Networking For VCE IT By Mark Kelly Manager, Information Systems McKinnon Secondary College Lecture notes: Vceit.com
Networking Year 11 IT – all of the topics can be covered at a relatively simple level ITA – topologies and protocols are not in the study design and cannot be directly assessed Software Development – all topics are relevant Note – several complex issues have been grossly simplified in this presentation. For example, detailed differences between switches and routers at the transport level are not necessary in any VCE course and will only serve to frighten sensitive souls.
What is a network – should I panic? At its simplest, a network is two or more computers that are connected so they can exchange information and share resources.
Networks can be classified by: Types of networks (LAN, WAN, Internet); (Client-Server, P2P);
Internet music sharing networks (e.g. Kazaa) are P2P - no central computer; data and software on users’ computers.
P2P built into Win, Linux, Mac
Protocols Communication protocols are agreed sets of rules and procedures for computers to exchange information. Like humans agreeing to speak the same language during a conversation. For two computers to exchange data, they must be using the same protocols.
during a phone call, saying “Uh huh”, “Mmmm” or “Yeah” while the other person speaks
nodding to show understanding
waiting for the other person to stop talking before you start
raising pitch of voice after a question
airline pilots speak English, refer to heights in feet, agree on which direction to turn to avoid collision, pronounce 9 as “niner”, spell out letters with words (Alpha, Bravo, Charlie etc.)
Network Protocols There is a standard protocol for each network communication task, such as: - how to send data over the Internet (TCP/IP) - how to send and receive email (POP, IMAP) - how to request and deliver web pages (HTTP) - how to request and deliver files (FTP)
Choosing Protocols Sometimes there is more than one choice of protocol for a task, such as how messages pass across a network (IPX/SPX vs TCP/IP, POP vs IMAP). As long as all the connected computers use the same protocol, it really does not matter which protocol is used (like diplomats agreeing on a language for negotiations) The internet only works because TCP/IP, POP, FTP and HTTP are universal standards, used by all shapes and sizes of computers.
The King of Protocols – TCP/IP The universal protocol for internet communications. The backbone of the internet. Made up of 2 complementary protocols… TCP (Transport Control Protocol) and IP (Internet Protocol)
Once a file has been chopped into packets, the IP protocol delivers each packet to its destination.
each packet can take a different route from A to B, bouncing from router to router getting more precise with each hop.
the route is dynamically chosen for each packet, based on internet conditions at that time.
Protocols – TCP/IP TCP again… At the packets’ destination the receiving computer’s TCP re-assembles packets back into the original file. Recalculates checksum to see if packet is OK If packets are damaged, lost or delayed in transit, TCP will request the server to send the packet again.
Packet Switching Any protocol that breaks files into packets (like TCP/IP does) is called packet switching . (Compare with circuit switching used by telephones where a full-time path is set up for the duration of the communication)
Like telephones, every node on a network must have a unique identifier so the file server knows who is requesting information, and who is to be sent information.
This unique network address is hardwired into the network card of each computer.
Also, every active node of the internet needs a unique identifying address so TCP/IP knows where packets are to be sent.
This is an I nternet P rotocol, or IP address.
Internet Addressing Humans like working with names (e.g. www.microsoft.com) but computers use IP numbers (e.g. 10.77.91.19). IP address has four ‘octets’ separated by dots, each octet can be between 0 and 255. Remember - all internet communications use IP addresses, not URLs. Only humans use URLs.
Addressing Domain name servers (DNS) – a distributed database on thousands of computers across the world - convert URLs into IP addresses. Like a phone book – look up a name (URL) to get a number (IP address).
A “Networking Technology” defines how packets are handled and what the hardware is like.
The only networking technology worth knowing is Ethernet
Used everywhere by everyone (except a few odd people who aren’t worth worrying about)
Uses coaxial, UTP, fibre-optic cable, and wireless.
Ethernet defines both protocols (CSMA/CD) and cabling (e.g. UTP, thick coax, fibre), speeds etc.
Ethernet – 10Base-huh? 10Base-T = twisted pair (e.g. UTP). 10Mbps , max length 100m, RJ45 connectors. 10Base2 = uses thin coaxial (RJ58) cable – max length 195m. 10 Mbps. BNC connectors. 10Base5 = uses thick coaxial – max length 500m. Used mainly for backbones, cable TV. 10Base-F – fibre optic cable on 10Mbps networks – can get up to 2,000 megabits/sec (2Gbps) on the right network. 10Base-35 – broadband coaxial cable. Max length 3,600m.
How Ethernet Works Network devices compete for attention using C arrier Sense Multiple Access with Collision Detection ( CSMA/CD ). Keep in mind: Only one signal can travel down a cable at a time. CS = Carrier Sense . Before transmitting over the network, a computer first "listens" and waits until there is no activity on the cable. When it sees its chance, it transmits.
