Databases and distributed systems

Question 1

what is a relational database?

Answer 1

multiple tables that contain related data that is linked using relationships

Question 2

relational database pros
(name 3)

Answer 2

less data redundancy (repeated data less common)
data consistency (fewer contradictory entries)
better security (can restrict viewing access)
better flexibility (easy to add new table)
data independence (data can be used for multiple things without customisation)

Question 3

relational database cons

Answer 3

greater complexity
can take longer to retrieve data if it is across multiple tables
can be overkill for simple sets of data

Question 4

primary key

Answer 4

unique identifier in a table

Question 5

foreign key

Answer 5

primary key from one table within another to link tables together

Question 6

data consistency

Answer 6

ensure that the constraints or limitations placed on the database area adhered to

Question 7

data reduncancy

Answer 7

duplication of data, data held in multiple locations
harder to maintain accuracy if the data changes

Question 8

data integrity

Answer 8

correctness of the data over time

Question 9

validation checks (name 4)

Answer 9

type check (correct data type)
length check
range check
format check (conforms to set of rules)
presence check
check digit

Question 10

verification check

Answer 10

double entry (enter data twice to ensure accuracy)
proof reading (read over input to manually check)

Question 11

field

Answer 11

one category of data in a table (a column)

Question 12

relationship

Answer 12

a link between two tables
can have one-to-many
can’t have one-to-one (except passwords) or many-to-many

Question 13

normalisation

Answer 13

the process of converting a flat file database into a relational database

Question 14

1NF

Answer 14

each field only contains one piece of data
all attributes dependent on the primary key
no data redundancies

Question 15

2NF

Answer 15

1NF + no partial dependencies

Question 16

3NF

Answer 16

2NF + no transitive dependencies

Question 17

what is a database management system (DBMS)?

Answer 17

piece of software that sits between user and the data
allows the user to manipulate the data and different users’ access to it

Question 18

name 4 DBMS functions

Answer 18

checks data for inconsistencies
provides security (access rights)
backing up data
performs calculations on data retrieved from DB
management of data dictionaries

Question 19

big data

Answer 19

set of data that is too large to be processed using traditional methods
can be classed as big data when:
-volume of data such that it can’t fit on single server
- rate at which new data arrives is very high
- data of different types to point it can’t be put into table/series of tables

Question 20

methods of working with big data

Answer 20

data warehousing
data mining
predictive analysis

Question 21

data warehousing

Answer 21

storing large, complex and not directly related data items in a way to make it readily retrievable and understandable

Question 22

data mining

Answer 22

process of retrieving data from large data sets with a view to identifying patterns

Question 23

predictive analysis

Answer 23

using techniques performed on big data to predict what might happen in the future

Question 24

distributed system

Answer 24

system in which both processing and storage functions are spread across multiple geographical locations

Question 25

distributed system pros (name 3)

Answer 25

• performance more consistent as no system-wide bottlenecks
• adding new node to system is easy as system already made of nodes
• errors more likely to be contained to a node
• even if one node fails the bulk of the system still works

Question 26

distributed system cons (name 3)

Answer 26

-without proper planning data transmission can be inefficient which will waste bandwidth
- more difficult to prevent data duplication
- security can be harder to enforce in every node + multiple entry points
- more complex to set up and maintain