Routing: Choosing the best pathways since 1976!

                        

   (Image Source: https://www.cisco.com/c/en/us/products/routers/what-is-routing.html)

Routing basically means ‘to route’. The aim of the routing process is to provide a way out for the network traffic to reach from the source to the destination and this destination can vary from 1 to multiple locations. Thus, Routing can be defined as the path/route for network traffic flow from the source to the destinations, being both in same network or different networks. Routing is controlled by the Router and router is the device which actually defines the whole path for routing.

 

Routing Process:

Routing depends on various factors like Administrative distance, ASN, Interface, next hop and mainly on Destination Network. Whenever the traffic flows out from the source and reach to the Router, the router at first check the destination IPv4 or IPv6 address and then proceed further by checking the Forward Information Base or FIB which consists of 3 main elements and those are Destination Network, Next Hop and Outgoing Interface. This FIB is generated by the RIB aka Routing Information Base which contains prefixes, routing tables, metrics, and next hop information. We will read about all this later in this document.

So once the Datagram reach to the router, the router will check the destination IP address and referring the FIB, it will send the information to the destined IP address and this network flow can be of unicast nature or multicast nature. It is not bounded to one form of flow only.

 

Routing Components:

1.)    Router: The router is a hardware device which functions to flow the network traffic in multiple or unicast way. It uses routing tables, and algorithms to decide the right path and to ensure to let the traffic reach its right destination.

2.)    Administrative Distance (AD): Numerical values assigned to different routes or protocols from 0 to 255, basis on which the preferred path is selected or rejected. It is a numerical value of trustworthiness of a routing information gained from different sources. The higher the AD value, lower will be the chances of its selection. Thus, can say that AD is inversely proportional to the numerical value. AD is one of the most important or the prior element checked by the router to forward any traffic. For example, if a router receives a route for a particular destination from two protocol one follows RIP with value as 120 AD and another static route as 10 AD, then the router will prefer the static route only as it is having lesser AD value.

3.)    Routing Protocol: Set of rules and procedures to make a protocol that function is to maintain Routing Tables is called Routing Protocol. Example: OSPF, BGP, EIGRP.

4.)    Routing Table: It is a database of a router which contains information like destination Network, network topology, or available routes in the network. Thus, this is very useful for the routers and on basis of this the RIB (Routing Information Base) is prepared and maintained which further generates the FIB (Forward Information Base).

5.)    Interface: A connection point located on a router device to connect to a network and each interface has its own IP address and subnet mask assigned. Signifies as G0/1 or other symbols but in this manner only. This interface can be physical or virtual as well. Each interface on a router can also have the configuration of other factors like default gateway, access control lists (ACLs), quality of service (QoS) policies etcetera.

6.)    Metrics: Metrics including various factors like hop count, bandwidth, or delay determines the best route for any data gram.

7.)    Path select Algorithm: By considering different factors like metrics, AD, policies, the path select algorithm provides and choose the best available paths and then the traffic to the destined IP is sent.

 

Types of Routing:

1.)    Static Route: Manually modifies, added and maintained by a Network Administrator only.

2.)    Dynamic Route: Gradual addition of routes that are learned by Network devices from the different routing protocols and they share the best possible route information with each other.

3.)    Default Route: These are the routes which are assigned as the default one when the device lacks the routing destination information in its routing table, then the device transfers the traffic to the default gateway or route which then send it the appropriate destination.

4.)    Black Hole Route: The main purpose of black hole route is security, thus whenever a selected segment or IP address from which the traffic is blacklisted try to hit the router, that IP address will by default fall down in the black hole route and get discarded. This is also called as null route.

5.)    Interior Route: The interior route is basically a route which is flowing inside a same Autonomous System (AS) only and managed by interior gateway protocols only, like the Route in Corporate Network.

6.)    Exterior Route: The routes that are learned from outside the AS like via the internet are exterior routes only and they follow the exterior gateway protocols.

7.)    Floating Static Routes: It is also called as backup route, because of its function that whenever the primary route fails to reach the destination, the backup or floating static route will function and let the datagram reach the appropriate destination. The AD of floating static route is higher than the primary ones.

