A-Level Computing Key - Terms & Concepts Cheat Sheet

by 0llieC (0llieC) via [Link]/38321/cs/12005/

Compon​ent​s/P​roc​essor Registers Compon​ent​s/P​roc​essor Registers (cont) Compon​ent​s/P​roc​essor Registers (cont)

Mother​boa The central interface for all the Arithmetic The ALU carries out arithmetic Memory Values fetched from memory are
rd components of a PC. Everything Logic Unit calcul​ations and compar​isons. Buffer sent to MBR.
connects to the mother​board via (ALU) The result of any calcul​ation is Register
slots, wires, readouts and sent to the Accumu​lator (MBR)
connec​tors.
Control The CU controls the operation of Memory The location in memory of the
Processor A combin​ation of registers than Unit (CU) the hardware, inc. input and Address current instru​cti​on/data being
(CPU) manipulate data between the output devices, it controls the Register fetched.
registers. The speed of a Fetch-​Dec​ode​-Ex​ecute cycle. (MAR)
processor is measured in the
Clock The clock is the part that regulates Current The instru​ction currently being
number of instru​ctions it can
the cycle of the CPU. It provides a Instru​ction execut​ed/​decoded
complete per second (Hz).
regular pulse of high voltage then Register
Modern computers speed is
low voltage. This high-low (CIR)
measured in GHz.
transition is a cycle, each cycle is Data Bus Carries the data between
Secondary Used to store programs and data. an instru​ction
memory and the MBR
Storage It can be partit​ioned to allow for
Program This register holds the address of Address Carries the memory location of
dual-b​ooting multiple operating Counter the next instru​ction to be
Bus the instru​cti​ons​/data being
systems. (PC) executed, the PC is automa​tically
received.
 implem​ented to the next
Control A bus with 2 states, set or enable,
Processor Structure instru​ction, unless the previous
Bus which govern if the data bus is
instru​ction was a jump.
 reading or writing to memory

Fetch Decode Execute

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 1 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Compon​ent​s/P​roc​essor Registers (cont) Compon​ent​s/P​roc​essor Registers (cont) Compon​ent​s/P​roc​essor Registers (cont)

Fetch Cache Cache is a small amount of very fast Pipelining Where the stages of the F-D-E
The PC contents are copied to the MAR Size memory. Repeatedly used cycle are 'stacked' so that they
Instruction at address in MAR  MBR instru​ctions and data is stored in the can be processed at the same
MBR  CIR cache for quick access. The bigger time. While one instru​ction is
 the cache, the more can be stored on being fetched, the previous is
Decode it thus reducing processing time being decoded. This may not
Instruction is decoded into: necess​arily increase processing
Clock The clock regulates the instru​ction
1. Operand  The data to preform an time but throughput is increased.
Speed execution rate. The faster the clock,
instru​ction on the more cycles completed per
2. Op-Code  The instruction Issues
second.
 • If an instru​ction requires the
Execute result of a previous instru​ction, the
Instruction executed CPU will remain dormant 
If data is being committed to memory, its held in leading to 'bubbl​es'​/pi​peline stalls
the MBR in the pipeline.
Cycle repeats until stop instru​ction • Jumps  lead to the pipeline
having to be flushed due to the

change in instructions
Factors affecting Processor Perfor​mance

 Hyper-Threading  Where the

CPU is intell​igent enough to fill the
Multiple The increase in number of cores,
bubbles caused by pipeline stalls
Cores allows for a greater throughput of
with other non-de​pendant
data. If the software is threaded - can
instru​ctions from separate threads.
use multiple cores - it will divide up
tasks to the different cores.
Howe​ver, it must be coded in, else
it will use the single core.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 2 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Compon​ent​s/P​roc​essor Registers (cont) Compon​ent​s/P​roc​essor Registers (cont) Compon​ent​s/P​roc​essor Registers (cont)

 Harvard The Harvard archit​ecture stores CISC CISC (Complex Instru​ction Set

Proc​essor Archit​ect​ure programs and data in sepa​rate Computing)

memory and uses the control unit at

the centre of the structure. Generally • One instru​ction can complete an
Von- The Von-Ne​umann archit​ecture is used in embedded systems. entire sequence - more complex
Neuman commonly used in most PCs. It • Higher Power Consumption
stores both programs and data in Pros • Powerful
the same memory. Using the F-D-E + Can complete an instru​ction in a • Generates more heat
cycle, it carries out a single single clock cycle (assuming RISC RISC (Reduced Instru​ction Set
instru​ction at a time. pipelining is used) Computing)
+ Modern
Pros + Can implement pipelining
• Only one value fetche​d/s​tored per
+ More Robust than Harvard (older)
instru​ction cycle
+ Single Storage Cons • Less Power Required
- Separate Memory • Used in smaller devices
Cons
 (Smartphones)
- Each Instru​ction takes 2 cycles • Generates less heat - requires less
(fetch​/decode and execute) RISC/CISC
cooling methods
- Cannot implement pipelining 


Flynn's Taxonomy

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 3 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Compon​ent​s/P​roc​essor Registers (cont) Compon​ent​s/P​roc​essor Registers (cont) Software Generation (cont)

SISD Single Instru​ction, Single Data- MIMD Multiple Instru​ctions, Multiple Data- General An applic​ation that allows the user
stream streams Purpose to produce their own solution to a
Single Core CPUs Multi-core CPUs Applic​ati problem. Most are sold as a
on package/a license.
NB: NB: Examples

