Computer Network
Distance Vector

Distance Vector Routing Algorithm

The Distance Vector algorithm is iterative, asynchronous, and distributed.

  • Distributed: Each node receives information from its directly attached neighbors, calculates, and shares the result back.
  • Iterative: The process continues until no more new information needs to be exchanged.
  • Asynchronous: Nodes don't need to operate in sync with each other.

The Distance Vector algorithm is a dynamic algorithm, used in networks like ARPANET and RIP.

Key Concepts of Distance Vector Routing:

  1. Knowledge about the whole network: Each router shares its knowledge with the entire network.
  2. Routing only to neighbors: Routers send their knowledge only to directly connected routers.
  3. Information sharing at regular intervals: Routers share their updated information every 30 seconds with neighbors.