4.9 communication and networking Flashcards
protocol
**a set of rules **defining how devices will communicate
what does the protocol need to define
- standards for physical connections and cabling
- bit rate or baud rate
- data format
- transmission a/synchronous
- error checking procedures
serial data transmission
data is sent one bit at a time over one communication line
serial data cables include?
the usb (universal serial bus)
parallel cables are?
a ribbon of several smaller cables used primarily for connecting internal components
parallel data transmission
several bits are sent simultaneously along separate lines, only reliable over short distances
serial vs parallel connectors
- serial connectors (USB) much smaller + cheaper than parallel connectors
- a PC has several USB ports for connecting peripherals
advantages of serial transmission over parallel transmission
requires less wires: lower cost, less difficult to manage when setting up the system
no risk of crosstalk/skew over long distances, as only one bit is transmitted at a time — less chance of errors, no limiting factor on transmission speed, no limiting factor on cable length (for skew)
skew
- bits transmitted across parallel links travel at diff speeds
- in synchronous data transmission, results in data falling out of sync with clock signal, data not read correctly
when is skew worst
- over long distances
- higher transmission speeds
- in extreme cases can lead to bits from different pulses overlapping, causing data corruption
crosstalk
- when comm lines tightly packed, signals from one line can ‘leak’ into another, causing data corruption
- worse with higher transmission speeds
synchronous transmission
- clock signal times when signals are sent
- signals, sent at regular intervals, received in same order that they were sent
- suitable for transmitting info in real-time systems
asynchronous transmission
- no clock signal
- start and stop bits used to indicate duration of byte transmission
- each byte sent separately the moment it is available, doesn’t wait for clock signal
- also has a parity bit
purpose of start bit in asynchronous data transfer
start the receiver clock;
synchronise the clock in the transmitter to the receiver clock;
purpose of stop bit in asynchronous data transfer
allows next bit to be recognised;
provides time for receiver to process the received data;
benefits/issues with asynchronous transmission
- relatively slow owing to the increased num of bits being sent
- cheap and effective form of serial transmission well suited to low-speed connections
bit rate equation
baud rate x no. of bits per signal
difference between physical and logical topology
physical: the architecture of the connections (between devices on the network);
logical: how the packets flow around a network;
physical bus topology
- connects clients to a single cable called a backbone
- terminator placed at either end of the backbone
- no need for a central hub, server connected to back bone like a client
operation of logical bus network topology
a node broadcasts data (to the entire network);
all nodes on the network receive the data;
a node examines the received data to check if it is the intended recipient;
only one node can transmit data at a time;
advantages of physical bus topology
- no central hub, reducing chance of network failure + decreasing cost of installation
- inexpensive to install as min length of cable is required
disadvantages of physical bus topology
- packets sent through shared backbone, every client on network can see packets not intended for them
- backbone used for communication by multiple clients, risk of collisions
- backbone failure = entire network failure
advantages of physical star network topology
- packets sent directly to recipient, over a cable connected only to the recipient. other clients can’t see
- easy to add/remove clients
- each cable has just one device comming over it, eliminating collisions
- failure of one cable does not affect entire network performance
disadvantages of physical star network topology
- central hub failure = all comms over the network are stopped
- expensive to install due to amount of cables
how can a physical star topology behave logically as a bus network?
