•As PC became more common, it became obvious one would need
to share data and peripherals between them efficiently.
•Floppy discs could not be the solution. Hence the idea of a
network.
•The necessities to have one database that many users can
access simultaneously made networking very relevant.
•Great demand to share devices ( e.g. printer sharing in companies).
How?
Basically a series of PC linked together via a cable system that
would transport the data flow.
Questions:
•How can access to data and peripherals be controlled?
•What kind of wire? How long can the cable be? What type of
connectors?
•How will each computer be identified? If two or more computers
want to talk at the same time, how do you ensure all conversations
are understood?
Network size: LAN, WAN
There are two different types of networks:
Topology
The term Topology is used to describe different configurations of cabling between computers.
Basically , there are two types of topologies:
the bus and the ring topologies.
Bus topology: a single cable conveys all the data, and the PCs are
connected to the cable via T-connectors.
Every network card reads the packet that is in transit in the cable.
The bus requires the use of terminators, special connections
that prevent the reverberation of the data in the cable.
Ring Topologies:
Using something called token passing, where every PC waits for the
token carrier to come round to start talking. This prevents data collision.
We can reduce the ring or the bus to a small device, called a “hub”.
Star-Bus Technology:
Today the most popular technology used in networking is the star-bus technology, all packed tight into the hub.
To connect more than two computers using Ethernet, you will need a hub like this.
Connections
Cables for Ethernet come in different flavours, mainly a speed and thickness factor.
The main type is 10BASE-T cabling. But you find now 100BASE-T.
The length between PC and Hub must not exceed 100 meters.
The connections used to link the PC to the network are usually BNC and RJ 45, the latter being the most commonly used.
IEEE 802.3
In the 70’s, manufacturers like Xerox, Intel developed a series of standard for the future computer networks. This standard is called Ethernet.
The control of the standard was given to the
The name of the Ethernet standard is now IEEE 802.3
Main Types of Network Organisation
To function a network must use a program called a network operating
system. The network operating system controls the sharing of resources
and data across a network and involves a lot of security measures.
Client/ Server:
a powerful computer is dedicated to serve other PCs with data and
resources. This computer does not run on Windows but on Novell
NetWare. It has no programs like Excel or CorelDraw. It only provides
data access to the other workstations or clients.
Peer-to-Peer:
some network operating systems like windows 2000 or Windows XP do
not require a dedicated server. A simple PC can act as both a client and
a server. Much cheaper, but cannot grow much further than 20/30 PCs.
Security is poor on Windows 9x.
Domain- based:
many servers co-exist on a network, one of them keeps the security
database of all systems. Windows NT, 2000 and XP will allow such a
centralisation via an account called “administrator”.
Protocols
In networking, the term protocol refers to a set of rules that
govern communications. Protocols are to computers what language
is to humans. Since this presentation is in English, to understand
it you must be able to read English. Similarly, for two devices on
a network to successfully communicate, they must both
understand the same protocols.
Protocols
In networking, the term protocol refers to a set of rules that govern communications. Protocols are to computers what language is to humans. Since this presentation is in English, to understand it you must be able to read English. Similarly, for two devices on a network to successfully communicate, they must both understand the same protocols.
Protocols are also typically in charge of sending the data to the right PC and make sure the data gets to the right program, and organise the data to be sent.
The address of the network cards, as well as the names and addresses of the PCs themselves are all checked and counter-checked via the protocols.
Protocol Types
There are 3 main network protocols most commonly used with Windows:
NetBEUI: (NetBIOS Extended User Interface)
A user friendly protocol provided by Microsoft but limited to this
environment. Only used within a small environment.
IPX/SPX: (Internetwork Packet Exchange/Sequenced Packet
Exchange)
A series of protocols used in Novell network. Works well with routers. Takes little RAM.
TCP/IP: (Transmission Control Protocol/Internet Protocol)
A combination of industrial standard protocols that allows to
communicate in various environments. It provides a protocol which
manages routable company networks and access to the internet.
Note: The AppleTalk protocol is for use only with Apple computers on a network.
How do I check which protocol is used by the network?
Exercise: change your protocol from NetBEUI to TCP/IP
1. In Network, click on Add
2. Select “Protocol”
3. Select “Microsoft”
4. Select TCP/IP.
5. Click “Next”
6. Click OK
Your network was using the very simple NetBEUI Protocol, you
have added the TCP/IP protocol to the database.
What is TCP/IP?
TCP/IP is the only protocol we support.
It is also the most commonly used by our prospective
customers.
TCP or transmission control protocol, is a set of rules explaining
how the data is transmitted within the cables.
Data moves from one PC to the other in discrete packets
called Frames.
When a frame is sent, TCP software on the PC checks who is
the sender, who it is addressed to, what data it contains, and
also if the info is coherent.
Data
Every single network card (NIC) that sends or receives frames must have a specific binary address called MAC address!
