Data Stores

Question 1

Concepts

What are the three types of data persistence?

Answer 1

Persistent, Transient, Ephemeral

Question 2

Concepts

What is a persistent data store?

Answer 2

Durable data that sticks around after reboots, restarts, power cycles, etc.

Glacier, RDS

Question 3

Concepts

What is a transient data store?

Answer 3

Temporary data that is stored and passed along to another process or storage
ex. SQS, SNS

Question 4

Concepts

What is an ephemeral data store?

Answer 4

A temporary store where data is lost after stop.
ex. EC2 Instance store, memcached

Question 5

Concepts

Explain IOPS

Answer 5

A measure of how fast we can read and write to a device

Question 6

Concepts

What is throughput?

Answer 6

A measure of how much data can be moved at a time

Question 7

Concepts

What are the two types of consistency models in data storage?

Answer 7

ACID and BASE

Question 8

Concepts

What is ACID consistency model?

Answer 8

Atomic - All or nothing
Consistent - Transactions are valid
Isolated - Transactions don’t interfere with each other
Durable - Transactions stick around and won’t disappear.

Question 9

Concepts

What is the BASE consistency model?

Answer 9

Basic availability - Values availability even if stale
Soft-state - Might not be consistent across all stores
Eventual consistency - Will achieve consistency eventually

Question 10

Concepts

What is the benefit of ACID over BASE?

Answer 10

Data is always consistent

Question 11

Concepts

What is the benefit of BASE?

Answer 11

It scales much better than ACID

Question 12

S3

What kind of Store is S3?

Answer 12

Object Store

Question 13

S3

What is the maximum object that can be stored in S3?

Answer 13

5TB

Question 14

S3

What is the maximum PUT size when uploading to S3?

Answer 14

5GB

Question 15

S3

When is it recommended to use multi-part uploads?

Answer 15

When the file size is larger than 100MB

Question 16

S3

The S3 “path” is not a file path, but what?

ex: s3://mybucket/finance/april/16/invoice_45675.pdf

Answer 16

A “key”

This key uniquely identifies the record in the file store. The record be

Question 17

S3

What security measures does S3 provide?

Answer 17

Recourse-based (object, ACL, bucket policy)
User-based (IAM Policies)
Optional multi-factor auth before elete

Question 18

S3

True or False: You can enable versioning in S3?

Answer 18

True

S3 will create a new version of the file with each write

Question 19

S3

What is the benefit to S3 versioning?

Answer 19

Enables “rollback,” and can be integrated with lifecycle management capabilities

Question 20

True or False: Versioning is not compatible with S3 lifecycle capabilities

Answer 20

False

Versioning IS compatible with lifecycle capabilities

Question 21

S3

What is the downside to versioning?

Answer 21

Old versions count as billable size until permanently deleted

Ensure you have a lifecycle management policy to control costs

Question 22

S3

True or False: S3 supports cross-region replication?

Answer 22

True

Question 23

S3

What are the benefits of cross-region replication?

Answer 23

Security Requirements
Compliance Requirements
Latency

Question 24

S3

Why should you consider latency for cross-region replication?

Answer 24

Customers accessing data closer to them results in faster response times

Question 25

S3

Tier: S3 Standard

Answer 25

Frequently accessed data with redundancy

Question 26

S3

Tier: Standard-IA

Answer 26

Long-lived, infrequently accessed data

Question 27

S3

Tier: One Zone-IA

Answer 27

Long-lived, infrequently accessed, non-critical data

Question 28

S3

Tier: Reduced Redundancy

Answer 28

Frequently accessed, non-critical data

Question 29

S3

Tier: Intelligent Tiering

Answer 29

Long-Lived data with changing or unknown access patterns

Question 30

S3

Tier: Glacier

Answer 30

Long-Term data archiving

Retrieval times ranfing from minutes to hours

Question 31

S3

Tier: Glacier Deep Archive

Answer 31

Long-term data archiving

Retrieval times within 12 hours

Question 32

S3

What is intelligent tiering?

Answer 32

Automatic moving of data between tiers based on usage

To reduced storage costs

Question 33

S3

Why should someone consider intelligent tiering when it’s more expensive?

Answer 33

Depending on the use case, you will make it up in reduced storage costs

Question 34

S3

True or False: Intelligent tiering will automatically archive unused data?

Answer 34

False: You have to configure archival policies

Question 35

S3

What are the 3 benefits of S3 lifecycle management?

Answer 35

Optimize storage costs
Adhere to data retention policies
Keep S3 buckets well maintained

Question 36

S3

In what two ways can lifecycle management transition your files?

Answer 36

Transition between storage classes
Transition to archive

Question 37

S3

What three ways can files be marked as lifecycle managed?

Answer 37

Prefixes
Tags
Current version vs previous version

Question 38

S3

What services are used with DataLakes?

S3 Analytics

Answer 38

Athena, Redshift, Spectrum, Quicksight

Question 39

S3

What service is used with streaming data?

S3 Analytics

Answer 39

Kinesis Firehose

Question 40

S3

What services support AI/ML

S3 Analytics

Answer 40

Rekognition, Lex, MXNet

Question 41

S3

What service supports Storage class analysis?

S3 Analytics

Answer 41

S3 Management Analytics