Domain 3: Cloud Technology and Services

AWS Certified Cloud Practitioner (CLF-C02) — Domain 3: Cloud Technology and Services

Exam Code: CLF-C02  |  Level: Foundational
Domain Weight: 34%  |  Total Domains: 4  |  Passing Score: 700 / 1000

---

Table of Contents

1. Compute Services
   • 1.1 Amazon EC2
   • 1.2 EC2 Instance Types and Families
   • 1.3 EC2 Pricing Models
   • 1.4 EC2 Auto Scaling
   • 1.5 AWS Lambda
   • 1.6 Container Services
   • 1.7 Additional Compute Services
2. Storage Services
   • 2.1 Amazon S3
   • 2.2 S3 Storage Classes
   • 2.3 S3 Key Features
   • 2.4 Amazon EBS
   • 2.5 Amazon EFS
   • 2.6 AWS Snow Family
   • 2.7 Storage Service Comparison
3. Database Services
   • 3.1 Amazon RDS
   • 3.2 Amazon Aurora
   • 3.3 Amazon DynamoDB
   • 3.4 Amazon ElastiCache
   • 3.5 Amazon Redshift
   • 3.6 Specialized Database Services
   • 3.7 Database Selection Guide
4. Networking and Content Delivery
   • 4.1 Amazon VPC
   • 4.2 Amazon Route 53
   • 4.3 Amazon CloudFront
   • 4.4 Elastic Load Balancing
   • 4.5 Amazon API Gateway
5. Developer Tools and Deployment
   • 5.1 CI/CD Services
   • 5.2 Infrastructure as Code
   • 5.3 Observability and Debugging
6. Analytics Services
7. Machine Learning and AI Services
   • 7.1 ML Platform
   • 7.2 AI API Services
8. Application Integration
   • 8.1 Amazon SNS
   • 8.2 Amazon SQS
   • 8.3 AWS Step Functions and EventBridge
9. Management and Governance
10. Migration and Transfer Services
11. Exam Tips and Quick Reference

---

1. Compute Services

1.1 Amazon EC2

Amazon Elastic Compute Cloud (EC2) provides resizable virtual servers (instances) in the cloud. It is the foundational compute service — the equivalent of renting a virtual machine with full control over the operating system.

Core concepts:

---

1.2 EC2 Instance Types and Families

Instance types are grouped into families based on workload optimization:

Instance naming convention:

c6i.xlarge
│ │  └─ Size (nano, micro, small, medium, large, xlarge, 2xlarge, ...)
│ └─── Generation (6)
└───── Family (c = compute optimized)

---

1.3 EC2 Pricing Models

Exam Tip: The three key attributes of Spot Instances are: up to 90% savings, can be interrupted by AWS with a 2-minute warning, and suitable only for fault-tolerant workloads. Do not use Spot for databases, web servers serving user traffic, or anything that cannot tolerate interruption.

Reserved Instance Payment Options (affects discount level):

---

1.4 EC2 Auto Scaling

Auto Scaling automatically adds or removes EC2 instances to match demand, ensuring availability and reducing costs during low-traffic periods.

Scaling directions:

Key Concept: AWS recommends horizontal scaling for high availability. Adding more smaller instances across multiple AZs is more resilient than making one instance larger.

Auto Scaling Group configuration:

Scaling policy types:

---

1.5 AWS Lambda

Lambda is a serverless compute service that runs code in response to events without requiring server provisioning or management.

How Lambda works:

Event Source (S3, API Gateway, SQS, etc.)
    │
    ▼
Lambda provisions compute automatically
    │
    ▼
Your function code executes
    │
    ▼
Resources released; you pay only for execution time

Key specifications:

Common Lambda event sources:

• API Gateway (HTTP requests)
• Amazon S3 (object events: created, deleted)
• Amazon DynamoDB Streams (table change events)
• Amazon SQS (message arrival)
• Amazon SNS (notifications)
• Amazon EventBridge (scheduled events, AWS service events)
• Amazon Kinesis (streaming data)

EC2 vs Lambda comparison:

---

1.6 Container Services

Core concepts:

AWS container services:

Key Concept: Fargate eliminates the need to provision or manage EC2 instances for containers. You define only the CPU and memory requirements; AWS handles the underlying infrastructure. It works with both ECS and EKS.

---

1.7 Additional Compute Services

---

2. Storage Services

2.1 Amazon S3

Amazon Simple Storage Service (S3) is object storage that stores any amount of data accessible from anywhere. It is not a file system or block storage — data is stored as objects (file + metadata) within buckets.

