🧾 AWS Cloud Practitioner Notes – Part 1

Keyword / Topic Details

Amazon Server / Amazon Elastic Compute Cloud (EC2) is a virtual server in the cloud. It's shared via
EC2 multitenancy.
Client-Server A foundational computing model where servers provide resources and services to
Model client devices.
AWS Cloud Leverages AWS experience and best practices to digitally transform and accelerate
Adoption business outcomes. CAF identifies specific organizational capabilities underpinning
Framework (CAF) successful cloud transformations.
AWS TCO Allows users to input their current infrastructure details and compare them with
Calculator estimated AWS costs.
Key Value "Pay for what you need" model — you are charged based on usage, not capacity
planning.
AWS Site-to-Site Uses IPSec to establish encrypted connections between on-premises networks and the
VPN AWS Cloud.
CloudHSM is a cloud-based service that provides dedicated hardware for securely managing and
(Hardware processing cryptographic keys
Security Module) easily generate and use your own encryption keys in the AWS Cloud

🧾 AWS Cloud Practitioner Notes – Part 2

Keyword / Details
Topic
Deployment Cloud-based deployment model:
Models • Run all parts of the application in the cloud.
• Migrate existing applications to the cloud.
• Design and build new applications in the cloud.

On-premises deployment (private cloud):

• Use virtualization and resource management tools.
• Increase resource utilization using app management & virtualization.

Hybrid deployment:
• Connect cloud-based resources with on-premises infrastructure.
• Integrate cloud resources with legacy IT apps.
Cloud - Trade CapEx for OpEx: Pay only for the compute resources you use instead of large
Computing upfront costs.
Benefits - Stop running and maintaining data centers: Save time and money on infrastructure
management.
- Stop guessing capacity: Scale in/out based on demand with services like EC2.
- Massive economies of scale: AWS aggregates demand from many customers.
- Increased speed and agility: Get new resources in minutes.
- Go global in minutes: AWS has a global footprint with low latency access worldwide.
Types of EC2 - General Purpose: Balanced compute/memory/networking. Used for app servers,
Instances backend servers, small DBs.
- Compute Optimized: High performance CPUs. Ideal for compute-heavy tasks like gaming,
 HPC. However, the difference is compute optimized applications are ideal for high-
performance web servers, compute-intensive applications servers, and dedicated gaming
servers. You can also use compute optimized instances for batch processing workloads
that require processing many transactions in a single group.
- Memory Optimized: Large in-memory datasets and real-time processing.
- Accelerated Computing: Use GPUs/FPGAs for floating-point calcs, graphics, ML.
- Storage Optimized: High IOPS needs like OLTP systems, data warehousing.

🧾 AWS Cloud Practitioner Notes – Part 3

Keyword / Topic Details

EC2 Purchase Options  - On-Demand Instances:
Ideal for short-term, irregular workloads. No upfront cost. Pay for what
you use.
Use cases: app development, testing, unpredictable workloads.

- Reserved Instances:
Billing discount for steady workloads. Available in 1- or 3-year terms.
• Standard Reserved Instances: Greater savings, less flexible.
 Reserved Instances require you to state the following
qualifications:
Instance type and size: For example, [Link]
 Platform description (operating system): For example,
Microsoft Windows Server or Red Hat Enterprise Linux
 Tenancy: Default tenancy or dedicated tenancy

• Convertible Reserved Instances: Flexible instance types & AZs.

- EC2 Instance Savings Plans:

Commit to an hourly spend for 1- or 3-year term. Up to 72% savings.
Flexible use within the instance family and region.

- Spot Instances:
Use unused EC2 capacity at up to 90% discount.
Workloads must tolerate interruptions.
Flexible times are okay

- Dedicated Hosts:
Physical servers allocated to you. Help with license compliance. Most expensive
EC2 option.
Amazon EC2 Auto - Ensures the right number of instances are running to handle the load.
Scaling - Two types:
• Dynamic scaling: Responds to real-time demand.
• Predictive scaling: Uses ML to forecast demand and scale automatically.
 Elastic Load Balancing - Distributes incoming traffic across multiple EC2 instances.
(ELB) - Acts as a single point of contact.
- Ensures no single instance handles all load—improves fault tolerance and
performance.
Amazon SNS (Simple - Publish/Subscribe (Pub/Sub) messaging service.
Notification Service) - Cashier gives orders to barista analogy: publisher to subscriber.
- Subscribers can be email, SMS, Lambda, HTTP endpoints, etc.
Amazon SQS (Simple - Message queueing service.
Queue Service) To decouple components of a monolithic application on AWS, Amazon SQS
(Simple Queue Service) can be utilized.
- Messages added to queue, processed independently.
- Example: Coffee shop orders go into a queue, processed one by one.
AWS X-Ray AWS X-Ray is a service that helps developers analyze and debug distributed
applications, particularly those built using microservices.

