Dynamic Routing Protocol Overview

• Internet is made up of a large number of independent networks or as they are called Autonomous Systems (ASes).
• Internal routing algorithm selection is done by individual organisations and usually called intradomain routing using Interior Gateway Protocols (IGP).
• Routing between independently operate networks, or interdomain routing, use Exterior Gateway Protocols (EGP).

• An AS typically belongs to one organization.

  • A single or multiple interior gateway protocols (IGP) may be used within the AS.

  • In either case, the outside world views the entire AS as a single entity.

• If an AS connects to the public Internet using an exterior gateway protocol such as BGP, then it must be assigned a unique AS number which is managed by the Internet Assigned Numbers Authority (IANA).

• AS numbers can be between 1 to 65,535.

  • RIRs manage the AS numbers between 1 and 64,512.

  • The 64,512 - 65,535 numbers are reserved for private use (similar to IP Private addresses).

  • The IANA is enforcing a policy whereby organizations that connect to a single provider use an AS number from the private pool.

• If an AS connects to the public Internet using an exterior gateway protocol such as BGP, then it must be assigned a unique AS number which is managed by the Internet Assigned Numbers Authority (IANA).

AS numbering

Politics in routing

Interdomain routing protocols need to deal with politics

Typical political policies:

1. Educational institutions do not want to carry commercial traffic.
2. Military, Secret service or some government traffic do not want to route via certain countries.
3. Traffic from Google to Apple and vice versa do not want to crossover
4. Certain ISPs do not want traffic to pass to another due to cost or poor perfomance.

• Router will prefer routes learned from paying customers first.

• Secondly will prefer routes learned from settlement-free peers.

• Last will prefer routes learned from provider networks.

  Thus:

• An AS will advertise all of its routes to customers, but it will not re-advertise  routes learned from a settlement-free peer or transit provider to other peers or providers.
  Except:

• If two ASes have made specialised money arrangments - PAID PEERING

routing policies

• OPEN: opens TCP connection to remote BGP peer and authenticates sending BGP peer

• UPDATE: advertises new path (or withdraws old)

• KEEPALIVE: keeps connection alive in absence of UPDATES; also ACKs OPEN requests

• NOTIFICATION: reports errors in previous msg; also used to close connection

BGP Messages

• Routing Money

• Usually rely on settlement free peering

• Can change to payment

• No agreement

  • Peering Dispute

Peering Disputes

Peering Dispute Examples

1. NTT Communications vs. Cogent Communications:
Context: NTT Communications (a Japanese firm) and Cogent Communications (a U.S.-based company) engaged in a disagreement about peering arrangements.
Issue: The dispute revolved around peering in Asia.
Result: As a consequence, Cogent de-peered NTT in Europe, leading to inefficient routing. Traffic had to traverse the Atlantic and back again between these two networks in Europe, causing latency issues for users unless broadband providers found workarounds.

2. Verizon vs. Cogent Communications:
Context: Verizon, a major telecommunications company, and Cogent Communications, a bandwidth provider, engaged in a peering dispute.
Issue: The dispute centred around unbalanced peering—essentially an arrangement where two networks exchange traffic reciprocally without payment.
Details: Cogent accused Verizon of causing poor video streaming quality for Netflix content due to unbalanced peering arrangements.
Resolution: Verizon defended itself, emphasizing that the issue was not solely its responsibility.

3. Netflix vs. Comcast:
Context: Netflix, being a major content provider, generates a significant amount of internet traffic.
Issue: Comcast (an internet service provider) and Netflix had disagreements over peering arrangements.
Result: In 2014, after contentious negotiations, Netflix agreed to pay Comcast for direct access to its network, improving streaming quality for Comcast subscribers.

Traffic Prioritasation

1. Latency-Sensitive Traffic Prioritization:

Objective: Ensure smooth performance for applications like gaming and interactive video.
Challenge: Prevent high utilization by other traffic types (e.g., large file transfers) from degrading user experience.
Methods: Network operators employ techniques like weighted fair queueing and low-latency queues for interactive apps.

2. Paid Prioritization Dilemma:

Issue: Money enters the equation.
Scenario 1: Paid prioritization—where a party pays an ISP for higher traffic priority—can be seen as anti-competitive.
Scenario 2: Transit networks may prioritize their own services over competitors’, impacting overall fairness.

3. Sensitive Flash Point:

Example: AT&T blocking FaceTime video calls.
Conclusion: Prioritization discussions often involve policy, competition, and user experience considerations.

• The autonomous system (AS) value is either an internally generated number (if not connecting
  to a provider network) or obtained from an ISP or RIR.

• It is a required parameter.

• It can be any positive integer in the range from
  1 to 65535.

Enabling BGP in routers

• The ip address is the destination address of the BGP peer.

• The address must be reachable before attempting to establish the BGP relationship.

• The AS value is used to identify if the session is with internal BGP (IBGP) peers or with external BGP (EBGP)

Defining BGP Neighbours

• The BGP network command determines which networks this router advertises.

• The mask parameter indicates that BGP 4 supports subnetting and supernetting.

• If the mask is not specified, this command announces only the classful network

• It is also important to note that the prefix must exactly match (address and mask) an
  entry in the IP routing table.

Identify BGP Networks