MA = Multiple Access . When one Ethernet station transmits, all the stations on the cable hear the transmission
CD = Collision Detection . Carrier sense does not guarantee that two devices will not sense the same silence and transmit simultaneously, and cause a ‘collision’. CD detects this event.
Each node involved in the collision waits a random number of milliseconds, then repeats the transmission attempt.
The random waiting time prevents endless further collisions.
A ‘node’ is any device attached to a network that is capable of requesting and sending packets (e.g. Usually a PC, network printer) When a node wants to communicate to another node, it transmits its addressed packet. The packet travels to every node on the segment . Each node inspects the packet to see if it is addressed to him. If not, the node ignores the packet. If so, the node opens the packet and reads its contents.
Huh? Network segments? A network segment is a self-contained section of a network bounded by a bridge, router, or switch. Using segments reduces network congestion. Like classrooms in a school.
Server (e.g. file server, proxy, DHCP, web servers)
Switches (rarely, hubs)
Routers – now SOHO (Small Office / Home Office) combined boxes contain an ADSL modem, a router, wireless access point (WAP), switch, print server, coffee maker etc)
The modem Modulator/demodulator Transmission speed is measured in bits per second ( not bytes per second!) 56Kbps modem downloads at a theoretical maximum of approx 56,000 bits per second (about 7KB/sec). Can only transmit (upload) at 33.6kbps. Modulate = turn digital data into analogue sound for transmission over phone network. (when uploading) Demodulate (when downloading) = convert sound back to digital data.
Expensive adaptors to convert digital <> electrical signals
Connections – Cables and wireless Many fibre optic cable (‘FOC’) threads can be bound into a slim, single cable without their signals interfering with each other, giving massive data throughput. FOC is replacing old, heavy, expensive copper cables to cross oceans Warning! Sharks can damage your network! Sharks get over-excited by the electromagnetic fields radiated by copper cable. FO is silent.
CPU Processing power – not very important in a file server
Backup – most servers have inbuilt high-capacity tape backup drives to protect against data loss. Tape drives usually use QIC (Quarter Inch Cartridge) DAT (Digital audio tapes) tapes.
Servers are the muscle men in the computer world
File Servers vs Desktops 2 Connectivity – servers often have two or more gigabit NICs to increase their data-throughput. Robustness - servers run all day for years, and need rugged high-quality components Scalability –the ability to increase the size and power of equipment and networks as required e.g. add 8 hard disks, two power supplies, two NICs, two CPUs, lots of RAM etc. Designing and engineering this expandability is expensive.
R.A.I.D. Redundant Array of Independent Disks ) arrays for reliability and/or speed. RAID uses a group of hard disks that work as a single disk under a RAID controller. Flavours of RAID: RAID0 to RAID10 (RAID 1 + RAID 0) offer reliability and/or speed (at ever-increasing cost). Includes mirroring (for reliability) and striping (spanning a logical single volume over several physical disks for greater performance – several simultaneous disk reads/writes are possible). RAID disks are usually "Hot Swap“ – no server downtime to replace sick disks. EXPENSIVE – needs justifying for small org 3-disk RAID array
central connection point (a switch) with cables branching to many computers.
Not a server with 4 NICs!!!
If a cable fails, only one node will fail.
prone to traffic bottlenecks at the centre of the star
RECOMMEND THIS for small networks!
Tree Topology* Combines bus and star topologies. It looks like a tree. Very common in larger networks. *not examinable e.g. one cable from a file server leads to a 24 port switch. Many cables branch from this switch to the computers in the computer room. They share the bandwidth of the incoming cable.
Mesh Topology* Multiple routes from one node to any other. Hardly ever found cabled in real-life in real networks: really only appears in The Internet to give near-perfect reliability. DO NOT RECOMMEND IT IN A TYPICAL CASE STUDY *not examinable
File server failure can severely affect network users.
Locked in air-conditioned, alarmed room with barred windows, restricted keys
No user access to server
Uninterruptible power supply (UPS) protects against blackouts, brownouts and voltage spikes.
Accessible fire fighting equipment.
Locked floppy disk drives
Network Electronic Security Passwords are not strong protection – they can be guessed, forgotten or stolen.
Network Security Daily backups are vital. Massive cost and effort to recover a single megabyte of lost data. Organisations need a data disaster recovery plan so they know what to do to recover from catastrophic data loss .
Makes data unreadable to unauthorised people even if a file is stolen.
Web browsers use encryption to connect to a “Secure” SSL (Secure Socket Layers) site.
Network Security TROJAN HORSES attempting to report ‘home’ or start a DOS/DDOS attack - can be blocked by a firewall. FIREWALLS in hardware (routers) or software (e.g. Zone Alarm) check for unauthorised incoming or outgoing network traffic, e.g. port scanning, being enslaved to help with a distributed denial-of-service (DDOS) or spam attacks. VIRUSES can disclose user passwords, steal information, destroy data, install “back doors” to let hackers in, clog print queues, disrupt Internet traffic, overload email servers etc. Keep scanners up to date.