• No parallelism • Multiple autonomous processors • Word Processing

• Single CU, fetches single instru​ction simult​ane​ously executing different • Desktop Publishing
instru​ctions on different data • Spreadsheets
SIMD Single Instru​ction, Multiple Data-
• Uses either one shared memory • Database Management
streams
space or a distri​buted memory space. • CAD/CAM
GPUs
• Presen​tation Software

Software Generation
NB:
Specific Examples
• One instru​ction preformed on many Specific A piece of software that has a • Microsoft Office
data-streams Applic​at specific purpose, such as order • Adobe Suite
• Naturally parall​elised operations ion entry, payroll, stock management
• Examples: Fractal Rendering, etc. It may be Besp​oke (made to
Graphics Processing (hence GPUs) - order) or Off-​the​-sh​elf (designed to
each pixel is indepe​ndant
be used in a variety of situat​ions).

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 4 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Software Generation (cont) Software Generation (cont) Software Generation (cont)

Open Open Source  Source code Translator Software that convert one Firstly the code is striped of anything uneeded
Source vs. is readable to anybody and Software progra​mming language into such as comments and redundant whitespace
Closed freely modifiable. another. There are 3 catago​ries:
Source Closed Source  Executable Compilers, Interp​reters and The code is then divided into Lexemes (the
only, source code is kept Assemb​lers. smallest 'unit' of code).
hidden.
Assemblers  Convert Tokens are then assigned to each lexeme
Pros of OS Assembly into machine specific indicating what it is. Some token examples:

• Free (usually) machine code. Assembly • Identi​fiers - for variables, subrou​tines, classes

• Community Coding/Bug Fixing language consists of mnemonics etc.

- Usually faster than any closed- that represent different • Keywords - new, if, for, while etc.
source instru​ctions. It is converted to • Operators - +, - , / , == etc.
• Customisable binary (machine code) • Literals - fixed numbers and strings
• Freedom to do what you like • Symbols - {}, (), ; etc.
Interpreters  Checks and
Pros of CS executes code line by line.
• Profes​sional Develo​pment Errors are caused when a lexeme cannot be

• Lower security risks Compiler  Checks all the assigned a token

• Well documented and codes syntax, semantics and logic Syntax Analysis
customer support and then converts the code into 
Object code (usually machine
The stream of tokens generated is then
code or similar low-le​vel). The
analysed to check they match the rules of the
compiled code is usually what is
language. Tree data structures are often used
distri​buted.
in this process.


Stages of Compil​ation An example of valid syntax would be:


String word = "​Hello, World!";
Lexical Analysis Datatype Literal Operator String​Literal Symbol


Errors occur when a series of tokens cannot

be matched to a rule, such as multiple
datatypes.

Semmantic Analysis


By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 5 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Software Generation (cont) Libraries, Linkers and Loaders Testing Strategies (cont)

The stage where code is checked for logical Bottom Testing each part of the applic​ation
errors. For example: Up indivi​dually then testing the parts
• Datatype mismatch - assigning a String to an that rely on the sectio​n/m​odule.
int
Usability Testing how easy a system is to use
• Undeclared Variables, or out of scope
(Beta) by testing it with real users. It shows
variables
how somebody without a working
• Multiple variable declarations knowledge of the applic​ation would
• Array out of bounds with an integer literal.
use the system and any problems
they might find.
Libr​ary  Generic name for a collection of
Errors occur when one of the rules are broken.
programs used in develo​pment. Some Test A range of test data must be used
NB Not all semantic errors can be caught
languages have native ones. Saves time as the Data to properly test a system. It should
during compil​ation. For example accessing an
developers don't have to create their own code. include:
array with a integer variable is logically fine, but
Linker  Combines object and library files • Normal Data
the integer value may be out of bounds causing
Loader  Loads the object code into memory • Boundary Data
a run-time error.
• Standard Incorrect Data - incorrect
to be executed
Intermediate Code Generation/Optimisation data that could easily be entered
 • Standard invalid data - e.g. text
Testing Strategies
The code is then converted into interm​ediate into numeric fields

code (Java to Java bytecode, where it remains Black Testing that examines the • Extreme data - data that would
until use - not all languages do this) Box functi​onality of a applic​ation, without never be entered normally, used to
(Alpha) looking at its internal test the limits of a system
Intermediate code is machine independent code/s​tru​ctures.

White Tests the internal struct​ure​/wo​rkings

The code is then optimised, so that it runs Box of a applic​ation rather than its
faster and requires less resources, but still functi​onality (opposite of black box)
having the same output.
Top Testing of modules and sections of
Machine Code Generation/Optimisation Down code that aren't yet implement.
 Testing the behaviour between
modules.
The final stage of compil​ation is the conversion
to machine code. This process has to be Top Testing of modules and sections of
repeated for each processor as it is machine Down code that aren't yet implement.

depend​ant. Specific optimi​sations are also Testing the behaviour between

modules.
done on the separate processors as code that
works well with one instru​ction set may not
work as well with another.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 6 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

OOP OOP (cont) Web Techno​logies (cont)

