0.3-0.7
- Created by: Schofe98
- Created on: 01-06-17 22:20
Data Transmission
When data is transmitted, regardless over a network or not, it is sent in the form of a packet; a data packet is split into 3 main parts: the message, header and footer.
- a packet has a max amount of data that it can transmit, defined bytheprotocolcurrentlyinuse
- Enforce andprovideservicestoenabledatatransmission and govern the rules that apply.
- A packet can be thought of as being like an envelope, with criticalinformationaboutthedestination and sender written on the outside of it.
Serial: connections always send one bit then another in order
Parallel: can send more than one bit at the same time.
Simplex: data can travel in only one direction
Half duplex: data can travel in both directions but only one at the time
Full duplex: data can travel in both directions at the same time.
Multiplexing
Data transmission speeds can be increased using multiplexing; a methods where several independent data sources are combined and sent along a single route to a single destination.
- Packet and circuit switching – method of breaking data files into small packets in order to send them across the network.
Communications Networks
Typical contents of packets: two different types of data are contained within a packet, namely the control information and actual data. The control information provides the data needed for the network for deliver the actual data.
- thesource and destination addresses
- Order number of packet
- Control signals
- Error control bits
Typically, control information is found in a packet’s header and trailer, with the actual data in between.
Network collisions – a data collision is natural occurrence on a network where two sets of data are detected simultaneously. Two transmitted packets are discarded when a collision is detected.
- A computerwaitsashort,random,timebefore attemptingtoendthediscarded packet again.
Datatransferrates:
- Bandwidth is not measured if speed as many people think, but a measure of how much data can be fitted onto a cable at any point in time.
- Bit rate is another way of expressing bandwidth and is used to describe software’s data transfer requirements.
Theinternetisaworldwidecommunicationsinfrastructure of interconnected packet-switched networks; these networks use the standard internet protocol.
Data Representation
A single bit is represented as either 1 or 0.
These bits are grouped together to represent information that is more meaningful, from small integers to complex graphics and programs.
A word is a group bits that can be manipulated as a single unit by the CPU. The size of a word usually consists of 16,32,64 or 128 bits.
- Large word sizes mean that the computer system can transfer and manipulate larger groups of data than a computer system with a smaller word size, which generally means faster operation.
Data Types:
- Character - single letter, digit, space, etc
- String - groups of characters
- Boolean - true or false
- Real - fractions or decimals
- Integeger - positive or negative whole number
Two's Compliment
Way of representing binary numbers as it allows you to use negative numbers without reducing the range of numbers.
- The easiest way to show a negative number using two’s compliment is to write it out as a positive number using the usualbinarymethod
Normalisation
The precision of floating point binary depends on the number of bits used to represent the mantissa.
Advantages and disadvantages of floating point:
- Integer formarestoredcompletelyaccurately
- There is a limited range of numbers that can be represented using the bits available to you.
- Greater range of positive and negative numbers can be stored in the same number of bits.
- Number aren’t normallystored completely accurately
- Require more complex processing
- No exact representation of zero
Truncation and Rounding
Methods for dealing with a situation where there are not enough bits to represent all of the numbers to be stored.
- Truncation cuts off the last digits
- Rounding ismoreaccurateasitisclosertotheoriginalvalue.
Organisation of Data
•Fixed length record – made up of a number of field, each one having a specific data type and size •Common datatypesincludeinteger, string, date and Boolean. •Estimatingfilesizerequiresthesizeofsinglerecordsto be calculated, multiplied by the number of records and 10% added on for metadata.
File Organisation
Files are opened in one of three modes: append, read and write. Files are read by iterating over records using end of file to end the loop
Serialfiles– new records are appended to the back of a file. All files are stored in chronological order.
- Tend tobeusedforsituationsthatdonotrequire an order or the number of records is small etc
Sequential and indexed sequential files – order records based on a primary key field.
- Insertingrecordsismorecomplexastheyhavetocopiedfrom the originally when inserting.
Random Files Random access involves jumping to a record directly without the need to search numerous records or use an index Hashingalgorithm will produce a position in a file; it must produce the same key every time and mustdoitregardless of what is already in the file. Direct accessfiles are split into blocks where each block contains a fixed number of records; hashing algorithm will produce a start position of the block and then a serial search can occur on the block to find the file.
- Deterministic – when given a key it should always produce the same result.
- Uniformity – keys should be spread evenly over the available block range with the same probability to reduce the number of records in the same block.
- Datanormalisation– keysshouldbenoramlisedbeforebeinghashed
- Continuity– keysthatdifferbyasmall amount should result in hash value that akso only differ by a small amount
- Non-invertinle – it does not hold true that the hash value can be reversed to get the original key.
Transcation and Master Files
Transaction files will store data collected over a short amount of time.
- Acts as a temporary file to help the system run quickly, but still sotring all of the required data
Master filesstoreallthedatarequiredtoperfrombatchprocessingoperations.
- Eg generatingbillsordatamining
- Will notbeusedonadaytodaybasisas they are too large
Backing Up and Archiving
Process of copying files from the main area to a separate area.
- Based on policy set up by a company
- It will consider how often the back up will occur, when it should be take, how long it will take and where it should be stored
Archiving is the process of moving old files to separate system location in order to speed up the main system
Databases
The ability to holdvast amounts of information.
Flat file – data is help within a single table; each column is an attribute and each row is a record.
- Generally usedfor small amounts of data
Relational– manysmallertables linked together
- Removes the repetition problems of flat file
- More complex to create and maintain
Primary key is a piece of information in each record that is unique to that record; often called a ‘uniqueidentifier’
Foreign key is any field in a table which is the primary key for another table.
Data Integrity
- Entity integrity – every table must have a primary key which is unique and not empty
- Referentialintegrity – foreignkeysmustpointto the primary key of another table but can be empty (meaning there is no relationship)
- Domain integrity – all attributes in the database are related to the overall domain that the database is working on.
- Normalisation is the process of converting a flat file into a relational database.
- Different stagesofnormalisation: the most common is 3NF which contains no transitive dependencies
- Indexing is a data structure used to shorten the length of time it takes to search a database
- For exampletheindexmightcontain aspecific column
Database and Big Data
Database views allow a restricted view of the entire database for performance and security reasons DMS allow users and databases to be managed
- Implement thedatadictionarywhichsotresallofthe metadataofthe database
- Databases are servers that run in the background and have no user interface
Contains thenetworkaddressoftheserver,username, password, schema name
Distributeddatabasessplitanorganisation’sdata need between multiple servers.
Data warehousing(datamining) – allows analysis of big data to uncover patterns and predict future outcomes. –Fraud detection: data mining has enabled company to spot suspicious transactions –Requires fairly advantaged technology and a large amount of data to be accurate
Related discussions on The Student Room
- probability heeeeeelp »
- What percentage/ average I need in third year to get a first. »
- What types of questions can appear with percentage calculations? »
- What will I be graduating with? »
- Studying at Pavol Jozef Šafárik University in Kosice, Slovakia »
- what is the degree weighting in cardiffmet for second and third year percentages?? »
- OCR A-level Mathematics B (MEI) Paper 2 (H640/02) - 13th June 2023 [Exam Chat] »
- Alevel Maths Hypothesis Testing »
- Chemistry question »
- Help maths a level urgent »
Comments
No comments have yet been made