 

Forwarding Information Base or FIB:

Forwarding Information Base (FIB) is a database table used by a router to know the next-hop address and interface for forwarding a packet. FIB is generated by the routing information base or RIB. When a packet arrives at the router, the router checks the destination IP address against and refer the FIB to determine where to forward the packet by seeing the destination network, next hop and outgoing interfaces and on basis of the information the router forwards the packet. FIB entries are typically stored in a hash table or a database structure, which allows for fast lookup and retrieval of the next-hop address or interface.

 

Routing Information Base:

It is a database where routes and route related metadata is stored by a routing protocol – allowing the routing protocol to select a ‘best’ path to a given destination. Each protocol has its own separate RIB. RIB functions as a backbone for FIB, without which the FIB can’t function. RIB consists of Routing tables, prefixes, next hop information and metrics.

Routing Protocols:

OSPF:

Open Shortest Path First, is a link-state routing protocol used to map the path with the shortest distance. It is a dynamic interior gateway protocol which uses link-state Algorithm and can work as OSPFv2 for IPv4 address using RFC 2328 and OSPFv3 for IPv6 address using RFC 5340. The OSPFv3 can be used for IPv4 and IPv6 as well by using RFC5838. The AD in case of OSPF is 110 and it is a fixed numerical value. The OSPF supports Hierarchical Routing. OSPF processes as by first giving a ‘Hello packet’ to the neighboring routers in same AS which will lead to exchange of topology among neighbors via Link-state advertisement. Once the hello is sent to all, then a topology map will be prepared for the network by creating a link-state Database. Using the database calculation for the best path is done and that is update in the OSPF tables. OSPF divides the routers into different areas starting from area0 to area n respectively. OSPF Router can Internal router which is for same area and External router which is for different areas. OSPF also supports other features like support for multiple paths to a destination, unequal cost load balancing, and authentication mechanisms to ensure secure routing information exchange.

 

Intermediate System to Intermediate System:

Intermediate System-to-Intermediate System (IS-IS) is a link-state, Interior gateway protocol that uses modified version of Dijkstra Algorithm. The AD value for IS-IS is 115. An IS-IS network has range of components, routers, areas, and domains. Just like OSPF it also organizes routers into areas and multiple areas together form a domain. It uses two network addresses, one is Network Service Access Point (NSAP) and other is Network Entity Title (NET).

Routing Information Protocol:

RIP is an Interior gateway protocol that also runs on Application layer of the OSI model. Like OSPF it has also two versions as RIPv1 and RIPv2. The former version functions to find network path based on IP destination and the hop counts by broadcasting IP tables to all routers in the network. While the later one or RIPv2 being more precise sends the IP tables to multicast addresses only. RIP, AD is with a fixed value of 120. RIP is not a suitable protocol for larger networks as it has limitations of hop count as 15.

Enhanced Interior Gateway Routing Protocol:

EIGRP is a distance vector and link-state routing protocol. Thus, also known as ‘Hybrid Protocol’. EIGRP is a Cisco proprietary protocol that was designed to follow on from the original IGRP protocol. EIGRP has features like bandwidth, reliability, maximize efficiency etc., whenever multiple paths to the same destination are available, EIGRP will select the path with the lowest metric, regardless of the administrative distance.  In EIGRP, the router takes information from the routing table and keep a record of the same, whenever a change or update occurs in the path the router informed the neighbors and they do update the tables accordingly The AD for EIGRP is 90 for internal EIGRP routes, and 170 for external EIGRP routes.

Broader Gateway Protocol:

BGP is distant-vector routing protocol designed to replace Exterior gateway protocol. The AD vale for BGP is 20 for eBGP (external BGP) routes and 200 for iBGP (internal BGP) and in the time of selection with multiple path value the BGP will always select the one with lower AD value independent of Metric value. BGP uses best path selection Algorithm. No auto-discovery of table like events happen in BGP case, i.e., user has to configure BGP manually.

Routing Algorithms:

Routing algorithms are the algorithms that implements different routing protocols by assigning a cost number to each link, which is calculated using various network metrics and aim is always to transfer the data packet with the lower cost value.

1)      Distance Vector Routing: This routing algorithm updates the best path information to all known destinations irrespective of same AS and different AS.

2)      Link State Routing: In Link State Routing, Same AS network discovery of best path among neighboring routers take place. Using the information, a map is created and best path is then calculated.