Class A 'bluep​rint', a combin​ation of Polymo​ A feature of a progra​mming JS JS is an interp​reted code that adds
attributes and methods that create rphism language that allows routines to use (JavaS​c intera​ctivity to websites. It works on
an object. variables of different types at ript) virtually all hardware and is used on
different times. For example, nearly all websites. For example
Object An instance of a class.
overlo​ading constr​uctors which there are currently 300,000 JS
Encaps​u Where attributes and methods are
behave differ​ently depending on their repos on Github, Java has 200k
lation wrapped in their objects. Access
parameters
Search A search engine searches through
modifiers control how the methods
Engine webpages, for certain keywords
and the attributes can be accessed,
Web Techno​logies Indexing and phrases. Problem Indexes are
whether that be by any class, or only
used, when a new document
within its own class. HTML The standard language for
(webpage) is added, the
Inheri​tan A relati​onship among classes where (Hypertext displaying webpages. A HTML
words/​phrases are tokenised, and
ce a child class shares Markup document starts with
added to the list.
method​s/a​ttr​ibutes with its parent Language) <!DOCTYPE html>

class. The child classes can also <html>

have their own indepe​ndent
attrib​ute​s/m​ethods but all child It consists of tags which are
nodes share the ones inherited from opened and closed <ta​g>

the parent. <​/ta​g>. Each document has a

head and a body.
Abstract A class which contains attributes
Class and methods like a normal class, but <h1>{{ml}}<h2>...
the class itself cannot be <a
instan​tiated. An example is an href="h​ttp​://​[Link]​ogl​e.c​om">Lin
Animal, you can write an abstract k Text</​a>
class, but you cannot create just an
<img scr="pa​th/​to/​ima​ge.j​pg​"
Animal.
alt="pi​ctu​re">

<p>​Normal Paragraph text</​p>

CSS The standard way to style

(Cascading webpages, whether internal or
Style extern​alised. Extern​alised
Sheet) styles​heets allow developers to
keep design and content
completely separate.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 7 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Web Techno​logies (cont) Web Techno​logies (cont) Data Structures (cont)

PageRank Google's algorithm that calculates Server​- Processing performed on the Queue A First In, First Out (FIFO) data
Algorithm the weighting of webpages. All Side server. The code is only viewable structure.
pages have an initial rank, but for Processing to people with access to the When coding a queue, there must
each link, it gives a certain server​-files. It processes be the possib​ility to:
amount to the webpage linked. requests and serves a webpage • Check if the queue is full
Other algorithms are also used to based on the requests • Read/R​emo​ve/​Return an element
give pages different rankings from the front of the Q
depending on what the user is Data Structures • Place a new element at the end of
searching for. the Q
Array A data structure used to whole
Client​-Side Processing preformed in the Circular The end of a queue linking back to
elements of the same data type.
Processing browser, usually JS. This allows Queue the beginning.
1D An Array with a single dimension, i.e.
user entered data to be checked Stack A First In, Last Out (FILO) data
Array it only has a given length
before sending it to the server,
structure.
which reduces the load on the 2D An array with 2 dimens​ions. It is When coding a stack, there must
server. For example, ensuring an Array commonly used to represent
be the possib​ility to:
email has an @, or a password is coordi​nates or a table, with the
• Check if the stack is full/empty
a certain number of charac​ters. indexes relating to rows/c​olumns
• Read/R​emo​ve/​Return an element
Anybody can view the code for
3D An array with 3 dimens​ions. It is used from the top of the stack (pop)
client​-side proces​sing, so its best
Array for repres​enting 3D space so is also • Add a new value to the top of the
for just verifi​cation
used for coordi​nates a lot. stack (push)

Linked A data structure where each element Graph A set of nodes/​ver​tices connected
List in the list points to the next one. This by edges.
makes is very easy to Direct​ion​ A graph where the edges have a
add/re​mov​e/r​eorder elements as only
al-​Graph‐ direction.
the pointer needs to change each
time.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 8 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Data Structures (cont) L.O.R. (*) Questions (cont) L.O.R. (*) Questions (cont)

Bi- A graph where the edges have 2 way Computer Level 1: Unauth​orised Access Copyright, Protects
Dir​ect​i direct​ions.  Misuse • Accessing secure parts of a Designs & indivi​dua​ls/​org​ani​sations
onal Act computer, that they are Patents Act intell​ectual data. It protects:
Graph unauth​orised to access • Income for the authors -
• In organi​sat​ions, accessing Allows the author to license the
Trees A tree is a simple un-dir​ected graph
secure parts that are beyond your data.
which contains no loops. A tree has a
rights. • Cost of creating the product -
root where all other nodes/​edges
some products can cost
originate from
Level 2: Unauth​orised Access thousands to produce
Binary A tree where each node has a
with intent to commit a Crime • Quality of Produce - pirates
Tree maximum of 2 sub-nodes. Nodes with
• Level 1 + intent to commit often alter products to bypass
no child nodes are called leafs, and
another crime. security
the edges, branches.
• Alteration Protection - Altering
Hash A table where the index system is the programs can have unintended
Level 3: Unauth​orised
Table data the person is looking for, but aftere​ffects.
Modification
Includes intent to:
L.O.R. (*) Questions • Impair the operation of any PC

Data 8 Principles: • Prevent or hinder access to a

Protection program
1. Personal data must be obtained
Act • Impair the operation of any
lawf​ully and fairly
program or reliab​ility of data.
2. P.D. must be held for a
spec​ified purpose
3. P.D. must be adeq​uate,
relevent and not excessive
4. P.D. must be kept up-t​o-d​ate
and accurate
5. P.D. must not be kept longer
than necessary
6. P.D. must be processed in
accordance with data subjects
rights
7. P.D. must be kept securely
8. P.D. must not be tran​sferred
outside the EU without
permis​sion

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 9 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

L.O.R. (*) Questions (cont) L.O.R. (*) Questions (cont) L.O.R. (*) Questions (cont)