use a bus transmission protocol;
use appopriate physical switching (hub transmits data to all devices);
client-server network
one or more central servers provide services to the clients on the network
features of a client-server network
resources are stored on the server;
centralised security management, including levels of access;
complex and expensive setup and maintenance;
servers accessible at all times;
centralised backups + security + software distribution + data storage;
expandability, able to deal with network growth;
features of a peer-to-peer network
resources stored on each individual device;
any device can access/share resources from any other (files can be distributed across the network;
each device can act as both client and server;
security management may be more difficult;
devices communicated directly with each other, no central server;
peer-to-peer on a WAN
- p2p configuration can also be used for file-sharing websites
- with thousands of people downloading, data passed between computer rather than from server
- p2p networks often used for torrenting, as they’re hard to close down
wi-fi
wireless networking tech providing high-speed internet and network connections
connecting to a wireless network needs
- ISP (internet service provider)
- WAP, modem and router
- device with a NIC / network interface card
wireless components
- wireless NIC
- stations (consisting of device w/ wireless NIC)
- WAP requires connection to a router, router requires a connection to a modem
advantages of a wireless adapter for a printer
easy to share printer between many devices;
can connect directly from devices with WiFi;
printer can be managed remotely;
MAC addressing
- every networked device contains a NIC
- each NIC attributed unique MAC address hard-coded in the manufacture
- a switch holds all MAC addresses for each device connected to it, uses these to direct packets of data to correct device
how can disabling SSID broadcasting increase the security of a wireless network
the SSID of the network will not be visible when trying to connect to a network;
therefore only users who know the SSID of the network can try to connect;
what is csma/ca
protocol used in wireless networks to avoid data collisions (by each station transmitting only when channel is idle)
hidden nodes
- sending device can only check if channel is idle if within its broadcast range
- a device can be out of range of the other device, but still on the network connecting them (i.e., within range of WAP)
- therefore they could both be sending data to WAP at same time, but would be unaware of the other device, meaning WAP unable to receive any data due to data collisions
bit rate
definition
the frequency at which bits can be transmitted per second
latency
definition
time delay between signal being transmitted and arriving
how can the use of a MAC address white list increase the security of a wireless network
a MAC address is unique to evert NIC;
a white list only allows those MAC addresses that have been authorised to connect to the network;
operation of a physical star topology
every device is connected to a central switch/hub;
every device sends data via the central switch/hub;
the switch sends packets of data to the intended recipient only // the hub sends every packet of data to every device;
can be either hub or switch in answer - learn both
baud rate
definition
the number of signal changes in a second
bandwidth
definition
the range of frequencies that can be transmitted across a network connection
how do WPA/WPA2 protect a network
WPA/WPA2 encrypted data reduces chance of unauthorised devices reading transmitted data;
if transmissions are intercepted, they cannot be understood by someone who does not have the key;
CSMA/CA and RTS/CTS
method - transmitting and receiving data
transmitting device checks for traffic;
if another transmission in progress, then transmitter continues to wait;
if channel detected as idle, transmitter sends RTS;
two devices could start transmitting simultaneously if they both detect there is no signal;
receiver responds with CTS signal;
if CTS not received, transmitter waits until end of transmission before resending RTS;
if CTS received, transmitter begins transmitting data;
receiver sends ACK ** after all data received**;
if no ACK received then data is resent;
majority voting during transmission
transmitter would send each bit an odd number of times;
receiver checks the bits received, and if they are not all the same, it assumes the one it received most copies of is the correct value;
advantages of majority voting over parity bits in data transmission
parity bits can only detect errors, not correct them;
majority voting can detect multiple errors;
majority voting is more efficient at detecting errors;
majority voting can detect an even number of errors;
how is it possible for the bit rate of a communications channel to be higher than its baud rate?
if more than one bit is encoded in each signal change
how will the receiver perform even parity error detection on a received byte
data transmitted in bytes : 7 data bits + 1 parity bit
if number of 1s in the byte is even, the data is received correctly;
if the number of 1s received is odd, the data has been corrupted;
what is a check digit and how is it generated
consists of a digit calculated;
from the other digits in the input sequence;
how would a parity bit be generated, using even parity?
number of 1s is counted;
if the total is even, parity bit is set to 0, otherwise it is set to 1;
the bits are XOR’d with each other;
the result is the parity bit
seemingly two different methods - only need to learn one i think
purpose of the DHCP system
to automate the configuration of hosts connecting to a (TCP/IP) network
advantages of the DHCP system
reduces the time required to configure hosts;
enables reuse of IP addresses;
avoids errors, such as duplicating IP addresses or programming incorrect subnet mask;
examples in second point are important for getting the mark
what happens when device communicates with DHCP server
host sends request to discover a DHCP server;
DHCP server offers configuration to host;
host accepts offer of configuration from DHCP server;
DHCP server confirms that configuration has been allocated to host;
how will an External Router be configured so that a web server can be accessed by computers outside the netowrk?
