Week 3: Big Data Modelling

Question 1

3 Components of a data model

Answer 1

• Structures
• Operations
• Constraint

Question 2

How is a data model characterized ?

Answer 2

• Structure of the data that it admits
• Operations on that strcutre
• A way to specift constraints

Question 3

Example of structued data

Answer 3

1. First name
2. Last name
3. DoB

Question 4

Example of semi-structured data

Answer 4

1. First Name
2. Last Name
3. DOB
4. Occupation
5. Salary

Question 5

Example of unstructured data

Answer 5

02023#02023#02023#02023#02023#02023#02023

• Hex dump

Question 6

NoSQL stands for

Answer 6

• Not Only SQL

Question 7

4 types of NoSQL database models

Answer 7

1. Key / Value
2. Document
3. Wide column stores
4. Graph

Question 8

4 Advanatages of Key / Value database

Answer 8

1. Simplest and most flexible
2. Stores a value indexed by a key
3. No Schema
4. Value is a binary large object which doesnt care about data type

Question 9

3 Key Value suitable use cases

Answer 9

1. Session information
2. User profiles and preferences
3. Shopping cart data

Question 10

3 Key Value unsuitable use cases

Answer 10

1. Data with relationships
2. Operations on sets of data
3. Transactions with multiple operations

Question 11

3 Document-oriented database suitable use cases

Answer 11

• Event logs
• CMS / Blogging
• Analystics

Question 12

3 Docuement-oriented database unsuitable use cases

Answer 12

• Complex cross-document transactions
• Queries that rely on fixed schema

Question 13

Wide column stores 3 use cases

Answer 13

• Event logs
• CMS / Blogging
• Analytics

Question 14

2 Wide column stores unsuitable use cases

Answer 14

• Complex cross-document transactions
• Queries that rely on fixed schema

Question 15

Graph database consists of what 2 elements

Answer 15

• Node - an entity
• Edge - a relationship

Question 16

3 Graph database suitable use cases

Answer 16

• Social networks
• Recommender systems
• Routing and location-base services

Question 17

Dont use graph databases for what use case

Answer 17

Operations on sets of data

Question 18

JSON is built on what 2 structures

Answer 18

• Collection of name/value pairs
• An ordered list of values

Question 19

What is a JSON object

Answer 19

an unordered set of name/value pairs

Question 20

What is an JSON array

Answer 20

an ordered collection of values

Question 21

Describe MongoDB Model

Answer 21

• Database
• Collections
• Documents

Question 22

Describe Relation DB Model

Answer 22

• Database
• Tables
• Rows

Question 23

Describe RDB vs MongoDB in terms of schema

Answer 23

Relation DB has a well deifne schema for the data it stores

MongoDb is schemaless

Question 24

Describe RDB vs MongoDB in terms of data rejection

Answer 24

• Relation DB will reject data that doesnt conform to schema
• Mongo DB can stored unstructured data where content is not known

Question 25

MongoDB 3 insert steps

Answer 25

1. db
2. use myDatabase
3. db.myCollection.insert( { “country” : “scotland” } )

Question 26

MongoDB remove operation

Answer 26

db.myCollection.remove( { “city” : “London” } )

Question 27

MongoDB update operation

Answer 27

db.myCollection.updateOne( {“country” : “USA”} , { $set : { “city” : “ New York” } } )

Question 28

An aggregation pipeline uses what 2 operations

Answer 28

1. $match
2. $group

Question 29

What is Map

Answer 29

Map is same computation across all items

Question 30

What is reduce

Answer 30

Combing the mapped values

Question 31

Design a MongoDB data model for a specified scenario and contrast it with the way the data would be modelled in a relational database (also in week 11)

Question 32

Retrieve MongoDB data from Python