A frame looks like this:
On a TCP/IP network, computers have no names. Instead they have
four sets of eight binary octets, separated by a period.
Ex: 11011101 . 11111000 . 10101110 . 00000101
In decimal language it translate as this :
00000000 = 0
00000001 = 1
00000010 = 2
…
11111111 = 255
So in theory, an IP address can vary from 0 . 0 . 0. 0 to 255.255.255.255
NOTE: IP address may vary in time for one device. The MAC address, or
hardware address, is “branded” on the device by the manufacturer. IP
addresses and MAC addresses go together to ensure proper identification
of a device.
Where do I see the IP address of my PC?
The TCP/IP Properties Screen
The IP address
An IP address is in fact divided in two portions:
The network ID (shared by every device on the network)
The Host ID ( unique to every device)
For instance : 134. 56 . 33 . 141
134. 56 . 145 . 36
could be the IP addresses of two devices on the same network,
whose Network ID is then : 134. 56
How do we choose a network ID?
There are public and private network IDs, depending whether you wish to have a network open to the web, or a private network within a building (intranet, for instance)
There are three classes of network, defined by their size and their
IP addresses:
Class A network : Network . Host. Host. Host
Class B network : Network. Network. Host. Host
Class C network : Network. Network.Network. Host
All PCs on a network share the same network ID.
The three classes of network
The first octet of the IP address is a give-away for the type of network you’re facing:
It seems like a lot of addresses are available for all the PCs in the
world. In reality some figures are forbidden or “booked out”.
For instance 256 is not to be used. And IBM has booked the use of
the 1.x.x.x IP addresses for instance.
With the explosion of PC usage, there is a strain on the number of IP addresses available. The necessity to share some IP addresses lead to the creation of a method called DHCP.
DHCP (Dynamic Host Configuration Protocol)
The biggest task in a large network is usually to affect and update IP addresses. This protocol is there to help you in this task, it offers a dynamic configuration of IP addresses and of the information associated to them.
The administrator controls the mode to assign those IP addresses, he can specify the time frame during which each computer keeps its IP address before requesting a new one again from the server.
This protocol offers a reliable and easy TCP/IP configuration which avoids address conflicts and allows the administrator to control the use of the IP addresses.
So again, if you are using a DHCP protocol, then, your computer will
not have a definitive or static IP address. Instead your server will
allocate you an dynamic address that it chooses from a pool of available addresses.
Finding your IP address in a DHCP environment
Subnet Mask
So when a frame reaches your network, the protocol looks for the
address in it, and tries to match it with the addresses existing on the network PCs.
But if the network had to query all the numbers from 0 . 0 . 0 . 0 to 255.255.255.255, it could be a long search till it gets to the right address.
Again, the address is divided in two portions: the network portion, shared by every PC on the network, and the Host portion which differs from host to host.
The Subnet Mask role is to differentiate between the Network part and the Host part in the IP address. It defines the range of IP address devices can have on a single physical segment of a network.
Default Subnet Masks
The Subnet mask defines the range of IP addresses available on a physical segment of the network.
All devices on this segment will share the same Network ID and the same Subnet mask.
There are Default Subnet Masks:
Class A network: 255.0.0.0
Class B network: 255.255.0.0
Class C network: 255.255.255.0
One can customise the subnet mask to divide a network in many small portion. But this is beyond the scope of this class.
Subnet Mask
Imagine a network with 100 PCs, set up by a company, you would need 100 IP addresses.
The company could register the following range of IP addresses :
from 125.125.125. 0
to 125.125.125.100
So, if you’re looking for a PC on this network, you need to only review the last digits of the IP address.
If we set the Subnet mask for your network to 255.255.255.0, because there are no number above 255, the only search plausible is on the last digits.
Subnet Mask
The search has been restricted to the existing PCs from . 0 to . 100!
Note: a subnet only applies to one segment of the overall network. (a floor, a building…)
Other functions of the TCP/IP Protocol
TCP/IP offers many in-built services.
The most famous is called HyperText Transfer Protocol (HTTP) and is the language of the Internet.
An another one is the PING function, which you access via a DOS prompt: it is used to check if you can access a PC , by sending a frame to its IP address:
Since our printer will also have an IP address you will be able to “ping” it as well. But because it is a printer, Lexmark has provided a special command: LEXPING, that you use the same way.
A typical Ping Screen:
Basic Network Troubleshooting
• Which network protocol? Any other protocols than TCP/IP won’t work.
• Are they using DHCP? If so, the network administrator can set up a fixed IP address for the printer.
• Do you see the other computers on the network? One important tool is the ping command. It is used to see if there is communication in the network. Go to start, run, then type ping + IP address (ex: ping 155.155.155.1) you can also use the name of the computer. Lexmark created their own ping command as the printer is not ping compatible so lexping + IP address (ex: lexping 155.155.155.5). If the command ping doesn’t work: I would say 70% of network problems are due to cable so make sure they verify their connectivity.
No comments:
Post a Comment