What is a Network?
- A network is any collection of independent computers(devices) which are connected to one another to share files and peripherals such as printers, modems, tape drives.
Every Network includes:
- At least two computer server or client work station
- Network Interface card(NIC)
- A connection medium(wired or wireless)
- Network Operating System software
Protocols:
- Protocols are the set of rules which defines how devices must communicate across a network.
Network Types:
LAN:
- Used in a small geographical area.
- High speed network
- Ethernet is a most common LAN technology
WAN:
- The network which connects multiple LAN’s
- Very large geographical area of network
- Uses services such as leased line, satellite links and data packet carrier services
MAN:
- A network that connects LAN’s across a city-wide geographical area.
VPN:
- A networking technology which help users to send and receive information securely via a public network.
Network Architecture:
- A host refers to any device connected to a network.
Basically three types:
Peer-to-peer architecture
- In this type of network there is no hierarchy, so each and every computers are equal commonly called as Peers.
- Peer-to-peer provides less administration overhead.
- It is used where no security of individual computers required.
- The network will have limited growth
Client-server architecture
- It refers to concept of sharing the work involved in processing data between the client computer and the server computer
- Used in Database and management of applications, document/network management
- Centralized File storage
- Restrictions of individual computers
Mainframe/Terminal
- The most common thin/client technology are:
- RDP – Remote Desktop Protocol – Microsoft
- ICA – Independent Computer Architecture – Citrix
To understand networking knowledge of OSI reference model plays a vital part
Open Systems Interconnection (OSI)
- OSI consists of two environments
- OSI environment – which is made up of seven layers of OSI protocols
- local system environment – which is the end computer system.
OSI facilitates the function to communicate between dissimilar or similar computer based systems. The local computer system environment has a closed operating system and performs its designed functions. Any application processes that do not require communicating with other systems to complete its tasks, will provide, the end result without any problems.
However when an application process needs to communicate with another application process located in a remote system, both systems must become open to the OSI environment
Important concepts to understand OSI Layering are:
- Each layer performs unique and specific task
- A layer only has knowledge of its immediately adjacent layers
- A layer uses services of the layer below
- A layer performs functions and provides services to the layer above
- A layer service is independent of the implementation
The Application layer, the topmost layer which is unique among the seven layers since it does not have any upper layer.
SEVEN LAYERS OF THE OSI MODEL AND THEIR FUNCTIONS
APPLICATION layer:
It provides services directly to applications. Responsible for identifying and establishing the availability of the intended partner, and required resources. It is also responsible for determining if there exist sufficient communication resources to reach the remote partner.
PRESENTATION layer:
Data encryption, decryption, compression and decompression are functions of this layer.
SESSION layer:
simplex, half-duplex and full duplex. Session is set up by connection establishment, data transfer and connection release.
TRANSPORT layer:
Segments data and also reassembles data from upper layers. Delivers data in a connection and connection less modes, commonly known as TCP and UDP
NETWORK layer:
Establishes a connection between two nodes by physical and logical addressing. Includes routing and relaying data through internetworks. This layer’s primary function is to deliver packets from the source network to the destination network.
DATA LINK layer:
Ensures hardware addressing of the device, and delivery to the correct device. Translates data messages from upper layers to frames, enabling hardware to transmit upper layer messages as a bit stream.
PHYSICAL layer:
Which transmits the bit stream and includes electrical signaling and hardware interface.
Process of Data encapsulation:
- User information is converted to data – Top 3 layers
- Data is converted to segments – Transport Layer
- Segments are converted to packets or datagrams – Network Layer
- Packets or datagrams are converted to frames – Data Link Layer
- Frames are converted to bits (1s and 0s) – Physical layer
The encapsulation type will vary depending on the underlying technology.
Common Data-link layer technologies include following:
- Ethernet – the most common LAN data-link technology
- Token Ring – almost entirely deprecated
- FDDI (Fiber Distributed Data Interface)
- 802.11 Wireless
- Frame-Relay
- ATM (Asynchronous Transfer Mode)
Ethernet:
- Ethernet cabling types
- Coaxial cable
- It is commonly used to deploy cable television to homes and businesses
- Twisted Pair cable:
- A RJ45 cable is used to connect a device to twisted-pair cable
- Fiber Optic cable:
- It used light to carry a signal
- Single mode fiber
- Multi-mode fiber
- Both carries high bandwidth over long distances, multi-mode is mainly used in Data center.
Packet switching:
Network layer of OSI plays important role in selecting the best route path to transmit packets and updates each time for the packets to be transmitted. After selectingt he best route the router switches the packets known as packet switching
Separate Protocol Routing:
A separate protocol routing is when the routing device, eg: a switch uses a routing table based on MAC address, and can accommodate only one encapsulation type
Multiprotocol Routing
Multiprotocol routing is carried out mostly by routers and similar devices because, the routing decisions are made at network layer and the routing tables are at network layer.
Classful vs Classless Routing:
- When using classful routing protocols, the subnet mask must remain consistent throughout the network.
- Classless routing protocols do send the subnet mask with their updates. Thus VLSM are allowed when using classless routing protocols.
- CLASSFUL: RIPv1 and IGRP
- CLASSLESS: RIPv2, EIGRP, OSPF
Here is a related post you maybe interested in: Routers Basics