Regulation This act allows government Automated Computers are starting to have Censorship Moral concerns are raised at
of agencies, to request access to Decision the ability to make decisions whether the internet should be
Invest​igator secure inform​ation. It makes Making based on input data. Usually it censored, would it be restri​cting
y Powers provisions for: can be better than any human the freedom of inform​ation. The
Act • Interc​eption of communication making the same decision. The issues arise when consid​ering
• Acquis​ition and disclosure of issue is what happens when the adult content, and piracy.
data wrong decision is made, who is to
Monitoring It is possible to monitor what
• Surveillance blame?
Behaviour indivi​duals are using a computer
• Access to electronic data
AI This is the one of the biggest for, there is a moral issue when
protected by encryp​tion. issues, as AI use is rising among consid​ering how much should an
 recent years. The issues are the individual be monitored, and the
same as automated decision issues based on a persons
Mora​l/E​thi​cal​/Social Issues
making, but more issues arise privacy.
Computers • Big Brother concern - an when you consider
Personal Rises a privacy concern.
in the employer could watch over
cognit​ive​/when is a computer
Inform​ation Computers can now monitor
Workplace employees considered alive?
peoples inform​ation and collect
• Reduced Produc​tivity -
Enviro​nme The increase in use of computers it. When does this become a
employees can do multiple things
nt = more RAW materials Another breach of privacy.
at once, which may reduce
issue is the disposal of old Piracy Breaking the law (C.D.P.) but
produc​tivity as employees may
parts/​dev​ices.
'waste' time people do it anyway.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 10 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

L.O.R. (*) Questions (cont) Input/​Out​put​/St​orage (cont) Input/​Out​put​/St​orage (cont)

Offensive Computers are general purpose, HDD/Ma​ Inform​ation is held in blocks SDD/Flash A storage medium that has no
Material what people do with them can be gnetic consisting of tracks and sectors. moving parts. It uses a data
considered offens​ive​/mo​rally wrong Each block contains the same controller to control the read/write
e.g. cyberb​ull​ying, which can have amount of inform​ation, therefore of data. 2 rules of the data
drastic effects inform​ation is more dense closer to controller:
the centre. 1. You can combine pages to
Input/​Out​put​/St​orage form a block, but a block cannot
Rotation Speed overwrite individual pages
Input A device (piece of computer hardware
• A HDD consists of a very fast 2. Before writing to a memory
Device equipment) that is used to provide
spinning disk (5400 - 7200 rpm) location, the page previously
data and control signals to an
• A reading head is suspended allocated must be erased.
inform​ation processing system such
above the disk due to the Bernoulli
as a computer or inform​ation
effect Pros
appliance. Examples of input
• Due to fast speeds, the housing • Low Latency Time
devices include keyboards, mouse, has to be evacuated • Fast Transfer speed
scanners, digital cameras and
joysti​cks. Capacity vs Cost Cons
Output is any device used to send data from • Largest HDD avaliable ~ 12TB. • More Expensive
Device a computer to another device or user. • Roughly 3p per GB.
Most computer data output that is (£0.00​000​000​0027915 per Byte) Capacity vs Cost
meant for humans is in the form of • Largest: ~4TB
audio or video. Examples include • Roughly 30p per GB
monitors, projec​tors, speakers, (£0.00​000​000​0291625 per Byte)
headphones and printe​rs.

Memory

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 11 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Input/​Out​put​/St​orage (cont) Input/​Out​put​/St​orage (cont) Input/​Out​put​/St​orage (cont)

Disc/O​‐ A storage medium that uses binary RAM The 'working' area of the computer. BIOS/UEFI Basic Input Output
ptical pits to encode data. A laser is Programs and data currently in use is System​/Un​ified Extensible
beamed at the disk and uses the stored in the RAM. On startup the BIOS Firmware Interface The BIOS is
diffra​ction of the light to detect a 0/1 loads the OS into the RAM. preforms the hardware
(troug​h/p​eak). initia​lis​ation during the bootup,
Characteristics and provides runtime services
Read-only: A laser is used to burn • Random Access - allows data items between the OS and hardware.
the disks, the data cannot be to be read or written in almost the UEFI was designed to be the
changed same amount of time irresp​ective of successor to the BIOS
Re-Writable: A dye is used where if the physical location of data inside the Virtual When the RAM is full, the OS
a high temp is used, it will go opaque memory. Memory uses some of the secondary
(creating a peak - 1) and if a higher • Vola​tile  emptied on power down storage as Virtual Memory. This
temp is used it goes transp​arent (a • ~1-16GB means the computer can
trough - 0). The disk is now reusable. continue to run. Pages (blocks of
ROM A permanent area of storage. The
Speeds Solid State: 200 to 2500 MB/s contents cannot be altered by data) are transf​erred to the virtual

Hard-Drive: 1030 MB/s software. Contents of ROM is written at memory when not needed thus
manufa​cture freeing up space, and returned to
Optical (x1 Speeds):
RAM when they are needed.
- Blue-ray: 4.29 MB/s
- DVD: 1.32 MB/s Characteristics

- CD: 0.15 MB/s • Read-Only Access Operating System

• Non-​Vol​atile  retains data at
 Operating Software that provides:
power down
System • Process Management
• Mainly used to store firmware or
• Memory Management
applic​ation software in plug-in
• Device Management
cartridges.
• User Interface
• ~4MB
• File Management
• Examples of ROM: Bootloader
(BIOS/​UEFI),

Fundamentally its software that

manage​s/i​nte​rfaces computer
hardware and software

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 12 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Operating System (cont) Operating System (cont) Operating System (cont)

