The internet is like a complex web of highways.
Highways that move your data from one end of the globe to another.
When driving the roads in your home Country, here we drive on the Autobahn, you need a map to know what roads to take to arrive at your destination.
The same happens on the internet. This map of roads is the BGP routing table. This is a map that tells border gateway protocol (BGP) where to route internet transit.
Here, we will take a tour in understanding the BGP routing table.
What is the BGP routing table?
The BGP routing table is a map of the internet. The routing table is where all internet transit routing information is stored. This information tells the routers when the next hop is in the AS_Path to an IP address.
The BGP routing table is not so different from a general routing table in networking. The size and scale of the routing table makes it unique. Routing tables are common in internal networks, enterprise networks and data center networks.
You can see updates of the BGP routing table here.
The stored information is to route network transit.
BGP routing table size and growth
The growth of the routing table has been astronomical. Also, there is much debate about whether it will be able to handle the increased growth of the internet.
In 2014 the BGP routing table experienced a milestone. The routing table hit the 512k mark of internet routes. At that point, the internet did experience growing pains.
The growing pains have prompted a push to IPv6, which is designed for the continued growth of the internet. This is one of the solutions to the BGP routing table growth.
What is the function of the BGP router?
The BGP routing table is to compile internet transit data to route traffic through the internet.
The BGP routing stores local preference data, peering data, and transit cost metrics. This allows the BGP router to choose the next hop in the AS_Path.
Thus, the BGP router is the network router that implements BGP for internet transit to move between autonomous systems. BGP routers are how internet service providers (ISPs) exchange information between them and AS’s.
BGP Routing Table vs. IP Routing Table
As with everything in internet and networking, you should always adhere to the rule of thumb: “it depends”. The same is with the BGP routing table vs. the IP routing table.
First, we should understand the IP routing table. The IP routing table works like other routing tables. The primary distinction is that the IP routing table routes to IP addresses. It is routing a particular data packet from one IP to the next.
The difference is that the BGP routing table is concerned with network prefixes and routing data between autonomous systems. Where the IP routing table is concerned with routing data packets between IP addresses.
A small distinction, but worth noting.
Peering agreements in the BGP Routing Table
Peering agreements are important to how the internet backbone works. They can greatly improve internet connectivity.
Also, peering agreements occur at internet exchange points. This is a more important factor in the speed of the internet than the material of your cables.
Thus, peering agreements are logged in the BGP routing table to alert the network router what the next hop on the AS_Path should be. Peering agreements with local providers in a given geographic area can improve the quality of service to that region.
This implementation of peering agreements happens at the internet exchange point, and the network router level.
Understanding BGP routing tables provides more insight into the current speed of the internet. The internet is large, complex, and full of peering agreements. The BGP routing table is what keeps everything up to date.
To learn more about networking terms as they relate to DevOps, you can download the DevOps networking eBook.