🧾 AWS Cloud Practitioner Notes – Part 4

Keyword / Topic Details

AWS Lambda Run code without provisioning or managing servers. Serverless compute. Pay only
for execution time.

Amazon ECS (Elastic Managed container orchestration service. Supports Docker. Use API calls to
Container Service) start/stop containers.
Docker Software platform for building, testing, and deploying applications in containers.
AWS supports Docker CE and EE.
 Amazon EKS (Elastic Fully managed Kubernetes service. Easily run, scale, and manage containerized
Kubernetes Service) applications using Kubernetes.
Kubernetes Open-source system for automating deployment, scaling, and management of
containerized apps. AWS works with the Kubernetes community.
AWS Fargate Serverless compute engine for containers. Works with ECS and EKS. No need to
manage servers; pay for what you use.
Amazon CloudFront Content Delivery Network (CDN) that delivers data, videos, apps globally with low
latency. Uses edge locations to cache content.
This content might be video files, photos, webpages, and so on.
Amazon Machine Template used to launch EC2 instances. Includes OS, application server, and apps.
Image (AMI)
Amazon Cloud Service to organize data across multiple hierarchies and dimensions. Useful for
Directory directory-based apps.
Serverless Stack Includes Lambda and DynamoDB—no infrastructure to manage.
Availability Zone (AZ) One or more data centers in a region. Each region has at least 3 AZs.
Region Geographical area with multiple AZs.
Edge Location Site used by CloudFront to cache content closer to users.

AWS Management Web interface for managing AWS services. Includes wizards and workflows.
Console
AWS CLI (Command Control multiple AWS services via terminal. Available on Windows, macOS, and
Line Interface) Linux.
AWS SDKs Software Development Kits for different languages and platforms. Use AWS services
from code.

🧾 AWS Cloud Practitioner Notes – Part 5

Keyword / Topic Details

AWS Elastic Platform-as-a-Service (PaaS) that handles deployment, from provisioning to
Beanstalk monitoring. You provide code and configuration, Beanstalk handles:
• Capacity adjustment
• Load balancing
• Auto scaling
• Health monitoring
AWS Infrastructure as Code (IaC). Write templates to provision AWS resources. Benefits:
CloudFormation • Safe and repeatable deployments
• Detects and rolls back on error
• Speeds up environment setup
Amazon VPC Isolated section of AWS Cloud where you launch resources in a virtual network you
(Virtual Private define.
Cloud) Components include:
• Subnets – public (web servers), private (databases)
• Internet Gateway – access to internet
• Virtual Private Gateway – private access
• AWS Direct Connect – private line to AWS from data center
Subnet Logical division within a VPC can group resources based on security or operational needs.
Subnets can be public or private.
• Public subnets: expose resources to the internet
• Private subnets: internal-only access
Subnets can communicate with each other.
Packet Unit of data sent over a network.
Security Acts as a virtual firewall. Stateful – return traffic automatically allowed.
Group
(Instance-
level)
Network Controls traffic at the subnet level. Stateless – rules must be explicitly set for both inbound and
ACL outbound traffic.
(Subnet-
level)
Amazon Scalable and highly available DNS web service.
Route 53 Resolves domain names to IP addresses.
Also provides health checking and traffic routing.
AWS Easily provision and manage SSL/TLS certificates for AWS services.
Certificat
e
Manager
(ACM)
Instance Temporary disk storage physically attached to EC2 host. Data lost when instance stops or
Store terminates. Good for:
• Temporary data
• Buffers and caches
Amazon Persistent block storage for EC2. Data survives stop/terminate. Available within a single AZ.
EBS
(Elastic
Block
Store)
EBS Incremental backup of EBS volume. Stores only changes since last snapshot. Useful for disaster
Snapshot recovery.
 An Amazon EBS volume stores data in a single Availability Zone.