Kernel The very core of the OS that provides Memory A OS must manage the Interrupts Interrupts are a form of error
the interface between the user and the Management computers memory including checking. If an error occurs, the
hardware. Applic​ations use the kernel adding​/re​moving programs and interrupt is stored in a priority
to send/r​eceive data from hardware. data from RAM, allowing queue. After the next instru​ction
multiple programs to be run at has been executed, the interrupt
the same time. The OS also queue is checked for any interrupt
reallo​cates memory when it is and the processor runs a set of
no-longer in use (i.e. when a instru​ctions called the Interrupt
program is closed) Service Routine (ISR), with each
interrupt having its own ISR.
Paging vs Segmen​tation & Before the ISR is run, the current
Virtual Memory values in the registers are stored,
• Segmen​tation.  Memory is so that the processor can return to
split into variable sized blocks, its previous position. Examples of
and programs are segmented, interrupt types are:
with each segment being a • I/O Interrupt  A status of a
logical divider. A segment channel has changed, Occurs
table then maps segments when an IO operation is complete
onto memory blocks. Generally or a device is ready.
slower than paging due to the • Timer Interrupt  Allows the
placement algorithm processor to preform tasks at
• Paging.  RAM is split into intervals
fixed sized blocks - frames. • Program Check  Most
Programs are split into same- commonly memory access
sized blocks - pages. Any page violations - accessing memory that
can be placed in any frame, doesn't exist or is not in use
easy to allocate as all equal • Machine Check  when
size. hardware
• If the RAM is full. Pages are
transf​erred to the secondary
storage acting as memory -
Virtual Memory. Pages are
moved in/out as needed.
• Thrashing  is when pages
are being constantly swapped
between RAM and [Link]. It
can cause speed issues as the
secondary storage's speed <<
RAM's speed.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 13 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Operating System (cont) Operating System (cont) Operating System (cont)

Process Involves the scheduling and Scheduling First Come, First Served Types Embedded
Management switching of programs and Techniques As the name suggests. of OS • Mostly hidden in devices, generally
threads. Modern PCs have • Poor Efficiency within the hardware themselves.
'multi​tas​king' but it is just clever • Built into objects
schedu​ling. Round Robin • Have a dedicated purpose
Each process has a set number • Little/no user interface
of processing time. Processor • Fully Autonomous
switches in a circular fashion • Use limited resources - only whats
• Easy Implementation required
• Can be inefficient
• Time can be lost waiting for Multi-Tasking
inputs • Several progra​ms/​pro​cesses at the
same time (concurrent).
Shortest Job First • Can either be process management
The process with the shortest or through parallel processing
processing time is processed • Most General Purpose OS' are now
• Long Process can be waiting a Multitasking
long time - processor starvation
Multi-User
Shortest time remaining • Must be a multi-​tasking OS too
The process with the shortest • Several users accessing the
remaining processing time is proces​sor​/pr​ogr​ams​/re​sources at the
processed. If another job with a same time.
shorter time remaining arrives, it • Usually a round robin approach.
will switch • Shared processing.
• Short jobs executed quickly
• Starvation can still occur Real-Time
• Inputs being processed under strict
Multi-Level Queue time limits. For requirements:
Processes are given a priority 1. Support Non-Se​que​ntial programs
when they arrive, dep. on their 2. Handle parallel and unpred​ictable
time remaining, process type and events
memory size. 3. Produce responses within the time
• Important jobs processed first limit
4. Have fail-safes to guarantee
Multi-Level Feedback Queue response time
Same as a MLQ but the
processor can change the priority Distributed
of a process, most likely due to A collection of indepe​ndent nodes,
a process taking up too much each with its own hardware. The OS
processing time. presents the systems as an
• Stops starvation indivi​dual. For example: AI; Weather
• Allows interactivity Foreca​sting; Online Shopping. Each
• Priorities can be changed may have the main system on one
server, and other things processed on
another. The pros of this are that it
reduces the load on one computer,
and if one fails, it may be able to
continue.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 14 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Operating System (cont) RAD Progra​mming paradigms

Device The OS can make devices Object Code is divided into objects which
Management acessible to other programs Orientated possess state and behaviour.
through the use of Device Follows the principles of
Drivers. It is a piece of software encaps​ula​tion, abstra​ction,
This develo​pment method​ology requires
that controls the hardware and inheri​tance and polymo​rphism.
minimal docume​nta​tion, but requires a high
provides the interface so that
amount of involv​ement of the client as a Logic The code consists of a series of
programs and the OS can use
prototype is created, then reviewed then rules which define a scenario.
the device. Devices cause
improved upon. Answers can be obtained by
interrupts on the processor and
asking questions in a specific
depending on its priority is
Extreme Progra​mming format.
when the interrupt is processed.
Data Queries to a database or other
Query data structure are specified by
Waterfall Method
Languages what is wanted rather than how
to get it.

Scripting Code is written to automate

processes rather than create
entire applic​ations. Scripting
languages are often embedded

Each of the stages are classified as into other systems.

miles​tones . Following the method​ology strictly Procedural Allows structured progra​mming

would mean the system is developed flowing with sequence, selection,
down the waterfall. Another version exists iteration and recursion. Code can
where there is iteration back up the steps. be made modular with the use of
This is one on the agile approaches to software proced​ures.
Spiral Method develo​pment. It allows for client changes
throughout the life cycle and the constant
review of progress and client involv​ement give
it its name as 'extreme'.

