AWS Cloud Practitioner Study Session Two

December 27, 2025

I am taking the AWS Cloud Practitioner Exam in approximately eight days and want to ensure I am prepared. This series will serve as non-exhaustive note taking for the information that I am internalizing as I go.

ChatGPT Summary:

AWS Certified Cloud Practitioner – EC2 Instance Types & Naming Summary

This section is very exam-heavy. AWS expects you to match workload needs to the correct EC2 instance family, not memorize specs.

How to Approach EC2 Questions on the Exam

When you see an EC2 question, immediately ask:

1. What resource is the bottleneck?
   • CPU?
   • Memory?
   • Storage (IOPS / throughput)?
   • GPU / acceleration?
2. Is the workload steady or spiky?
3. Is cost optimization a priority?

Then select the instance family optimized for that constraint.

EC2 Instance Categories (Big Picture)

Think of EC2 families as falling into what they are optimized for:

General Purpose Instances

Families

• T
• M

Use Cases

• Web servers
• Application servers
• Development & test environments

Key Idea

• Balanced CPU, memory, and networking
• Default choice when nothing special is required

🧠 Memory Tip:
M = Middle-of-the-road (balanced)
If the question doesn’t strongly emphasize CPU, memory, or storage → General Purpose

Compute Optimized Instances

Families

• C

Use Cases

• High-performance web servers
• Batch processing
• Scientific modeling
• CPU-bound workloads

Key Idea

• High CPU-to-memory ratio

🧠 Memory Tip:
C = Compute
If the question mentions “CPU-intensive”, “high performance processing”, or “number crunching” → C

Memory Optimized Instances

Families

• R
• X
• z

Use Cases

• In-memory databases
• Real-time big data analytics
• High-performance relational and NoSQL databases

Key Idea

• Large amounts of RAM

🧠 Memory Tips:

• R = RAM
• X = eXtra memory
• Databases + “low latency” + “in-memory” → Memory Optimized

Storage Optimized Instances

Families

• I
• D
• H

Use Cases

• Data warehousing
• Log processing
• Distributed file systems
• NoSQL databases

Key Idea

• High disk throughput and IOPS

🧠 Memory Tips:

• I = IOPS
• D = Dense storage
• If the question emphasizes “large amounts of data on disk” or “high read/write performance” → Storage Optimized

Accelerated Computing Instances

Families

• P
• G
• Inf
• Trn

Use Cases

• Machine learning training and inference
• High-performance computing
• Video rendering
• Financial modeling

Key Idea

• Uses GPUs or specialized accelerators

🧠 Memory Tips:

• P = Powerful GPUs (ML training)
• G = Graphics / inference
• Inf = Inference
• Trn = Training

If you see ML, AI, GPUs, or acceleration, this category is the answer.

Burstable Performance Instances

Families

• T4g
• T3
• T2

Use Cases

• Low-traffic websites
• Microservices
• Dev/Test environments

Key Idea

• Baseline performance with the ability to burst using CPU credits

🧠 Memory Tip:

• T = Tiny traffic (most of the time)
• Spiky workloads + cost sensitivity → T instances

High Performance Computing (HPC) Instances

Families

• Hpc6a
• Hpc7g

Use Cases

• Weather modeling
• Computational fluid dynamics
• Genomics

Key Idea

• Extremely high compute performance
• Often used for scientific simulations

🧠 Memory Tip:

• If the question sounds like supercomputing, think HPC

Arm-Based Instances

Families

• T
• M
• C
• R (Arm-based variants)

Use Cases

• Linux workloads
• Containers
• Microservices

Key Idea

• Uses AWS Graviton (ARM) processors
• Better price/performance for supported workloads

🧠 Exam Tip:

• If the question mentions cost efficiency, Linux, or containers, Arm-based instances may be the best choice.

EC2 Naming Conventions (Very Testable)

🧠 One-Line Memory Trick:

C runs hot, R remembers, I stores, P powers AI

Final Exam Takeaways

• EC2 questions are about matching workloads to instance families
• Look for keywords:
  • CPU → Compute Optimized
  • RAM → Memory Optimized
  • Disk → Storage Optimized
  • GPU / ML → Accelerated Computing
  • Spiky traffic → Burstable
• General Purpose is the safe default
• Naming conventions are intentionally descriptive—use them

Study materials:

• Free Code Camp Preparation
• AWS Certified Solutions Architect Practice Tests
• AWS Cloud Practitioner Essentials
• AWS Documentation
  • What is Cloud Computing?
  • Shared Responsibility Model
  • Regions and Availability Zones
• ChatGPT

Raw Input Notes:

• Different types of EC2 instances:
  • General Purpose
    • Families: T, M
    • Use Cases: Web servers, App servers, Development and test environments
  • Compute Optimized
    • Families: C,
    • Use Cases: High performance web servers, Batch processing, Scientific
  • Memory Optimized
    • R, X, z
    • Use Cases: In-mem databases, Real-time big data analytics
    • High performance relational / non-relational dbs
  • Storage Optimized:
    • I, D, H
    • Use Cases: Data warehousing, Log processing, Distributed file systems, NoSQL
  • Accelerated Computing:
    • P, G, Inf, Trn
    • Use Cases: Machine learning training + inference, HPC, Video rendering, Financial modeling
  • Burstable Performance:
    • T4g, T3, T2
    • Use Cases: Low-traffic websites, Microservices, Dev / Test Envs
  • High Performance Computing:
    • Hpc6a, Hpc7g
    • Use Cases: Weather modeling, Computational fluid dyanmics Genomics
  • Arm-based Instances:
    • T, M, C, R
    • Linux based workloads, containers, Microservices

• Naming Conventions:
  • T = Burstable
  • M = Balanced
  • C = Compute Optimized
  • R = Memory Optimized
  • X = Extra Memory
  • I = High IOPS Storage
  • D = Dense Storage
  • P = GPU (ML training)
  • G = GPU (Graphics / inference)
  • Inf = ML Inference
  • Trn = ML Training HPC = High-Performance Computing