web server has non-routable IP address, cannot be accessed directly from outside the network. therefore, access to web server facilitated by External Router, which supports Network Address Translation (NAT) and port forwarding
traffic arriving on the HTTPS port;
must be forwarded (by the External Router) to the IP address of the web server;
how will the transport layer of the TCP/IP stack determine which application layer software of the server should deal with a received request?
the transport layer will use the port number, by adding the port number to the request;
functions of the network layer of the TCP/IP stack
adds source IP / destination IP addresses to datagrams;
determines where to send data to using destination IP address;
creates checksum for datagram;
splitting data into datagrams // reassembling data from datagrams;
why could JSON be better than XML?
more compact;
facilitates faster transmission;
uses less memory;
structure understood directly in some languages;
support for arrays;
easier for humans to write and understand;
how can multiple devices connected to the internet have the same IP address and still communicate with each other?
the computers have private/non-routable IP addresses;
NAT will be performed (so that computers can communicate on the Internet);
second point has a longer, more complex explanation: as data passes onto Internet, private IP address replaced with public IP address of router;
different ways a firewall can protect a LAN
block/allow (traffic on) specific ports;
block/allow (traffic from) specific IP addresses;
block/allow certain types of packet;
firewall maintains information about current connections and only allows packets relevant to these connections through;
act as a proxy server;
identify unusual behaviour from a host;
rules are written to specift conditions under which to block/allow;
how can a checksum be used to determine if a received packet has been changed during transmission?
checksum produced when packet transmitted;
checksum calculated from packet contents;
MOD operation used to limit magnitude of checksum;
checksum transmitted with packet;
receiving device recalculates checksum;
received and calculated checksums compared;
if checksums match, packet contents are accurate;
how is a packet routed across the Internet?
heirarchical organisation of routers, eg passed up to national router, transferred internationally and then passed back down a heirarchy;
path to take selected by each router;
route may change as a result of congestion;
possible repackaging of packet to use different protocol;
route determined using the IP address;
router decrementing “time to live” of packet;
source and destination MAC addresses changed at each router;
NAT will occur at router(s);
how can a subnet mask be used to determine whether or not a device can send a packet to another device across LAN or the Internet?
AND operation of subnet mask with device A’s IP address;
AND operation of subnet mask with device B’s IP address;
result of each AND operation is the subnet ID;
subnet IDs compared;
as they are different, then packet must be sent via Internet;
if they were the same, packet can be sent directly to device B;
(if they were the same, device B is on the same subnet)
relationship between bit rate and baud rate for a system that can transmit 4 different signals
(each different signal represents 2 bits of data)
bit rate is double the baud rate
relationship between bit rate and the bandwidth of the transmission medium
they are proportional // the greater the bandwidth, the higher the bit rate;
why has IPv6 been introduced to replace IPv4?
not enough unique addresses in IPv4;
eliminate need for NAT;
more efficient routing is possible;
improved facilities for multicasting;
automatic configuration possible without DHCP;
allows bigger packet sizes;
devices can move between locations and keep same IP address;
improved support for prioritising traffic by type;
advantages of a thin-client system over a thick-client system
clients are cheaper to purchase;
less configuration of clients is necessary;
simpler installation of software;
more secure as fewer settings can be changed;
workstations consume less power;
workstations need less maintenance;
what is thin client computing?
applications are executed on an application server
purpose of a DNS system
translates Fully Qualified Doman Names into IP addresses
how does a DNS system work
DNS stores a databse of FQDNs and corresponding IP addresses;
individual mappings are only known by some DNS servers;
DNS servers organised into a heirarchy;
if one DNS server cannot resolve a lookup, the query will be passed to another DNS server;
hardware requirements for a thin-client system
(all opposite for thick-client system!)
higher bandwidth network connection required;
client: slower processor needed;
client: reduced RAM needed;
client: no secondary storage required in workstations;
server: multiple processors needed;
server: a lot of RAM needed;
server: many secondary storage drives needed;
