Deployment Strategies

Question 1

In-place

Answer 1

Update your application on existing instances instead of replacing with new instances

Question 2

Rolling

Answer 2

Gradually roll out updates to your application to some portion of the production fleet

Question 3

Immutable

Answer 3

Stand up a new infrastructure with the new application version

Question 4

Blue/Green

Answer 4

Systems use the same backend (parallel environment)/ cloning and swapping parallel environments to reduce downtime

Question 5

What are the continuous deployment strategies?

Answer 5

In-place
Rolling
Immutable
Blue/Green

Question 6

deployment configurations

Answer 6

Amazon EC2/On-premises compute platform
Amazon ECS
AWS Lambda

Question 7

Amazon EC2/On-premises compute platform

Answer 7

deployment configurations:
AllAtOnce
HalfAtATime
OneAtATime

Question 8

Amazon ECS

Answer 8

deployment configurations:

ECSAllAtOnce

Question 9

AWS Lambda

Answer 9

deployment configurations:
Canary
Linear
All-at-once

Question 10

What tools can be used for Blue/Green deployments?

Answer 10

Elastic Load Balancing
Amazon Route 53
Amazon EC2 Auto Scaling

Question 11

Segmentation

Answer 11

Can assist you with your deployments and lower your risks. By breaking up deployments into small chunks, you minimize the impact of failures to only a small portion of your environment. You can also limit the traffic. Another reason to segment is to potentially catch issues before your users do. You also want to be able to roll back as quickly as possible if necessary.

Question 12

Segmentation options

Answer 12

Common segment types:

• region
• Availability Zones
• Subzones
• Single hosts (canary)

Create deployment groups per segment using:

• tags
• auto scaling groups