AWS Cloud Practitioner Essentials T3.3

AWS Compute Services Overview

• AWS provides compute services for running applications and workloads, offering flexibility, scalability and efficiency.

Amazon EC2 (Elastic Compute Cloud)

• EC2 provides virtual servers (instances) for running applications in the cloud.

• Instance types vary based on needs (CPU, memory, storage).

• Pricing is pay-as-you-go: pay only for instance usage.

  • General Purpose: Balanced for CPU, memory and networking; suitable for web servers (t3, m5 families).
  • Compute Optimized: High-performance processors for CPU-intensive tasks (like machine learning) (c5 family).
  • Memory Optimized: More memory for large datasets or in-memory databases (r5 family).
  • Storage Optimized: Optimized for high-speed or large storage needs (i3 family).
  • GPU Instances: Accelerate tasks like deep learning (p3 family).

• EC2 is best for applications needing full control over virtual machines.

Auto Scaling

• Auto Scaling automatically adjusts the number of EC2 instances based on demand.
• Maintains high availability and saves costs.
• Suitable for fluctuating workloads (like e-commerce during sales).

Elastic Load Balancers (ELB)

• Load Balancers distribute incoming traffic across multiple servers (e.g., EC2 instances).

• Keeps applications available and handles large traffic.

  • Application Load Balancer (ALB): Works at the application layer (HTTP/HTTPS), routing traffic based on URLs.
  • Network Load Balancer (NLB): Works at the network layer (TCP/UDP), ideal for high-performance, low-latency traffic.
  • Classic Load Balancer (CLB): A legacy option for basic load balancing.

• ELBs are crucial for distributing traffic for high availability and performance.

Container Services

• AWS offers services to manage and deploy containerized applications.

  • Amazon ECS (Elastic Container Service): Fully managed container orchestration for Docker containers.
  • Amazon EKS (Elastic Kubernetes Service): Managed Kubernetes service for container orchestration (advanced scaling and deployment).
  • AWS Fargate: Serverless compute engine for running containers without managing servers.

• ECS is suitable for simpler container needs, EKS offers advanced Kubernetes management, and Fargate avoids server management.

Serverless Compute

• AWS serverless compute (no infrastructure management) options:

  • AWS Lambda: Executes code in response to events (like HTTP requests), paying only for used compute time.
  • AWS Fargate: Runs containers without server management.

• Lambda is best for event-driven applications, while Fargate handles container workloads without infrastructure management.

Key Differences Between Compute Options

• A summarized comparison of compute services highlighting their key benefits and best use cases. (See table in original text).

Real-World Scenarios

• E-commerce: Use EC2, Auto Scaling, and ELB to handle fluctuating traffic.
• Microservices: Use ECS or EKS for container orchestration.
• Weather App: Use AWS Lambda for hourly scheduled data updates.

Key Takeaways

• Choose EC2 instance types based on workload requirements.
• Auto Scaling adjusts resources automatically.
• Load Balancers distribute traffic efficiently.
• Choose container services (ECS, EKS, Fargate) based on needs.
• Serverless options (Lambda, Fargate) are suitable for applications without infrastructure management.

Explore the various AWS compute services, focusing on Amazon EC2. Learn about the different instance types available for specific workloads including general-purpose, compute-optimized, and memory-optimized instances. Understand the fundamentals of auto-scaling and the benefits of using AWS for application deployment.