Databases 101

Question 1

What is a relational database analagous to?

Answer 1

A traditional spreadsheet

Question 2

What does RDS stand for?

Answer 2

Relational Database Servers

Question 3

What are the 6 flavors of relational databases engines that are compatible with Amazon?

Answer 3

• SQLServer
• MySQLServer
• Aurora
• Oracle
• PostgreSQL
• MariaDB

Question 4

What are the 2 key features of RDS? What are they primarily for?

Answer 4

• Multi-AZ (for distaster recovery)
• Read Replicas (for performance)

Question 5

Does RDS Multi-AZ require changing the connection string?

Answer 5

If you lose access to your primary database, AWS can automatically point incoming requests to your secondary database without changing the connection string

Question 6

How do Read Replicas work in RDS?

Answer 6

• There is no automatic failover. If your primary instance goes down, you’ll need to create a new URL to connect to the secondary instance
• All writes to the primary instance are copied over to the read replica(s)

Question 7

Suppose you want to be able to handle a surge of incoming traffic to your RDS instance and scale out your application. What key feature of RDS would you use to do this?

Answer 7

Read Replicas

Question 8

What is the maximum number of read replicas that can be made from a single primary RDS instance?

Answer 8

up to 5 copies

Question 9

What is a non-relational database best analagous to?

Answer 9

A JSON Object

Question 10

What does OLTP stand for?

Answer 10

Online Transaction Processing

Question 11

What does OLAP stand for?

Answer 11

Online Analytics Processing

Question 12

What is the difference between OLTP and OLAP?

Answer 12

• Big difference is in the types of queries you will run
• OLTP is for a specific transaction
• OLAP will pull in a lot of data

Question 13

What is Amazon’s data warehousing solution?

Answer 13

Redshift

Question 14

Does RDS run on virtual machines?

Answer 14

Yes

Question 15

How can you log in to the OS of an RDS instance?

Answer 15

You can’t

Question 16

How can you patch the OS or database of an RDS instance?

Answer 16

This is Amazon’s responsibility (you can’t)

Question 17

Is RDS Serverless?

Answer 17

No, (except for Aurora)

RDS runs on virtual machines

Question 18

What is AWS DMS used for?

Answer 18

DMS allows you to migrate services from one source to AWS

Question 19

What is the difference between a homogenous migration and a heterogeneous migration?

Answer 19

• Homogenous migrations go between the same database engines
• Heterogenous migrations go between two different database engines

Question 20

What types of infrastructures can be used as a DMS Source?

Answer 20

The source can be:

• On-premises
• inside AWS itself
• another cloud storage provider (like Azure)

Question 21

What does SCT stand for?

Answer 21

Schema Conversion Tool

Question 22

Do you need AWS SCT for a homogenous migration?

Answer 22

No

Question 23

Do you need AWS SCT for a heterogenous migration?

Answer 23

Yes

Question 24

What does EMR stand for?

Answer 24

Elastic Map-Reduce

Question 25

What is AWS EMR?

Answer 25

EMR is the industry-leading cloud _big data platform_ for processing vast amounts of data using open-source tools like Apache Splunk, Apache hive, Apache HBase, Apache Flink, Apache Hudi, and Presto

Question 26

At what scale can AWS EMR run analysis?

Answer 26

EMR can run **petabyte-scale** analysis

Question 27

What is the cost-saving estimate for EMR over traditional on-premises solutions?

Answer 27

EMR offers analysis at **less than half the cost** of traditional on-premises solutions

Question 28

What is the estimated speed increase of EMR over Apache Spark?

Answer 28

**3x** faster

Question 29

What is a **cluster** in AWS EMR?

Answer 29

A **collection of EC2 instances**, each of which is called a **node**

Question 30

What are the three node types in EMR?

Answer 30

* Master Node * Core Node * Task Node

Question 31

What is the purpose of a master node in EMR?

Answer 31

The master node manages the cluster and tracks the status of tasks. ## Footnote **Every cluster has a master node**

Question 32

What is the purpose of a core node in EMR?

Answer 32

A core node **runs tasks and stores data in HDFS**. Multi-node clusters have at least one core node

Question 33

What does **HDFS** stand for?

Answer 33

**H**adoop **D**istributed **F**ile **S**ystem

Question 34

What is the purpose of a task node in AWS EMR?

Answer 34

A task node runs tasks, but does NOT store data in HDFS they are optional for your cluster.

Question 35

Suppose your EMR cluster wants to be able view log files even after the master node terminates. Is this possible?

Answer 35

**Yes**, **you can configure a cluster to periodically archive the log files stored on the master node to S3**. This ensures log files are available after the cluster terminates, whether through a normal shutdown or due to an error. Note that **this is only possible when you are creating the cluster for the first time**

Question 36

How often does EMR archive log files to S3?

Answer 36

**5-minute intervals**

Question 37

By default, how is log data stored in EMR?

Answer 37

By default, log data is stored **on the master node**

Question 38

Can you encrypt an unencrypted database snapshot?

Answer 38

**No**