Method​ology Comparison

The method starts in the centre and spirals

outwards. The purpose is to elimin​ate​/reduce
any project failures by constantly returning to
each of the milest​ones. The review stage is
where the client is consulted with to determine
the progress.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 15 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Progra​mming paradigms (cont) Compre​ssion (cont) Encryption and Hashing (cont)

Functional Code is divided into isolated Dictio​ Uses a substring search to match Client​- The client generates a session
functions. There is no global nar​y- strings in the file to be compressed to Server key and uses the public key to
state, only arguments and return based those stored in a dictio​nary. If a match Commun​ encrypt it
values are important. Closely is found then the string is substi​tuted ication 
linked to mathem​atics. for the dictionary index. If no match is The server decrypts the session
found, the string is added to the key with the private key
Assembly One to one corres​pon​dence
dictionary
Languages between lines of code and 
processor instru​ctions. Unlike raw Client-Server now commun​icate
machine code however, you can Encryption and Hashing using symm​etric encryption with
have variable names and labels. the session key
Symmetric Encryption that uses the same
Encryption key to both encrypt and decrypt. Private The private key consists of 2 very
Compre​ssion Uses: Encrypted Harddrives Key large prime numbers

Lossless Compre​ssing a file without the loss Asymmetric Encryption where different keys Public The public key is the product of the
of data Encryption encrypt and decrypt the data. Key 2 prime numbers making the private

Lossy Compre​ssing a file by removing Uses: Online transa​ctions key. As no efficient non-qu​antum

redundant data. integer factor​isation algorithm

exists, it is practi​cally impossible to
Run RLE identifies repeating patterns of
crack the private key by brute force.
Length data and stores a copy of the
Encoding inform​ation and how many times it
(RLE) occurs in succes​sion.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 16 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Encryption and Hashing (cont) Databases Databases (cont)

Hashing Using an algorithm to map data of Database A structured system to hold data Secondary A key that is indexed to allow for
any size to a fixed size. Unlike Key faster searching. There can be
Relational a database structured to
encryp​tion, hashing cannot be multiple secondary keys and they
Database recognize relations between
undone, it is therefore a lossy don't have to be unique.
stored items of inform​ation.
process. Inner Join Combining columns from one+
Flat File A flat file database is a database
Uses: tables by using values common to
Database that stores data in a plain text file.
• Rapid data access in a hash table each.
Each line of the text file holds one
• Error checking and corruption
record, with fields separated by
detection - such as downloads SELECT [Link]​lumn1,
delimi​ters, such as commas or
• Password verifi​cation - the plain- table2.column2...
tabs. While it uses a simple
text password would not have to be FROM table1
structure, a flat file database
stored INNER JOIN table2
cannot contain multiple tables like
a relational database can. ON [Link]​mmo​n_field =
A good hash algorithm:*
[Link]​mmo​n_f​ield;
• Same message = Same hash Entity Any item about which data is
• Quick to compute stored e.g. Student, Pizza, Stock 

• Impossible to generate a message etc. Normal Forms

from the hash Attribute A feature of the entity 
• A small change = a big hash
Foreign A unique identifier to each record 1NF • Each row is unique - it has a
change
Key held in the relational database. primary key
• Impossible to find 2 messages with
Composite A combin​ation of 2+ fields that act • Each column has a unique
the same hash
Primary as a primary key name
Key • No columns with similar or
repeated data (i.e. choice1,
Foreign A way to build a relati​onship
choice2 etc.)
Key between 2 tables, the foreign key
• Each data item cannot be
is another tables primary key
broken up any further - no
commas in the data

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 17 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Databases (cont) Databases (cont) Databases (cont)

2NF • 1NF Atomicity Transa​ctions are either done, or Serial​isatio Create a clone of the data item,
• If the primary key is a not done. Never partially n so the user can make changes,
composite of attributes (contains applied then upload a copy of the clone
multiple columns), the non-key to the database. This will ensure
Consis​tenc Refere​ntial Integrity and other
attributes (columns) must depend that no updates or changes can
y constr​aints must be adhered to
on the whole key. be lost due to uploading a copy
Isolation Transa​ctions preformed of the local version.
3NF • 1NF
simult​ane​ously must have the
• 2NF Timestamp A non-lock way of concurrent
same result as if they were
• There are no non-key attributes Ordering access, so multiple people can
preformed sequen​tially
that depend on other non-key access the data at one time. The
attributes Durability Transa​ctions that have been main process is that the lower
committed must be done fully timestamps occur first.

and remain so.
CRUD
 Networks

Concurrent Access​ing
Standard A definition or a format that has
CREATE INSERT INTO tableName
 been approved by a recognised
(field​Names) VALUES (values)
standards organi​sation.
Concurrent Is ensuring that more than one
READ SELECT fieldNames FROM de jeur (by force of law) or de facto
Access user can at least view data at the
tableName WHERE fieldName = a standard that has just been
same time.
value ORDER BY fieldName accepted over time
Record Making a file read-only to
UPDATE UPDATE tableName SET
Locking anybody else who opens the file
fieldName = value WHERE
while changes are being made.
fieldName = value
Deadlock When 2 separate transa​ctions
DESTROY DELETE FROM tableName
lock the file the other transa​ction
where fieldName = value
needs, thus both are in a state of
waiting.
DROP tableName

ACID Principles (Trans​act​ions)

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 18 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Networks (cont) Networks (cont) Networks (cont)

