27:08
Lecture 12: Routing protocols (Part II)
NPTEL IIT Kharagpur
Overview
This lecture continues the discussion on routing protocols, focusing on the distinction between interior and exterior routing protocols. It introduces the concept of Autonomous Systems (AS) as a way to group networks under a single administrative domain. The lecture then delves into two interior gateway protocols (IGPs): Routing Information Protocol (RIP) and Open Shortest Path First (OSPF). RIP is explained as a distance-vector protocol with limitations in convergence speed and bandwidth usage, while OSPF is presented as a more efficient link-state protocol that uses Dijkstra's algorithm to find the shortest path. The lecture concludes by briefly mentioning the header format and packet types used in OSPF.
How was this?
Save this permanently with flashcards, quizzes, and AI chat
Chapters
- Routing protocols are categorized as interior (within an Autonomous System) or exterior (between Autonomous Systems).
- An Autonomous System (AS) is a collection of networks managed by a single entity, identified by a unique AS number.
- Interior protocols (like RIP and OSPF) manage routing within an AS, while exterior protocols (like BGP) manage routing between ASs.
- The AS graph is typically connected, with routers within an AS using common routing protocols and routers between ASs using exterior protocols.
Understanding the distinction between interior and exterior protocols, and the concept of Autonomous Systems, is crucial for comprehending how large-scale networks like the internet are organized and how routing information is exchanged across different administrative domains.
An organization with multiple departments, each managing its own networks, can be considered a single Autonomous System if all are under one administrative control. Routers within this organization use interior protocols, while communication with an external ISP would use an exterior protocol.
- RIP is an interior routing protocol that uses a distance-vector approach to update routing tables.
- It exchanges information about the 'distance' (number of hops) to reach other routers.
- RIP suffers from slow convergence, meaning it takes a long time to update routing tables after network changes (like link failures).
- A significant drawback is the 'counting to infinity' problem, where incorrect routing information can propagate slowly, consuming bandwidth.
RIP illustrates a foundational, though now largely outdated, approach to routing. Understanding its limitations helps appreciate why more advanced protocols were developed to address issues like slow convergence and inefficient bandwidth usage.
If a link fails, RIP might take many updates to propagate this information across all routers, during which time some routers might still try to send traffic through the non-existent link, leading to delays or packet loss.
- OSPF is a widely used interior routing protocol that employs a link-state approach.
- Instead of hop counts, it exchanges information about the status and cost (e.g., delay, bandwidth) of network links.
- Each router builds a complete map (database) of the AS topology and uses Dijkstra's algorithm to calculate the shortest path to all destinations.
- OSPF uses 'Hello' packets to maintain neighbor adjacencies and Link State Advertisements (LSAs) to share network topology changes.
OSPF represents a significant improvement over RIP by offering faster convergence and more efficient use of network resources, making it a more suitable protocol for modern, complex networks.
When a link's delay increases, OSPF routers exchange this information, recalculate the shortest paths based on the new costs, and update their routing tables quickly, ensuring traffic is rerouted efficiently.
- OSPF routers exchange various packet types, including Hello, Database Description, Link State Request, Link State Advertisement, and Link State Acknowledgment.
- Hello packets are sent periodically to verify neighbor reachability.
- LSAs describe the state of links and are flooded throughout the AS to build a consistent network map.
- OSPF uses authentication to ensure the integrity of routing information and prevent malicious updates.
Understanding OSPF's packet types and operational mechanisms reveals how it achieves efficient and reliable routing by maintaining a consistent view of the network topology and ensuring the security of routing updates.
If a router detects a link failure, it sends out a Link State Advertisement (LSA) detailing this change. Other routers receive this LSA, update their network map, and recalculate their shortest paths, then send an acknowledgment to confirm receipt.
Key takeaways
- Autonomous Systems provide a framework for organizing large networks, enabling distinct routing policies within and between them.
- Interior Gateway Protocols (IGPs) manage routing within an AS, while Exterior Gateway Protocols (EGPs) manage routing between ASs.
- RIP's distance-vector approach is simple but prone to slow convergence and 'counting to infinity' issues.
- OSPF's link-state approach provides faster convergence and more efficient routing by maintaining a complete network map and using shortest path algorithms.
- The choice of routing protocol depends on the network's size, complexity, and administrative requirements.
- Network protocols rely on specific packet types and exchange mechanisms to maintain accurate and up-to-date routing information.
- Security is a critical consideration in routing protocols to prevent manipulation of routing tables and ensure network stability.
Key terms
Routing ProtocolAutonomous System (AS)Interior Gateway Protocol (IGP)Exterior Gateway Protocol (EGP)Routing Information Protocol (RIP)Distance-Vector RoutingOpen Shortest Path First (OSPF)Link-State RoutingDijkstra's AlgorithmLink State Advertisement (LSA)ConvergenceHop Count
Test your understanding
- What is the primary difference between interior and exterior routing protocols, and how do Autonomous Systems relate to this distinction?
- How does RIP's distance-vector approach differ from OSPF's link-state approach, and what are the implications for network performance?
- Why is OSPF considered more efficient than RIP, and what mechanisms does it use to achieve this efficiency?
- What is the role of Dijkstra's algorithm in OSPF, and how does it contribute to finding the shortest path?
- Explain the purpose of 'Hello' packets and Link State Advertisements (LSAs) in the OSPF protocol.