Amazon 📦 Amazon S3 Storage Classes - Summary Table

S3 Storage Best For Availabil Cost Retrieval Special Notes
(Simple Class ity Speed
Storage Zones
Service) S3 Frequently 3+ High Immediat Default, high
Standar accessed data e durability
d (websites,
analytics)
S3 Infrequently 3+ Lower Immediat Lower storage,
Standar accessed, but e higher retrieval
d-IA needs fast cost
access
S3 One Infrequently 1 Lowest Immediat Risk of AZ failure,
Zone-IA accessed, cost- among e for easily
saving above reproducible
data
S3 Unknown/ 3+ Auto- Immediat Charges
Intellige changing adjusti e monitoring fee,
nt- access patterns ng (depends moves objects
Tiering ) between tiers
S3 Archived data 3+ Very Milliseco For quick archival
Glacier that needs fast Low nds access
Instant access
Retrieva
l
S3 Archived data, 3+ Lower Minutes For backup or old
Glacier not often to hours records
Flexible needed
Retrieva
l
S3 Rarely 3+ Lowest 12 to 48 For long-term
Glacier accessed data hours archiving
Deep (1–2
Archive times/year)
S3 On-premise On-prem Varies Local For data
 Outpost AWS storage access residency/compli
s needs ance
requirements

🔑 Key Points to Remember

 Durability: All storage classes (except One Zone-IA) replicate across 3+ AZs →
99.999999999% durability.
 📁 Max object size: 5 TB per object.
 ✅ Unlimited storage.
 🔁 Write once / Read many supported (WORM).
 📊 Access pattern automation: Intelligent-Tiering handles moving data between Standard
and IA.

🧾 AWS Cloud Practitioner Notes – Part 6

Keyword / Topic Details

Amazon EFS (Elastic File System) Scalable file storage that works with AWS and on-premises resources.
REGIONAL
• Automatically grows/shrinks with file system usage
• Can scale to petabytes
• Stores data across multiple AZs
• Accessible from multiple instances and on-prem systems via Direct
Connect
Amazon RDS (Relational Managed relational database service.
Database Service) • Automates provisioning, patching, backups, and scaling
• Supports multiple engines:
- Amazon Aurora (MySQL & PostgreSQL compatible)
- PostgreSQL, MySQL, MariaDB
- Oracle, SQL Server
• Encryption in-transit and at-rest supported
Amazon Aurora High-performance relational DB engine compatible with
MySQL/PostgreSQL.
• Up to 5x faster than MySQL, 3x faster than PostgreSQL
• Replicates 6 copies of data across 3 AZs
• Continuously backs up to S3
Amazon DynamoDB Serverless key-value and document NoSQL database.
• Single-digit millisecond latency
• No need to provision/manage servers
• Automatically scales throughput
• Built-in security, backup & restore, in-memory caching with DAX
Amazon Redshift Data warehousing service for big data analytics.
• Can query petabytes of structured/semi-structured data
• Integrates with business intelligence tools
Amazon DocumentDB Managed document database compatible with MongoDB.
• Stores data in JSON-like format
Amazon Neptune Graph database optimized for highly connected data (e.g., fraud
 detection, recommendation engines).
Amazon QLDB (Quantum Ledger Immutable ledger database for tracking changes in app data.
Database) • Maintains complete history of changes
• Cryptographically verifiable
Amazon Managed Blockchain Create and manage scalable blockchain networks using open-source
frameworks.
Amazon ElastiCache Adds in-memory cache layer to databases.
• Supports Redis and Memcached
• Improves read speeds for frequent queries
DynamoDB Accelerator (DAX) In-memory cache for DynamoDB that improves performance from
milliseconds to microseconds.
Amazon EMR (Elastic Managed big data framework for running tools like Hadoop & Spark.
MapReduce) • Used to process and analyze large datasets in the cloud
AWS Security – Shared - Customer is responsible for security in the cloud:
Responsibility Model • Managing data, identity, access, OS patches, firewall, app security

- AWS is responsible for security of the cloud:

• Protecting infrastructure: hardware, software, networking, and
facilities
• Physical security of data centers, virtualization, and network layers

To promote AWS adoption

 AWS Professional Services assists customers in achieving their desired business outcomes

 AWS Partners.

🧾 AWS Cloud Practitioner Notes – Part 7

Keyword / Topic Details

AWS IAM (Identity and Used to securely manage access to AWS services and resources.
Access Management) • IAM Users: Identities for individual people or apps with specific
credentials
• IAM Policies: JSON documents that allow or deny permissions
 • IAM Roles: Similar to job roles; temporary access. Used when assuming
roles across services (e.g., EC2 assumes role to access S3)
• IAM Groups: Collections of users sharing the same permissions
• Multi-Factor Authentication (MFA) supported for extra security
IAM User Permissions By default, new users have no permissions. You must attach policies to
grant access (e.g., S3 read-only, EC2 full access).
AWS Organizations Centralized account management system.
• Group accounts into Organizational Units (OUs)
• Apply Service Control Policies (SCPs) to root, OU, or individual accounts
• SCPs restrict services/actions IAM users or roles can access, even if
granted in IAM
Monitoring & Auditing Tools Amazon CloudWatch:
• Monitoring service for AWS resources and custom metrics
• Create alarms that auto-trigger actions when thresholds are crossed

AWS CloudTrail:
• Records all API calls
• Tracks: who called, when, source IP, and what was done
• Think of it as a “breadcrumb trail” of user and service activity

AWS Trusted Advisor:

• Provides real-time guidance on best practices in 5 categories:
• Cost optimization
• Security
• Fault tolerance
• Performance
• Service limits
Amazon Inspector Automatically assesses applications for vulnerabilities and deviations from
best practices. Scans EC2 instances for security flaws.

