Distance-Vector vs Link-State Routing
Compare distance-vector and link-state routing — how RIP and OSPF differ in convergence, topology awareness, and scalability.
Expected Interview Answer
Distance-vector routing (e.g., RIP) has each router share its entire routing table with only its direct neighbors and pick paths based on hop count or a simple distance, while link-state routing (e.g., OSPF) has every router flood a description of its own links to the whole area and independently compute the shortest path using a full topology map.
In distance-vector protocols, a router does not know the full network shape — it trusts its neighbors’ summarized distance claims ("routing by rumor"), which is simple and low-overhead but converges slowly and is prone to routing loops, historically mitigated with techniques like split horizon, route poisoning, and holddown timers. In link-state protocols, every router originates a link-state advertisement describing its own directly connected links and floods it unchanged to all routers in the area, so every router ends up with an identical topology database and runs Dijkstra’s shortest-path-first algorithm independently to compute its own best routes. This gives link-state protocols faster, loop-free convergence and better scalability through hierarchical areas, at the cost of higher CPU/memory usage and more complex configuration. RIP is the classic distance-vector example; OSPF and IS-IS are the classic link-state examples; EIGRP is a hybrid that behaves more like an advanced distance-vector protocol with faster convergence via DUAL.
- Distance-vector is simple, low-overhead, easy to configure
- Link-state converges faster and avoids routing loops via full topology awareness
- Link-state scales better with hierarchical areas
- Choosing between them trades simplicity against convergence speed and scale
AI Mentor Explanation
Distance-vector routing is like a fielder only knowing “the boundary is about 40 meters that way” from what the fielder next to them said, passing along a rough distance without ever seeing the boundary rope themselves. Link-state routing is like every fielder independently studying the same complete ground map showing every rope, tree, and sightscreen, then calculating their own best path to any point. The map-based fielder reacts instantly and correctly to a moved sightscreen, while the rumor-based fielder has to wait for the update to trickle down neighbor by neighbor.
Step-by-Step Explanation
Step 1
Distance-vector exchange
Each router periodically sends its entire routing table (destination + distance) to its direct neighbors only.
Step 2
Distance-vector selection
A router picks the neighbor-advertised path with the lowest distance, without seeing the full topology.
Step 3
Link-state flooding
Each router originates a link-state advertisement about its own links and floods it unchanged to every router in the area.
Step 4
Link-state computation
Every router builds an identical topology database and independently runs Dijkstra SPF to compute its own best routes.
What Interviewer Expects
- Correctly contrasts “routing by rumor” vs full topology awareness
- Names RIP as distance-vector and OSPF/IS-IS as link-state examples
- Explains why link-state converges faster and avoids loops
- Mentions tradeoffs: simplicity/overhead vs CPU/memory and complexity
Common Mistakes
- Claiming distance-vector protocols know the full network topology
- Forgetting EIGRP is a hybrid, not a pure distance-vector protocol
- Not knowing link-state protocols use Dijkstra/SPF specifically
- Assuming distance-vector is always worse — ignoring its lower overhead for small networks
Best Answer (HR Friendly)
“Distance-vector routing is like passing directions along by word of mouth — each router only tells its immediate neighbor how far things are, without seeing the big picture. Link-state routing is like every router having the same full map and working out its own best route. The map approach reacts faster and more accurately to changes, but takes more memory and computation to maintain than the word-of-mouth approach.”
Code Example
# Distance-vector example: RIP only knows hop count from neighbors
# vtysh -c "show ip route rip"
# R>* 10.1.1.0/24 [120/2] via 192.0.2.1, eth0, 00:00:12
# Link-state example: OSPF builds a full link-state database
# vtysh -c "show ip ospf database"
# vtysh -c "show ip route ospf"
# O 10.1.1.0/24 [110/20] via 192.0.2.1, eth0, 00:00:03
# Notice OSPF’s cost (20) reflects cumulative link bandwidth,
# not just hop count like RIP’s metric (2)Follow-up Questions
- Why do distance-vector protocols converge more slowly than link-state protocols?
- What is split horizon and how does it prevent routing loops?
- How does OSPF use areas to improve scalability?
- Why is EIGRP considered a hybrid rather than pure distance-vector protocol?
MCQ Practice
1. What does a distance-vector router share with its neighbors?
Distance-vector routers periodically exchange their whole routing table (destination and distance) with direct neighbors.
2. What algorithm do link-state protocols like OSPF use to compute best paths?
Link-state protocols build a full topology database and run Dijkstra shortest-path-first to compute routes.
3. Which is a classic example of a distance-vector protocol?
RIP (Routing Information Protocol) is the classic distance-vector protocol, using hop count as its metric.
Flash Cards
Distance-vector routing basic idea? — Routers share their full routing table with direct neighbors only, without seeing full topology.
Link-state routing basic idea? — Routers flood their own link info to all routers in the area and independently compute shortest paths.
Distance-vector example protocol? — RIP.
Link-state example protocols? — OSPF and IS-IS.