sending.

Carrier Sense Multiple Access/Collision Detection (CSMA/CD)

Carrier Sense Multiple Access/Collision Detection (CSMA/CD); is the most widely used transmission method in half-duplex Ethernet networks.  On Ethernet, any device can attempt to send a frame anytime. Each device senses whether the line is idle and available for use. If it is, the device begins to send its first frame. If another device has tried to send simultaneously, a collision occurs, and the frames are discarded. Then, each device waits a random amount of time and retires until it successfully gets its transmission again.

Carrier Sense Multiple Access with Collision Avoidance

CSMA/CA (Carrier Sense Multiple Access/Collision Avoidance) is a method for carrier transmission in 802.11 networks. If a collision has occurred in the wired network, then the energy of the received signal almost doubles, and the station can sense the possibility of collision. In the case of wireless networks, most of the energy is used for transmission and the energy of the received signal increases by only 5-10% if a collision occurs. The station can’t use it to sense collision.

Therefore, CSMA/CA has been specially designed for wireless networks. It uses a method similar to CSMA/CD to detect if the media is clear. CMSA/CA also uses additional techniques. This method does not detect collisions but attempts to prevent collisions before they happen. Each device that transmits includes the time duration that it needs for the transmission. All other wireless devices receive this information and know how long the medium will be unavailable.

In CSMA/CA, when a node receives traffic for sending, it checks to ensure the channel is clear (no other node is transmitting at the time). The packet is sent when the channel is clear. If the channel is not clear, the node waits for a randomly chosen period of time and then checks again to see if it is clear. This period of time is called the backoff factor and is counted down by a backoff counter.

The node transmits the packet if the channel is clear when the backoff counter reaches zero. If the channel is not clear and the backoff counter reaches zero, the backoff factor resets again, and the process repeats. Carrier sense multiple access with collision avoidance (CSMA/CA) is the least popular of the access methods. The WLANs now use this access method.