🧾 AWS Cloud Practitioner Notes – Part 8

Keyword / Topic Details

AWS Budgets Allows you to plan your service usage, cost, and instance
reservations.
• Set custom alerts when usage exceeds thresholds.
• Plan for monthly/quarterly/annual spending limits.
AWS Cost Explorer Visual tool to view and analyze AWS spending and usage.
• See trends over past 7, 30, or 60 days.
• Get recommendations for Savings Plans based on past EC2 usage.
• Helps identify cost-saving opportunities.
AWS Support Plans Five levels of support based on business needs:
1. Basic (Free):
• Account & billing support
• AWS documentation and whitepapers

2. Developer:
• Best practice guidance
• Building-block architecture support
 • Limited general guidance

3. Business:
• 24/7 access to Cloud Support Engineers
• Use-case guidance
• Full access to Trusted Advisor checks
• Limited 3rd-party software support

4. Enterprise On-Ramp:
• Technical Account Manager (TAM) pool access
• Concierge billing support
• One consultative review and one cost workshop/year

5. Enterprise:
• Designated TAM
• Infrastructure Event Management
• Custom reporting, automation workflows
• 15-minute response time for critical issues
TAM (Technical Account Manager) Assigned to Enterprise or On-Ramp customers.
• Primary AWS contact
• Provides architectural guidance
• Coordinates access to AWS experts and programs
• Ensures cost efficiency and operational excellence
6 Strategies for Cloud Migration (6 Common approaches when moving applications to the cloud:
R’s)
1. Rehosting (“Lift and shift”)
• Move apps without changes.
• Fastest option for legacy migration.

2. Replatforming (“Lift, tinker, and shift”)

• Small cloud optimizations.
• No core architectural changes.

3. Refactoring / Re-architecting
• Redesign app using cloud-native features.
• Adds scalability, flexibility.

4. Repurchasing
• Move to SaaS (e.g., from local CRM to Salesforce).

5. Retaining
• Keep apps that require major refactoring or are mission-critical.

6. Retiring
• Decommission apps that are no longer needed.
 AWS Cloud Practitioner Notes – Part 9

Keyword / Topic Details

AWS Snow Family Hardware devices for transferring large datasets to AWS when network isn’t
feasible.

1. AWS Snowcone:
• Small, portable edge device
• 2 CPUs, 4 GB RAM, 14 TB usable storage

2. AWS Snowball Edge (Storage Optimized):

• For large-scale transfers, local compute
• 80 TB HDD + 1 TB SSD
• 40 vCPUs, 80 GiB RAM

3. AWS Snowball Edge (Compute Optimized):

• For ML, analytics, and video
• 104 vCPUs, 416 GiB RAM, optional GPU
• 80 TB HDD + 28 TB NVMe SSD

4. AWS Snowmobile:
• Exabyte-scale transfer
• 45-foot shipping container moved by truck
• Up to 100 PB per Snowmobile
AI Services Amazon Transcribe: Converts speech to text.
Amazon Comprehend: Finds patterns in text.
Amazon Fraud Detector: Detects fraud.
Amazon Lex: Creates voice/text chatbots.
Amazon Textract: Extracts text/data from scanned documents.
AWS DeepRacer: 1/18 scale car for reinforcement learning model testing.
Machine Learning Services Amazon SageMaker: Build, train, and deploy ML models easily.
• Removes need to integrate complex tools manually.

Amazon Q Developer:
 • ML-powered code generator
• Analyzes your code & comments
• Gives suggestions matching your coding style
• Helps write functions faster in IDE
Well-Architected Based on 6 pillars for cloud best practices:
Framework
1. Operational Excellence
• Run and monitor systems, improve over time
• Use CloudFormation for infrastructure as code

2. Security
• Protect data, systems, and assets
• Implement IAM, encryption, monitoring

3. Reliability
• Recover quickly from failures
• Design systems that adapt to demand

4. Performance Efficiency
• Use IT resources efficiently
• Evolve with tech changes

5. Cost Optimization
• Eliminate unneeded costs
• Pay only for what you use

6. Sustainability
• Reduce environmental impact
• Use managed services & efficient architectures