How can EGP help achieve policy-based routing and traffic engineering across domains?

1 What is EGP and how does it work?

EGP is a generic term that refers to any routing protocol that operates between ASes, which are groups of routers that share a common routing policy and administration. EGP allows ASes to advertise their reachability to other ASes, and to learn about the reachability of other ASes. EGP also allows ASes to apply their own policies and preferences to select the best routes to reach other ASes, based on factors such as cost, bandwidth, latency, security, or load balancing. The most widely used EGP is BGP, or border gateway protocol, which is the de facto standard for interdomain routing on the internet.

2 How can EGP help achieve policy-based routing?

Policy-based routing is the process of applying specific rules or criteria to route packets differently based on their source, destination, or content. Policy-based routing can be used to implement various network objectives, such as enforcing access control, prioritizing traffic, diverting traffic, or avoiding congestion. EGP can help achieve policy-based routing by allowing ASes to communicate their policies and preferences to other ASes, and to influence the routing decisions of other ASes. For example, an AS can use EGP to advertise different routes to different ASes based on their relationship, such as customer, provider, or peer. An AS can also use EGP to filter or modify the routes received from other ASes based on their attributes, such as origin, path, or community.

3 How can EGP help achieve traffic engineering?

Traffic engineering is the process of optimizing the utilization and performance of network resources, such as bandwidth, delay, or reliability. Traffic engineering can be used to improve the quality of service, reduce the operational costs, or increase the resilience of the network. EGP can help achieve traffic engineering by allowing ASes to control the distribution and allocation of traffic across multiple paths or links between ASes. For example, an AS can use EGP to balance the load across multiple links to the same destination AS, or to prefer a link with lower cost or higher bandwidth. An AS can also use EGP to reroute the traffic around a link failure or congestion, or to isolate a link for maintenance or testing.

4 What are the advantages of using EGP?

Using EGP for interdomain routing has several advantages, such as scalability to handle the large and dynamic nature of the internet, flexibility to accommodate the diversity and heterogeneity of the internet, autonomy to preserve the independence and sovereignty of the ASes, and efficiency to improve network performance. With EGP, thousands of ASes can exchange routing information and update their routes frequently, while each AS can decide its own routing policy and preference without being dictated by a central authority or a common protocol. This allows ASes to optimize their interdomain routing based on their own criteria and objectives.

5 What are the limitations of using EGP?

Using EGP for interdomain routing has some limitations, such as complexity and overhead, instability and inconsistency, and insecurity. Complexity is introduced as ASes need to manage and exchange a large amount of routing information and attributes, and to apply complex policies and filters to their routes. Instability can be caused by conflicting or incompatible policies or preferences, or route flapping or oscillation due to frequent changes or errors in their routes. Insecurity can occur due to attacks or misconfigurations that can compromise their routes or traffic, such as prefix hijacking, route leaking, or denial of service.

6 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?