A computer network includes software and hardware that can transmit and get data from one device to another. The purpose of hardware is to determine the physical material required to communicate and get data. Simultaneously, software represents the collection of information that uses the hardware material for data transmission.
Simple communication of data includes different phases at several layers of the computer network. In computer network models, we will explain the models to learn how the data is moved and taken at a computer level.
In a layered architecture, a computer network is partitioned into interconnected end systems using a subnetwork. The communication process is decomposed into the hierarchical functional layer, and hence a computer network has a layered architecture. There can be N functional layers in a computer network end system. In general, each layer has an active element that can be a piece of hardware or software to carry out each layer’s various functions. It is known as a layered entity.
In a layered architecture, two types of communications take place −
Hierarchical communication is the communication between layers of an end system for requesting and receiving services from the lower layer, while peer to peer communication works between peer layers of two end systems for carrying out an assigned set of functions. The service interface definition specifies the procedures and rules for hierarchical communication, while peer-to-peer communication is called the protocol. The message exchange between adjacent layers during hierarchical communication is called Interface control information, while the message exchanged between the peer layers are called protocol control information.
A computer network’s layered architecture is shown in the figure, and hierarchical & peer to peer communication.
For hierarchical communication, direct paths are available between two layers of a subsystem. But for peer to peer communication, direct paths are not available between peer layers. Hence protocol control information is exchanged using the services provided by the lower layers.
Below we will see, why we need Layered architecture −
Divide-and-conquer technique creates a design phase so that the unmanageable functions are broken into small and manageable functions. This technique can decrease the difficulty of the design.
Layered design is more modular. Modularity supports the independence of layers, which is simpler to understand and execute.
Easy to modify
It provides the layers of independence to change execution in one layer without moving different layers.
Easy to test
Every layer of the layered mechanism can be evaluated and checked individually.