Peering basics: Public vs. private peering

When we make Leaseweb Service Status announcements, we often mention our peering connections. We assume that the readers of those messages technically understand peering. But as they say, assumption is the mother of all mistakes. Sometimes it’s good to start with the basics of things. In this blog, we’ll take a closer look at peering and examine the differences between public and private peering. 

What is peering?

So, what is peering? In its most simple form, it is defined as the interconnection of two different networks so they can exchange traffic between them. This means the traffic goes directly from one network to the other, allowing them to reach each other’s customers. As this is mutually beneficial for those involved, the network owners usually don’t charge each other to peer.  

Now, I should note that having many peering connections doesn’t mean that you are able reach the whole internet – you only reach the networks (and their customers) you peer with. This is why a quality network should also be connected to at least two transit providers (preferably tier 1 networks). However, transit providers charge a sizable fee, making peering a cost-effective alternative. 

Read More: To peer or not to peer – reflections about IP Interconnection from the Global Peering Forum 14 in Montreal 

Public vs. Private Peering

There are two types of peering – public and private: 

1. Public peering is performed across a shared network called an Internet Exchange Point (IX or IXP). Through an Internet Exchange you can connect to many other peers using one or more physical connections, thereby optimizing the cost per peer when sending traffic to many different networks. Internet Exchanges often charge a port and/or member fee to keep their infrastructure intact. 

1. Private peering is performed by creating a direct physical connection (usually consisting of one or more 10GE fibers) between two networks. The connection is made from only one network to another, for which you pay a set fee to the owner of the infrastructure that is used (such as a data center). This makes private peering a sensible option when you need to send large volumes of traffic to one specific network, as the cost per megabit goes down when more traffic is exchanged. 

Because of the way these types of peering are designed, each offers several (dis)advantages: 

	Public Peering	Private Peering
Advantage	Efficient usage of ports Hundreds of peers available at the larger IXs Easy to administrate New peers added instantly on a daily basis	Guaranteed capacity Easy to monitor More reliable than public peering More secure than public peering
Disadvantage	Port and/or member fees for the IX	Only cost-effective for large volumes of traffic Takes more time to setup new peering connections

Conclusion

When you’re a network engineer, you’re constantly looking for fast and reliable ways to ensure data reaches its destination. As peering is a cost-effective solution for this fascinating puzzle, usually enabling us to offer you bandwidth at low prices. There are also other advantages to peering, which we will look at in the next peering blog. In the meantime, if you have any questions, don’t hesitate to drop me a line in the comments below. 

Read more: Network economics: more bandwidth for your buck 

Editor’s note: This blog post was originally published on 24 October 2012 and was updated on 1 November 2022.