Protocol An agreed​-upon format for WAN Wide Area Network Network Bus Topology
exchanging data between devices. It Topologies • All devices connected to a
determ​ines: • Geogra​phi​cally remote (across a central cable (backbone)
• The error checking used countr​y/b​etween contin​ent​s/the • Devices have equal rights
• Compre​ssion method, if any w.w.w.) • Collisions can occur if multiple
• How the sender will indicate end of • Connects LANs together with third devices send data at once
transmission party teleco​mmu​nic​ation equipment
• How the receiver will indicate the • Slower speed than LAN Star Topology
data has been received. • Uses layer 3 devices - router​s/m​ult​i- • A hub at the centre of the
LAN Local Area Network layer switches network. Requests are sent to all
other devices connected to is
• Geogra​phi​cally Small (build​ings/a • The hub reads the packets and
site) determines the MAC address of

• Equipment is generally owned by the recipient

the compan​y/p​eople using it

• Generally Faster Ring Topology
• Uses layers 1 and 2 devices - • A token is passed around the
hubs/switches ring until one of the devices
requests to use it.
• The token is filled with the
frame of data
• It is passed around the network
to each device until it reaches the
recipient
• Recipient acknow​ledges the
data has arrived.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 19 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Networks (cont) Networks (cont) Networks (cont)

Client​- • One entity (client) requests services Peer- • All computers have equal rights and Packet A messag​e/data is broken into a
Server from another (server) t​o- act as both a client and server Switching number of parts (packets) which
• Server stores security inform​ation Peer • Popular applic​ations include the are sent indepe​nde​ntly, over
e.g. logins and permissions. BitTorrent Network, and BitCoin whatever route is optimum for
each packet, and reasse​mbled at
Pros the destination.
Pros + Easy to set up
+ Centra​lised control + More reliable as central Pros
+ Single data storage depend​encies are eliminated + Efficient use of a network
+ Easy backing up and restoring + No-need for a system admini​strator + Can easily circumvent broken
+ Remote access as every user is the admin of their sections of a network
+ Can define security rights and machine + Network only has to increase

permissions + Cheaper to implement and maintain. slowly as demand does

Cons Cons Cons

- Too many requests can cause - Difficult to administor as there is no - Time taken to rebuild packets is

congestion central dependency variable - an issue for time-

- If the server fails, whole network - Less security therefore viruses and s​ens​itive data

goes down other malware can easily be - Not good for small data.

- Expensive to install and manage transmitted

- Requires profes​sionals to install and - Data recovery is difficult as there is
manage no central storage, each computer
requires ots own backup system
- "​(Lots of movies, tv shows and music
are transf​erred using P2P, via
torren​ts)​"

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 20 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Networks (cont) Networks (cont) Networks (cont)

Circuit Commun​ication where a dedicated IPv4 Most commonly used IP version. Its a The network layer is where the IP of the sender
Switching channel (or circuit) is establ​ished 32-bit system, so there are 232 is attached, so the recipient can send a
for the duration of a transmission. addresses available. message saying the packets were received. It
also attaches the recipients IP. This is also the
IPv6 IPv6 is a 128-bit address, so there are
Pros layer where the Time To Live (TTL) is added to
2128 addresses available.
+ Data arrives in order sent the header. It governs how many times the
 packet can hop before deleting itself, this
+ No additional inform​ation has to
be added - e.g. headers TCP/IP Stack ensure infinite loops don't occur.

 Link Layer
Cons 
Application Layer
- Portion of the network is
 This is the layer where the MAC address of
unavai​lable while in use
both the sender and recipient is attached,
- Data is easily interc​epted. The Applic​ation layer ensures the data is sent
allowing the packets to be directed to a specific
in an unders​tan​dable format by the recipient. It
Domain A system that converts the web NIC.
formats the data to meet the standards of the
Name address: [Link]​bsi​te.s​om​ething
protocol. 
System into the IP address of the host
Transport Layer Internet/Network Protoc​ols
(DNS) server.
 
MAC Unique 6-byte identifier that is
Address given to NICs. Assigned to the NIC The transport layer takes the data and splits it HTTP Defines how webpages are
by the manufa​cturer. into data packets. Each one is given a number, (Hypertext transf​erred from server to the
specifying the order so it can be recons​tru​cted. Transfer client. The HTTP will make a
The port number is also added depending on Protocol) request to the IP and the server
the applic​ation being used for example HTTP is responds with a webpage. There
port 80. are 8 different HTTP commands
Network Layer including GET, POST and
 CONNECT*

HTTPS Connects via a different port and

(Secure encrypts the data between the
Hypertext HTTP and the TCP protocols.
Transfer
Protocol)

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 21 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Networks (cont) Networks (cont) Networks (cont)

FTP The protocol used to download and CSMA/CA Is a transm​ission protocol that Virus A embedded program intended to
(File upload files. Most modern browsers (Carrier prevents packet collis​ions. Once cause damage to a PC. It copies
Transfer have built in FTP Sense it receives a packet, it checks itself onto the disk and hides itself. It
Protocol) Multiple whether the channel is clear, if it attempts to duplicate itself and
Access is not available it will generate a spread to other computers.
POP3 Allows emails to be received from a
with random wait time, when it will
(Post server. The protocol connects to the Worm A virus but it is contained within its
Collision check again.
Office email server, downloads a local own program.
Avoidance)
Protocol copy then deletes them from the
Trojan A non-se​lf-​rep​lic​ating virus hidden in
If a packet is larger than the
3) server.
a downloaded file, and unleashed
permitted size is needed to be
SMTP Used for sending emails on execution.
sent, a hands​hake needs to
(Simple
occur first - the RTS/CTS Ransom A trojan​/worm that encrypts data
Mail
(Request to Send/Clear to Send) Ware and then charges the owner to
Transfer decrypt it.
protocol. This protocol only
Protocol)
occurs when the packet is larger
SSH Remote​-access protocol, allows than the threshold.
(Secure secure commun​ication between a

