AWS Technology Selection: Compute, Storage, Database, and Networking

This guide covers the critical decision-making process for selecting the right AWS services based on specific business and technical requirements, aligned with the AWS Certified Solutions Architect – Associate (SAA-C03) exam objectives.

Learning Objectives

After studying this guide, you should be able to:

• Evaluate compute options based on workload duration, statefulness, and management overhead.
• Select storage solutions (Object, Block, File) based on performance, access patterns, and durability needs.
• Differentiate between relational and non-relational database use cases and caching strategies.
• Design high-performing network architectures using VPC, Route 53, and Global Accelerator.
• Apply architectural patterns for decoupling and scalability.

Key Terms & Glossary

• RPO (Recovery Point Objective): The maximum acceptable amount of data loss measured in time.
• RTO (Recovery Time Objective): The maximum acceptable downtime for a service after a failure.
• IOPS (Input/Output Operations Per Second): A common performance measurement used to benchmark computer storage devices like EBS.
• Decoupling: An architectural pattern where components remain autonomous so that a failure in one does not cause a total system collapse.
• Serverless: A cloud computing execution model where the provider runs the server and dynamically manages the allocation of machine resources.

The "Big Idea"

Selecting the right AWS technology is not about finding the "best" service, but the best fit for the workload's "personality." Workloads are typically characterized as compute-oriented, storage-focused, or memory-driven. A Solutions Architect must map these technical requirements to the specific performance tiers and cost structures of AWS services to ensure the architecture is efficient, scalable, and resilient.

Formula / Concept Box

Disaster Recovery (DR) Strategies

[!IMPORTANT] High-bandwidth EC2 instances can support enhanced networking (ENA/EFA) reaching speeds up to 100 Gbps for high-performance computing (HPC) workloads.

Hierarchical Outline

1. Compute Solutions
   • Amazon EC2: Virtual machines for full control; various instance types (Compute Optimized, Memory Optimized).
   • AWS Lambda: Serverless, event-driven functions; best for short-lived, bursty tasks.
   • Containers (ECS/EKS/Fargate): Orchestration for microservices; Fargate provides a serverless container experience.
   • Specialized Compute: AWS Batch (large-scale batch jobs), Amazon EMR (Big Data/Hadoop).
2. Storage Solutions
   • Amazon S3: Object storage; high durability (99.999999999%); used for data lakes and backups.
   • Amazon EBS: Block storage for EC2; high performance; persistent volumes.
   • Amazon EFS: Managed File System; supports concurrent access for multiple EC2 instances.
3. Database Technologies
   • Amazon RDS: Relational (SQL); supports Multi-AZ for HA and Read Replicas for performance.
   • Amazon DynamoDB: NoSQL; key-value pair; single-digit millisecond latency at scale.
   • Amazon ElastiCache: In-memory caching (Redis/Memcached) for read-intensive workloads.
4. Networking & Content Delivery
   • Route 53: DNS with latency-based and geolocation routing.
   • CloudFront: CDN for caching content at edge locations to reduce latency.
   • VPC Endpoints: Private connection to AWS services without traversing the public internet.

Visual Anchors

Compute Selection Logic

High Availability Architecture (Multi-AZ)

\begin{tikzpicture} [node distance=2cm, box/.style={rectangle, draw, minimum width=2.5cm, minimum height=1cm, align=center}]

\node (ALB) [box, fill=orange!20] {Application\Load Balancer};

\draw [dashed] (-4,-1) -- (4,-1); \node at (4.5, -0.8) {Public Subnet};

\node (AZ1) [box, below left of=ALB, xshift=-1cm, fill=blue!10] {EC2 Instance$AZ A)}; \node (AZ2) [box, below right of=ALB, xshift=1cm, fill=blue!10] {EC2 Instance$AZ B)};

\node (RDS) [box, below of=ALB, yshift=-3cm, fill=green!10] {Amazon RDS$Primary + Standby)};

\draw [->, thick] (ALB) -- (AZ1); \draw [->, thick] (ALB) -- (AZ2); \draw [->, thick] (AZ1) -- (RDS); \draw [->, thick] (AZ2) -- (RDS);

\node at (0, -6) {\textbf{Multi-AZ Architecture}}; \end{tikzpicture}

Definition-Example Pairs

• Amazon S3 (Object Storage): Storage for discrete files (objects) that include metadata and a unique identifier.
  • Example: Storing millions of user profile images for a social media application.
• Amazon EBS (Block Storage): Storage that treats data as blocks within sectors; used as a hard drive for an operating system.
  • Example: A boot volume for a Windows EC2 instance running a legacy accounting software.
• AWS Lambda (Serverless Compute): Running code without provisioning servers; billed only for execution time.
  • Example: Automatically resizing an image immediately after it is uploaded to an S3 bucket.
• Amazon CloudFront (CDN): A web service that speeds up distribution of static and dynamic web content.
  • Example: A global video streaming platform using edge locations to cache movie files closer to users in different countries.

Worked Examples

Example 1: Read-Intensive E-commerce Site

Scenario: An e-commerce site experiences performance lag during sales because the relational database (RDS) is overwhelmed by product catalog queries.

• Requirement: Improve read performance without migrating from SQL.
• Solution:
  1. Implement Amazon RDS Read Replicas to offload read traffic from the primary instance.
  2. Deploy Amazon ElastiCache (Redis) to cache frequently accessed product details in memory.

Example 2: Massive Batch Data Processing

Scenario: A financial firm needs to process 10TB of data every night between 2 AM and 4 AM. The process is computationally heavy but can be interrupted.

• Requirement: Minimize cost while completing the task within the 2-hour window.
• Solution: Use Amazon EC2 Spot Instances managed by AWS Batch. Spot instances provide up to 90% savings, and AWS Batch will handle the distribution of the 10TB workload across a fleet of instances.

Checkpoint Questions

1. Which AWS storage service provides a managed file system that can be mounted simultaneously by hundreds of EC2 instances?
2. What is the difference between an AWS Region and an Availability Zone?
3. For a serverless application that needs to maintain a stateful session but scale infinitely, which database is the best choice?
4. Which networking service allows for private, dedicated connectivity from an on-premises data center to AWS?
5. How does a "Warm Standby" DR strategy differ from a "Pilot Light" strategy?

▶Click to see answers

1. Amazon EFS (Elastic File System).
2. A Region is a geographical area; an Availability Zone (AZ) is one or more discrete data centers within a region with redundant power and networking.
3. Amazon DynamoDB (often used with DynamoDB Accelerator/DAX).
4. AWS Direct Connect.
5. A Pilot Light keeps only the data live (the "pilot light" of the furnace), while the Warm Standby keeps a scaled-down but functional version of the entire application running at all times.