Explain Network Segmentation and Basic Traffic Management Concepts
A LAN without any network segments is a single broadcast and collision domain. In a broadcast domain a group of devices on a network hears all the broadcasts that are sent on that network segment. This may cause unnecessary traffic on the LAN.
A collision domain on the other hand can be defined as a network scenario where one device sends a packet on a network segment and forcing other devices on the network segment to pay attention to the packet sent, thus causing collision by causing other devices on the network to transmit the data at the same time.
When a network has too many network hosts, the traffic congestion becomes a problem because too many packets are transmitted on a network. The performance of the network may degrade to an extent that no packet is delivered. To avoid network traffic congestion, a large network is usually broken into small network segments. The breaking of a network into smaller network segments is called subnetting.
Subnetting or Segmenting a LAN improves its performance. Consider a single department has more traffic than other two departments in an office. In such a scenario, entire traffic meant for different departments move on the same network and would cause network congestion and reduced performance. If the same LAN is segmented for different departments then the performance would increase as the unnecessary traffic would not move on each network segment.
Segmentation also allows you to isolate network problems, improve network performance, solve network congestion issues, and maximize the available bandwidth. In addition, the segmentation allows hardware and software failures to restrict to a network segment instead of affecting the entire network.
Once a LAN is segmented, the distances are no longer limitations because each segment can be created with maximum distance for a specific Ethernet type. With segments the collisions and latency is also decreased, throughput is improved and bandwidth per user is increased.
Segmenting a Network
The segmenting of a network is possible only through network devices such as switches, routers and bridges.
The bridges just send all the data to all the computers connected to it but they have just one source and one destination to deliver data packets. A bridge has two interface devices. It receives data from one interface and delivers it to the other interface, it is connected to. Thus a bridge creates two collision domains. A bridge is also called an intelligent hub because it considers the destination of data before delivering it.
Bridges are mainly used to break the collision domains and filter frames on the basis of layer 2 fields. The bridges can be programmed to reject frames from a particular source and may help you to get rid of unnecessary broadcast and multicast packets. The bridges available these days may have up to 16 ports.
Just like bridges, the switches are also data link layer devices that allow multiple LAN segments to be interconnected into a single larger network. Switches performed on hardware instead of in software and therefore much faster than routers that perform on software. Switches forward and flood traffic based on MAC addresses. They learn the MAC address of the requester and the port or the location of the device which responded to the request, almost instantly.
The first time, a request received by a switch is sent to all the computers connected to it. However, as soon as the request is responded by a computer, the switch learns the network location of the port that responded to the request and the MAC address of the source computer to handle the similar subsequent requests.
Switches create address tables to handle network traffic. They have the capability to intelligently monitor the network traffic. Switches can provide nonblocking service, which allows multiple conversations between two ports to occur simultaneously.
Switches create many collision domains on the same segment. They make each port its own collision domain. Switches can also be used to create VLANs or virtual workgroups by allowing you to create virtual segments instead of physically segmenting the network.VLANs logically break a LAN in such a way that each segment is a separate Layer 2 broadcast domain. The packets in a VLAN are switched only between the ports that are a part of the same VLAN.
A VLAN is a bridging domain and provides a complete isolation between other VLANs. All broadcast and multicast traffic is contained within a VLAN. The VLANs communicate with each other with the help of routers.
VLANs allow easy administration of network, easy management of network broadcasts and good security. Each time a user moves from one network to another, the cables need not be pulled and rearranged for the user. The network changes can be made by simply configuring the VLAN with the new port.
Routers are the most intelligent among all the network devices. They can be programmed to use the most efficient route to transmit the data to the desired computers. They operate on the Network Layer 3 of OSI model and can route data packets from one network to another based on their IP address. Also they don’t forward broadcasts by default.
Routers make each of its port into a separate segment and a separate collision domain. Thus routers have many segments and many collision domains. The routers can separate broadcast domains unlike switches that create separate collision domains but same broadcast domains.
By default, routers break the broadcast domains and keep all devices separate on a segment from other segments. This means, if messages are sent on a segment, the devices on that segment will be able to get that broadcast message and not all other segments of the same network. The functions of a router are: Path selection, Packet filtering, Packet switching, and Internetworking. The routers perform packet switching and internetworking using logical addressing and packet filtering using access lists. Routers use routing tables or map of the internetwork to make path selection and send data packets to remote networks.
Which of the following is not an advantage of using routers for segmenting?
- Routers do not forward broadcasts and multicast frames by default.
- Routers only forward packets to the network segment to which it is intended for.
- Routers typically perform more functions before a packet is routed.
- None of the above.
‘Routers performing more functions before routing a packet’ feature is definitely not an advantage of using routers for segmenting because this causes delays and latency to the network.