What is the need of routing

IP routing

The Internet Protocol (IP) is a routing-capable protocol and ensures that data packets find their way to other hosts across network boundaries. It can convey the data over any type of physical connection or transmission system. The high flexibility contrasts with a high degree of complexity in finding the way from the transmitter to the receiver. The process of finding the way is called routing.

What is routing

Routing is a process that determines the route to the next station in a data packet. The focus is on the choice of route from the available routes that are stored in a routing table.

Parameters and criteria for routing

Various parameters and criteria can be important for the choice of a route:

  • Connection costs
  • necessary bandwidth
  • Destination address
  • Subnet
  • Connection type
  • Connection information
  • known network addresses

Why is routing necessary?

The basic connecting element in an Ethernet network is the hub or switch. All network participants are connected to it. When a host sends data packets, the packets are sent to and accepted by all stations in the hub. However, only the addressed host processes the packets further. However, this also means that all hosts have to share the total bandwidth of this hub (e.g. 10 Mbit or 100 Mbit). Although the physical structure and cabling of the hub is a star with point-to-point connections, the logical structure corresponds to a bus. In other words, a single line to which all network participants are connected. If two or more hosts want to transmit at the same time, a collision occurs which leads to a general pause in transmission on the bus. Then the hosts try to send again until the transfer is successful. This process is called CSMA / CD. The maximum number of hosts on an Ethernet bus is 1,023. The more hosts connected to a hub, the more frequent collisions occur that overload the network.

In order to exclude the disadvantages of Ethernet in connection with CSMA / CD, a switch is selected as the coupling element and Fast Ethernet is used (no more CSMA / CD). The switch remembers the hardware addresses (MAC addresses) of the stations and only forwards the Ethernet packets to the port behind which the station is located. If a switch does not know the hardware address, it forwards the data packet to all of its ports (broadcast) and functions like a hub at this point. In addition to the limited memory size of the switch, many unknown hardware addresses have a negative impact on the performance of a network.

Neither a hub nor a switch is suitable for connecting large networks. For this reason, a network is divided into logical segments or subnets by routers and IP addresses.
The addressing by the Internet Protocol is designed in such a way that the network traffic remains within the subnets and only leaves the network when the destination is in another network.

In particular, the following problems in an Ethernet network make IP routing necessary:

  • Avoid collisions and broadcasts by limiting the collision and broadcast domains
  • Routing over different network architectures and transmission systems
  • Packet filter and firewall
  • Routing via backup connections in the event of a power failure

Avoid collisions and broadcasts by limiting the collision and broadcast domains

Broadcast-based protocols are often used to find the way from sender to receiver in local networks. For example NetBIOS in Microsoft-based networks and ARP in the TCP / IP stack. The logs keep sending out broadcasts to find the way to an unknown station. Broadcasts put a strain on a network. Routers prevent broadcasts from being forwarded if they are not dependent on using them themselves. Routers reduce the load on the network from broadcasts.

Routing over different network architectures and transmission systems

Network cabling is usually laid out in a structured manner. A distinction is made between primary, secondary and tertiary cabling that use different architectures and transmission technologies (Ethernet, Token Ring, FDDI, ATM, ISDN, WLAN, etc.). A router can be able to mediate between different architectures. This also includes the fragmentation of the data packets.

Packet filter and firewall

Security aspects do not go past routers either. Unwanted or unsafe data traffic can be filtered and prevented using IP addresses or TCP and UDP ports. Special firewall routers or routers with firewall functions are often used.

Routing via backup connections in the event of a power failure

The use of routers often creates a close-knit network that offers the data packet several routes to its destination. If a router fails, the routers communicate with each other and the data packets simply take a different route to their destination. If a line fails between two routers, B. Establish a backup connection. For example, a dial-up connection via the telephone network.
In large and modern networks, error detection and handling play a major role. Routers can log the network traffic and send messages to the network management via SNMP or execute commands from the network administrator.

IP routing algorithm

The IP routing algorithm applies not only to IP routers, but to all hosts that can receive IP data packets. The received data packets run through this algorithm until the data packet can be assigned or forwarded.

The first question of the routing algorithm is the question "Is the data packet for me?". If the destination address of the data packet matches its own IP address, then the data packet has reached its destination and can be processed.
If the address does not match, the second question is asked: "Is the data packet for my subnet?". The destination address is masked with the subnet mask. The remaining address portion is used to determine whether the data packet belongs in its own network section (subnet or subnet).
If the subnet is also incorrect, the third question is asked: "Do I know the route to the recipient of the data packet?". Sometimes the hosts know the route for certain IP addresses. If the route is known, the data packet is forwarded via this route.
If the route is not known, the fourth question is asked: "Do I know a standard gateway where I can forward the data packet?". The standard gateway is usually a router that distributes incoming data packets to its routing outputs based on the destination address and some rules. If there is no standard gateway, this leads to an error message. The data packet is discarded.

Overview: Internet Protocol

Other related topics:

Everything you need to know about IPv6.

Collection: IPv6

A PDF file of all articles on Internet Protocol Version 6 from this website. The compilation takes into account the introduction to the basics of IPv6 with detailed descriptions and numerous tables and figures. Learn more about the possibilities and relationships in the IPv6 network.

More information and to download

Everything you need to know about networks.

Network technology primer

The network technology primer is a book about the basics of network technology, transmission technology, TCP / IP, services, applications and network security.

I want that!

Everything you need to know about IPv6.

Collection: IPv6

A PDF file of all articles on Internet Protocol Version 6 from this website. The compilation takes into account the introduction to the basics of IPv6 with detailed descriptions and numerous tables and figures. Learn more about the possibilities and relationships in the IPv6 network.

More information and to download