Core characteristics:

---

2.2 S3 Storage Classes

Exam Tip: S3 Intelligent-Tiering is the best choice when you cannot predict access patterns. It automatically moves objects between access tiers based on usage — there are no retrieval fees, only a small per-object monitoring fee.

---

2.3 S3 Key Features

---

2.4 Amazon EBS

Amazon Elastic Block Store provides block-level storage volumes for use with EC2 instances — similar to attaching a physical hard drive to a server.

Key characteristics:

• Attached to a single EC2 instance at a time (multi-attach available for io2 volumes in limited configurations)
• Locked to a single Availability Zone — must be in the same AZ as the EC2 instance
• Persists independently from the EC2 instance lifecycle (survives stop, start, and reboot)
• Supports snapshots — incremental backups stored in S3; can be used to create new volumes or share data across Regions

EBS volume types:

---

2.5 Amazon EFS

Amazon Elastic File System is a fully managed, elastic NFS (Network File System) that can be mounted concurrently by multiple EC2 instances across multiple AZs.

Key characteristics:

• Shared file system — multiple instances can read and write simultaneously (unlike EBS)
• Automatically scales — grows and shrinks without provisioning a fixed size
• Regional service — accessible from any AZ within a Region
• Supports Linux-based instances only (NFSv4 protocol)

EBS vs EFS comparison:

---

2.6 AWS Snow Family

Physical devices shipped by AWS used to migrate large volumes of data when internet bandwidth is insufficient or impractical.

Migration process:

1. Request the device via the AWS Console
2. AWS ships the encrypted device to your location
3. Connect locally and transfer data onto the device
4. Ship the device back to AWS
5. AWS uploads the data to your S3 bucket
6. AWS performs a software erasure of the device after transfer

Exam Tip: If a question describes a company that would take weeks or months to transfer data over the internet, AWS Snowball is the answer. General rule: if the transfer would take more than one week over available bandwidth, use Snow Family.

---

2.7 Storage Service Comparison

---

3. Database Services

3.1 Amazon RDS

Amazon Relational Database Service is a fully managed service for relational databases. AWS manages the database engine, hardware, OS patching, and automated backups.

Supported engines:

• Amazon Aurora (AWS-native, discussed separately)
• MySQL, PostgreSQL, MariaDB (open-source engines)
• Oracle, Microsoft SQL Server (commercial engines)

What RDS manages vs what the customer manages:

Multi-AZ Deployment vs Read Replicas:

Exam Tip: Multi-AZ = availability (protect against AZ failure). Read Replicas = performance (scale read traffic). These are complementary features often used together, but they serve different purposes.

RDS backup types:

---

3.2 Amazon Aurora

Aurora is AWS's cloud-native relational database, compatible with MySQL and PostgreSQL but designed from the ground up for cloud-scale performance.

Key advantages over standard RDS:

Aurora-specific features:

---

3.3 Amazon DynamoDB

DynamoDB is a fully managed, serverless NoSQL key-value and document database designed for single-digit millisecond latency at any scale.

Key characteristics:

• Serverless — no servers to provision, patch, or manage
• Single-digit millisecond read/write latency regardless of table size
• Highly available — data automatically replicated across 3 AZs within a Region
• Scales to trillions of items and hundreds of TBs
• Supports ACID transactions

Data model:

• A table contains items (equivalent to rows)
• Each item contains attributes (equivalent to columns)
• Each item has a Primary Key: a Partition Key alone, or a Partition Key + Sort Key combination
• Schema is flexible — attributes can vary between items

DynamoDB key features:

---

3.4 Amazon ElastiCache

ElastiCache is a fully managed in-memory caching service that improves application performance by retrieving data from a fast, managed cache instead of slower disk-based databases.

Supported engines:

Typical caching pattern:

Application → Check ElastiCache → Cache Hit: return data immediately
                                 → Cache Miss: query database → store result in cache → return data

---

3.5 Amazon Redshift

Redshift is a fully managed data warehouse service for analytical queries across petabyte-scale datasets.

Key characteristics:

• Based on PostgreSQL but optimized for OLAP (Online Analytical Processing), not OLTP
• Columnar storage — optimized for aggregation and analytical queries across many rows
• Massively Parallel Processing (MPP) — distributes query execution across multiple nodes
• Performs analytics 10x faster than traditional data warehouses at one-tenth the cost

Redshift ecosystem:

---

3.6 Specialized Database Services

---

3.7 Database Selection Guide

---

4. Networking and Content Delivery

4.1 Amazon VPC

A Virtual Private Cloud (VPC) is a logically isolated section of the AWS cloud where you launch resources in a virtual network that you define and control.

Core VPC components:

---

4.2 Amazon Route 53

Route 53 is AWS's scalable and highly available Domain Name System (DNS) service and domain registrar.

