EC2 101 Flashcards
What does EC2 stand for?
Elastic Compute Cloud
At a high level, what does EC2 provide?
EC2 is a web service that provides resizeable compute capacity in the cloud
Why is EC2 useful?
- EC2 reduces the time required to obtain & boot new server instances to minutes.
- This allows you to quickly scale capacity (both up and down) as computing requirements change
What are the 4 core principles behind the EC2 pricing model?
- Pay as you go
- Pay for what you use
- Pay less as you use more
- Pay even less as you reserve capacity
What are the 4 EC2 Pricing Models?
- On Demand
- Reserved
- Spot
- Dedicated Hosts
How does On Demand Pricing for EC2 instances work?
Pay a fixed rate by the hour (or second) with no time commitment
A user is developing and testing an application using EC2 for the first time. Which EC2 pricing model best suits their use case?
On Demand Pricing
At a high level, how does Reserved Pricing for EC2 instances work?
Reserved Pricing provides you with a capacity reservation, at a significant discount on the hourly rate.
How long is the contract on a Reserved Instance in EC2?
Can be 1 year or 3 years
A user’s application requires a reserved capacity, and they are able to make upfront payments to save money. Which pricing model makes the most sense in this scenario?
Reserved Pricing
What are the three types of Reserved Instances Classes?
- Standard Reserved Instances
- Convertible Reserved Instances
- Scheduled Reserved Instances
What is the key difference between a Standard Reserved Instance and a Convertible Reserved Instance?
A convertible reserved instance is more flexible on the fly. In general, you can change the attributes of the instance as long as the change results in a reserved instance class of equal or lesser value.
What is the best way to purchase compute capacity reservations that do not run continuously, but do recur on a regular schedule (daily, weekly, monthly)
Use Scheduled Reserved Instances
(Source)
How do EC2 Spot Instances work?
When EC2 has unusued capacity, Amazon drops the price of instances, so you can bid whatever price you want for instance capacity (“Free market”)
An application has an urgent computing need that calls for large amounts of additional compute capacity. Which EC2 pricing model makes the most sense for this use case?
Spot Instances
(On-Demand might also work here but will be costlier. Spot is better since we need additional compute. Spot should not handle 100% of your compute capacity)