Shell) client and server
Network Security

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 22 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Networks (cont) Networks (cont) Networks (cont)

Firewall The purpose of a firewall is to control WPA/WPA2 It requires you to enter a Router A device that forwards packets
the traffic flowing in and out of a password when accessing a from one network to another.
network. It can be hardware or network. It acts as layer of
Gateway The entrance and exit of networks.
software based, and sometimes is a protec​tion.
The main use is to connect multiple
combin​ation of both. It can be setup
 networks with different
to block individual website addresses
archit​ectures
Network Hardware
or specific computers.

Proxy f a user requests a service from the Search Engines
Server network, it is first passed to the Hub It receives a signal from a node
Overall Crawling the web with 'spiders' TF-
proxy, before the proxy server then and transmits it to all the other
Search IDF (Term Frequency - Inverse
performs the request on the behalf of nodes. It is cheap and effective
engine Document Frequency) The
the network user. If the resource is for small networks but for larger
PageRank algorithm Other factors,
banned the request can be rejected, networks causes too many
such as domain name, page age,
There is never any direct contact collisions
mobile friend​liness
between user and resource, as the
Switch A switch has a small amount of
proxy acts as a "​middle man".             
internal memory, that allows it to
generate a look-up table. When Page PageRank works by counting the
data is sent the switch finds the Rank number and quality of links to a page
approp​riate node. Unlike a hub, to determine a rough estimate of how
it does​n't send the data to all important the website is. The
the nodes, just the receiver. underlying assumption is that more
important websites are likely to
receive more links from other
websites.

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 23 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Data Types Data Types (cont) Data Types (cont)

Integer A whole number Hexade​ci 0  0000  0 Two's A improved way of showing negative
mal (Base 1  0001  1 Comp numbers. If the first bit is a 1, it is
Real/Float Float is a term is used in various
16) 2  0010  2 taken as the negative version, and all
progra​mming languages to define
3  0011  3 following numbers are added to it. If it
a variable with a fractional value.
4  0100  4 is a zero, it behaves just like a
Numbers created using a float
5  0101  5 negative number.
variable declar​ation will have
6  0110  6
digits on both sides of a decimal
7  0111  7 To change a normal +ve binary
point. This is in contrast to the
8  1000  8 number to twos comp.  Flip the
integer data type, which houses
9  1001  9 bits, and add 1 
an integer or whole number.
10  1010  A E.g.
Boolean A value with a True or False
11  1011  B 01101010 = 106
condition - can possibly use 0/1
12  1100  C 10010101 (Flip the bits)
instead
13  1101  D 10010110 = +1
Character A single keyboa​rd/​unicode 14  1110  E 10010110 = -106
character 15  1111  F
Fixed A method of showing floats. The
String A set of charac​ters, used to store Denary Normal number formats with the Point decimal is fixed so there is a set
text. (Base 10) positional numbering in powers of Binary amount of integer bits and a set
Date/Time A repres​ent​ation of time. Can be 10. number of fractional bits. The
repres​ented in either text or Sign & A form of showing negative binary fractional parts follow the same
number format Magnitude numbers where the first bit is the positional numbering (powers of 2)
sign (0 = +ve, 1= -ve). but the negative versions i.e. 2-1, 2-2,

Immedi​ately you have reduced etc.
Number Bases
the range of values as one of the
Binary Groups of bits in 1 and 0 (2 bits is reserved for the sign.
(Base 2) possible values). It uses positional
numbering but with powers of 2,
not 10 (denary numbers - normal)

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 24 of 25. [Link]
 A-Level Computing Key - Terms & Concepts Cheat Sheet
by 0llieC (0llieC) via [Link]/38321/cs/12005/

Data Types (cont) Logic Gates

Floating A method of showing binary Half Add two single bits, produces an
Point composed of 2 parts: the Adder output S, and a carry signal C. It
Binary mantissa, and the exponent. The consists of an AND gate (C) and a
mantissa is the actual number, XOR gate (S) in parallel.
and the exponent is the number of
Full Full adders are a combin​ation of 2+
units to move the floating point up
Adder half adders. Where the C of both half
or down by.
adders, connects to a OR gate
(Cout) and the sum of one connects
10101011 | 0011
as the input of another.
Mantissa | Exponent
Basically treating the first HA as
another input.
1.0101011 x 2 0011
Basic Takes a set and a reset signal. The
1.0101011 x 23
Flip- idea is that the FF stays in one state
1.0101011
Flop until the change signal is sent.
 
1010.1011 = 10.6875 D Stores the signal it receives if it is
(Data) enabled. It takes in an extra input D.
The rule for powers: Type
 Flip
 Flop
(Decrease the power, per jump
right)

Underflow When very small numbers, and

the boundary of what the computer
can store is reached. For example,
128-1 x 128 -1 requires 14 bits to
be stored.

Overflow A calcul​ation that results in a

number too large to be stored. For
example, any numbers past
x10100 on most calcul​ators

By 0llieC (0llieC) Published 20th February, 2019. Sponsored by [Link]

[Link]/0lliec/ Last updated 12th June, 2017. Measure your website readability!
Page 25 of 25. [Link]