Skip to main content

Layered protocols, headers and trailers

Short answer

Layering divides communication into manageable modules, enabling modularity, standardisation, easy troubleshooting, abstraction, and reuse. Each layer adds a header to data from above, and the peer layer removes it. The main drawback is added overhead from headers and some duplication.

What are the reasons for using layered protocols (need of networking software in the form of a hierarchy)?

A layered protocol architecture splits the complex job of communication into smaller tasks, each handled by one layer. Network software is organised as a hierarchy (stack) of layers. Each layer offers services to the layer above and hides how they are done.

Reasons for using layered protocols

  1. Divide and conquer: a big problem becomes several small, manageable ones.
  2. Modularity: a layer can be changed (for example, Ethernet replaced by Wi-Fi) without changing other layers, as long as the interface is the same.
  3. Standardisation: each layer has clear functions, so products from different vendors interoperate.
  4. Easy troubleshooting and testing: a fault can be traced to one layer.
  5. Abstraction: upper layers need not know details of lower ones (an email program does not care about cable type).
  6. Easy design and teaching: the stack is simple to understand, and different teams can develop different layers independently.
  7. Reuse: one layer, such as IP, can serve many upper protocols.

How layering works

Each layer on the sender adds its own control information (header) to the data from the upper layer. The peer layer on the receiver removes it and uses it.

 Layer n+1 data --> [H_n | data] --> layer n
 (header added going down, removed going up)

Drawback: each layer adds header overhead and some duplication of functions, which slightly reduces efficiency.

✓ Answer checked by Claude Opus 5.5 (high effort)

Related questions

All Computer Networks and Security chapter 1 questions →