DNS routing policies:

---

4.3 Amazon CloudFront

CloudFront is AWS's global Content Delivery Network (CDN) that caches and delivers content with low latency through a network of 400+ Edge Locations.

How CloudFront reduces latency:

User (Tokyo) requests content
    │
    ▼
Route 53 directs to nearest Edge Location (Tokyo)
    │
    ├── Cache HIT: Content served from Edge Location immediately
    │
    └── Cache MISS: CloudFront fetches from Origin (e.g., S3 in us-east-1)
                    → Caches content at Edge Location
                    → Serves to user
                    → Future requests served from Edge Location (cache hit)

CloudFront origins:

Key CloudFront security features:

---

4.4 Elastic Load Balancing

Elastic Load Balancing (ELB) automatically distributes incoming traffic across multiple targets (EC2 instances, containers, Lambda functions, IP addresses) in one or more AZs.

---

4.5 Amazon API Gateway

A fully managed service to create, publish, maintain, monitor, and secure APIs at any scale.

• Supports REST APIs, HTTP APIs (lower latency and cost), and WebSocket APIs
• Integrated with Lambda, EC2, DynamoDB, and any HTTP endpoint
• Built-in: throttling, rate limiting, authentication (IAM, Cognito, Lambda authorizer), usage plans, and API keys

Common serverless API pattern:

Client → API Gateway → Lambda function → DynamoDB table

---

5. Developer Tools and Deployment

5.1 CI/CD Services

Complete CI/CD pipeline flow:

CodeCommit (Source)
    │
    ▼
CodeBuild (Build and Test)
    │
    ▼
CodeDeploy (Deploy to Target)

All orchestrated by CodePipeline

---

5.2 Infrastructure as Code

---

5.3 Observability and Debugging

---

6. Analytics Services

---

7. Machine Learning and AI Services

7.1 ML Platform

Amazon SageMaker is AWS's fully managed platform for building, training, and deploying machine learning models at any scale.

---

7.2 AI API Services

Pre-built AI capabilities accessible via API — no ML expertise required.

---

8. Application Integration

8.1 Amazon SNS

Amazon Simple Notification Service is a fully managed publish-subscribe (pub/sub) messaging service.

How it works: A publisher sends a message to a topic. The topic immediately pushes (fans out) that message to all subscribers simultaneously.

Subscriber types:

• Amazon SQS queues
• AWS Lambda functions
• HTTP/HTTPS endpoints
• Email and email-JSON
• SMS text messages
• Mobile push notifications (iOS, Android)

Common patterns:

---

8.2 Amazon SQS

Amazon Simple Queue Service is a fully managed message queue service for decoupling and scaling microservices, distributed systems, and serverless applications.

How it works: A producer sends messages to a queue. One or more consumers poll the queue, process messages, and delete them upon successful processing.

Queue types:

Key SQS concepts:

SNS vs SQS comparison:

---

8.3 AWS Step Functions and EventBridge

AWS Step Functions

A serverless workflow orchestration service that coordinates multiple AWS services into a sequence using a state machine definition. Provides built-in error handling, retry logic, and parallel execution. Maximum workflow duration of one year.

Amazon EventBridge

A serverless event bus that routes events from AWS services, custom applications, and third-party SaaS platforms to target services.

• 90+ AWS services emit events to EventBridge natively
• Create rules to filter events by source, detail type, or content
• Route events to Lambda, SQS, SNS, Step Functions, API Gateway, and more
• Supports scheduled rules (cron expressions) for time-based automation

---

9. Management and Governance

---

10. Migration and Transfer Services

---

11. Exam Tips and Quick Reference

Scenario-to-Service Mapping

---

Common Exam Traps

• EC2 vs Lambda: EC2 is for long-running, stateful, or workloads exceeding 15 minutes. Lambda is for short-duration, event-driven, stateless functions. Do not select Lambda for something that runs continuously.
• S3 vs EBS vs EFS: S3 = object storage accessed via API (not mountable as a file system directly). EBS = block storage for one EC2 instance. EFS = shared file system for multiple Linux EC2 instances simultaneously.
• Multi-AZ vs Read Replicas: Multi-AZ = availability (standby is not readable). Read Replicas = performance (replicas are readable).
• SNS vs SQS: If the scenario says "notify multiple systems at once" → SNS. If it says "decouple producer from consumer" or "buffer requests" → SQS.
• Redshift vs RDS: Redshift = analytics on historical data (OLAP). RDS = transactional operational database (OLTP). Do not use Redshift for transactional workloads.

---

End of Domain 3. Continue to Domain 4: Billing, Pricing, and Support →