Compute (EC2) Flashcards
Name the 6 EC2 components?
- AMIs
(templates for virtual machines (EC2 instances) - EC2 Instance types
(different hardware config) - EC2 Instance purchasing options
(On-demand, reserved, spot) - EC2 Tenancy
(shared / dedicated) - EC2 User data
(custom data passed at launch) - EC2 Storage options
(EBS volumes and instance store (ephemeral))
Name 5 EC2 instance types
General purpose
Memory optimized
Storage optimized
Compute optimized
Accelerated computing
Name 5 EC2 purchasing options
On-demand
Reserved
Scheduled
Spot
On-demand capacity reservations
What are the two types of EC2 tenancy that exist?
Shared
Dedicated
What are the 2 EC2 storage options?
Ephemeral storage
Elastic Block Storage
Name the 2 types of EC2 security?
Security Groups
NACLs
What do you need to do before you can create an Auto Scaling group?
you need to create a launch template with parameters (ie., AMI, ID and instance types)
then you can create an auto scaling group
Which is the new way of autoscaling EC2 components
Using a launch template is the new way
The old way was launch config
Launch templates use autoscaling instead of instances
Name 4 types of autoscaling policies
Manual
Dynamic (step and target)
Predictive
Scheduled
You create a EBS volume on your EC2 instance to store and manage data.
What should you do next?
You should create a file system to organize and manage files and directories on the volume
Describe the 4 steps to follow to create an EBS (Elastic Block Store) snapshot between AWS accounts?
2 AWS accounts: A is the source account and B is the destination account
- A shares EBS snapshot with B (adds B to its snapshot permissions)
- B accepts snapshot from A (copies EBS volume)
- B creates an EBS volume (contains data from original snapshot)
- EBS Volume attached and used by B
When does EC2 Autoscaling check the health status of the instance?
After the grace period ends. The default grace period is 300 seconds (5mins)
Gives instances a chance to stabilze before the Auto Scaling group begins checking their health
What does the cooldown period do when Auto Scaling?
The cooldown period is the waiting period between scaling activities
This prevents excessive or rapid scaling by allowing instances time to stabilize before further adjustments are made
Does AWS have access to the OS of your app?
No
Who is responsible for patching the system?
The customer is responsible for patching the system
What is the default number of EC2 instances that you can launch?
Default number of instances you can launch is 20
What happens if you try to create more EC2 instances than your account limit?
If you try to create more EC2 instances than your account limit, instance creation will fail and you will receive the error:
status=start_failed
Name the required and optional parameters when creating a launch template
Required params:
Instance type
Optional params:
Instance ID
AMI
Key Pair
Security Groups
UserData
IAMRole
etc
How would you back up multiple EBS volumes attached to an ec2 instance?
You would need to create a multi-volume snapshot (ie., create a snapshot for each of the EBS volumes attached to the ec2 instance)
To do this: choose “Instance” as the Resource Type in the Snapshot navigation pane
When would you create a multi-volume snapshot instead of a snapshot?
A snapshot and multi-volume snapshot both create backups for EBS volumes
There is no separate action for a multi-volume snap
You just choose if you want to create a snapshot for a single EBS volume or multiple EBS volumes attached to the same EC2 instance
Name 2 storage options for EC2 instances?
Instance Store (ephemeral)
EBS-Backed
What do you pay for an EC2 instance in a hibernation state?
EBS Storage charges and Elastic IP address charges
You don’t pay for compute charges because it is in hibernation (not being used)
Can Reserved Instance be queued to be purchased at a later date?
Yes
Can be queued for regional instance (can be used across multiple availability zones within the same AWS region)
(not zonal reserved instance or instance brought from other sellers)
Examples of when u would use scheduled scaling
Everyone logs in a 9am so at 8.45 start scaling up