Instance Types, AMIs, and Storage
Choosing the right instance type is an architectural decision.
EC2 instance types are grouped by workload patterns:
- General purpose (balanced CPU/memory2)
- Compute optimized
- Memory optimized
- Storage optimized
- GPU/accelerated computing
Instance selection affects:
- Performance
- Cost
- Scalability
- Application stability
For example, a memory-intensive analytics engine on a compute-optimized instance will underperform. The architecture must align with workload characteristics.
An Amazon Machine Image (AMI) defines the template for your instance. It includes:
- Operating system
- Preinstalled software
- Configuration settings
Production insight:
Never manually configure instances in production. Instead:
- Create a hardened base AMI.
- Bake dependencies into the image.
- Version your AMIs.
- Deploy immutable infrastructure.
Storage considerations:
EC2 commonly uses Amazon EBS (Elastic Block Store) for persistent storage. Instance store volumes provide temporary storage tied to the instance lifecycle.
Architectural implications:
- EBS volumes survive instance termination (if configured).
- Instance store does not.
- Separate compute from data when possible.
In this section, I learned:
0 of 4 completed