Computer Networks
Lecture 1 – Exploring the network
Client server network
A computer network: are several computers connected to each other, that way they can share
resources/ information.
Client: is a computer that requires one or more services from a server
Server: is a computer that provides one or more services to other clients
P2P network: is a network with several hosts that can both act as client/ server
Network components
Types of networks
LAN (Local Area Network): network limited to a smaller geographical area
WLAN (Wireless Local Area Network): ^ but wireless
WAN (Wide Area Network): network providing access to other networks in larger geo. areas
SAN (Storage Area Network): intended for storage
MAN (Metropolitan Area Network): intended for cities (LOCATED BETWEEN WAN & LAN)
Intranet & Extranet
Intranet: private computer network from an org. etc.
Extranet: is a network that provides secure access to business data for non-company users
Internet access
SOHO users use: FTTH (Fiber To The Home), cellular, distribution cable, DSL, Satellite connection
Companies: one option is added: SDSL (Symmetric Digital Subscriber Line)
Converged networks
In the past diff. networks were used for different types of data (video, telephony, …), nowadays all
of this data is sent over one single network. In this case we talk about a ‘Converged Network’.
Network architecture
Is the whole of the physical network infrastructure and the services and protocols used in com.
4 basic characteristics
Fault tolerance: ensures the limitation of damage and quick repair in case of malfunction
Scalability: ensures that a network can expand without compromising user performance
Quality Of Service: ensures that (important) time-sensitive data is faster than low priority
Security: protect s network usage by means of hardware and software
Network Security
Good network security should ensure:
Unauthorised users can’t access the network
Information can’t be stolen
Unauthorised users can’t modify information
No DoS (Denial Of Service) attacks can occur
,Lecture 2 – NOS Configuration
Introduction
Like end devices, intermediary devices and routers are equipped with an (I)OS (operating system)
usually operated via the CLI (Command Line Interface).
Access
You can access the IOS of a Cisco device in various ways:
Locally via dedicated port (= serial connection)
o Only for device management
o Connection computer console port via special cable and use of special software
o Mainly used for initial configuration
o Can be used if there isn’t network connectivity
Done with a cable to connect a computer to the console port of a router, …
Remotely via SSH (Secure Shell)
o Possible if computer is not connected directly but via intermediary device
o Possible if the network functions work correctly
IOS modes
Basic modes
User Executive Mode
o Mode which you enter the CLI
o Limited actions! – VIEW ONLY
o Prompt sign: “name device”>
Ex. switch>
Privileged Executive Mode
o Can perform configuration and management comm. in addition to monitoring comm
o Prompt sign: “name device”#
Ex. switch#
Other modes
Global Configuration Mode
o Used to configure devices
o Only be reached from ‘privileged executive mode’
o Prompt sign: “name device” (config)#
Ex. switch(config)#
Sub Configuration Mode
o Configure specific components of a device
o Only be entered from ‘global configuration mode’
o Prompt sign: “name device” (config-if)# , …
Ex. switch(config-if)#
,Lecture 3 – Network Protocols
To communicate over a network you need:
1. Message source
2. Message destination
3. Channel = transmission medium
Communication protocol tasks
Message encoding and decoding: the information to be sent must be converted into signals that can
be transmitted via the medium (=encoding). The recipient must convert these signals back into
usable information (=decoding).
Message size: information is minced into small pieces before they are sent
Message formatting and encapsulation: before the pieces are sent, they are cast in a specific format
(frame) that provides additional address information and encapsulation; receiver does the reverse.
Message timing: necessary to agree when to communicate, but also how a too fast transmitter can
be slowed down by the recipient (flow control) and how long a transmitter must wait for an answer
to a question asked. (response time-out).
Message delivery options: it is necessary to agree to whom dispatched frames can be delivered:
through casting (to one recipient), multicasting (to several recipients) or broadcasting (to everyone).
Protocol suites
In a network, diff protocols work together to achieve a successful communication. (= protocol suites)
P.S. are implemented in hardware and software by hosts and network devices.
P.S. is hierarchical: several layers, the higher one uses services provided by lower layer.
OSI (reference model)
Is rather theoretical model that describes what network functionality is required in each layer but
does not specify how this should be implemented concretely. It offers a “framework”.
1. Application layer: process-to-process communication
2. Presentation layer: ensures that the info can be interpreted
3. Session layer: ensures synchronization of data exchange
4. Transport layer: segments, transfers, reassembles data in the comm.
5. Network layer: sends data packets over a route between 2 devices
6. Data link layer: provides physical address information
7. Physical layer: ensures the physical bit tfransmission on the comm medium
TCP/IP (protocol model)
Is a protocol model that is used on the internet and is named after TCP (Transmission Control
Protocol) & IP (Internet Protocol).
1. Application layer: sets data in readable format for the users
2. Transport layer: supports communication between diff devices through a network
3. Internet layer: determines the best path that packets can follow through a network
4. Network Access Layer: checks the hardware of the network
, Data exchange - TCP/IP (‘server wants to send something’)
At the sender side, each layer will add a header with the corresponding protocol info. The process by
which each layer of the sender adds a protocol header to the received data, before its given to lower
layer, is called data encapsulation.
At the receiver, the reverse process takes place: each layer will remove the corresponing protocol
info. The process by which each layer of the receiving system removes a protocol header of the
received data, before it’s transmitted to a higher layer is called data de-encapsulation.
Segmentation
The transport layer splits the data it receives from the application layer into smaller pieces. This is
referred to as segmentation.
+ Several communications can take place almost simultaneously, no long waiting times
+ In case of network obstruction, segments can be sent along a different path.
+ If a packet is lost, only one small segment must be re-transmitted
- More complex management: segments must be provided with additional information
Header information
the transport header contains amongst other things, the port numbers corresponding to the sending
and receiving application.
The internet header contains the logical IP addresses of the original transmitter and the final
receiver. An IP address consists of 32 bits that are represented as 4 decimal numbers between
0 & 255. It is called the dotted decimal notation ex. 192.168.0.1
The network access layer header contains the MAC addresses of the receiving and sending network
interface in the same network. An MAC address consists of 48 bits that are represented as 12
hexadecimal digits 0 – F, ex. 12-34-56-78-90-AB. For IP communication you need IP, DNS, default
gateway, subnet mask. These can be assigned diff:
Manual static IP address: usually assigned to servers, switches, routers, printers, etc.
Automatic dynamic IP address: via DHCP – usually assigned to ordinary
Communication network
If the transmitter and the receiver are in the same network, the transmitter must know the IP- &
MAC-address of itself and of the receiver
How does the transmitter know the recipient IP-address
o If used directly in the sending application
o Via DNS if the sending application uses the name
How does the transmitter know the recipient MAC-address
o Via ARP (Address Resolution Protocol)
If the transmitter and the receiver are not in the same network, they belong to networks connected
to each other via 1 or more routers.
In this case, the transmitter will send the packets to the default gateway. The router will
either deliver the packets directly to the receiver or forward it to another router if directly
connected.
The transmitter must know the MAC-address of the default gateway, (also via ARP) not the
MAC address of the receiver.