Curriculum Overview: High Availability and Scalability

This curriculum provides a foundational understanding of how cloud computing ensures that applications remain accessible and performant under varying loads. As a core component of the AZ-900 Microsoft Azure Fundamentals path, this module focuses on the pillars of cloud reliability.

Prerequisites

Before starting this module, learners should possess a basic understanding of the following:

• Basic Cloud Concepts: Understanding what cloud computing is (on-demand delivery of compute, database, and storage).
• Shared Responsibility Model: Awareness that security and management are shared between the provider (Azure/AWS/GCP) and the consumer.
• Computing Basics: Familiarity with Virtual Machines (VMs), IP addresses, and the concept of an application "outage."

Module Breakdown

Learning Objectives per Module

Module 1: Defining Availability

• Define High Availability (HA) and its role in maintaining 24/7 access.
• Identify common causes of downtime: network outages, system failures, and power loss.
• Understand the purpose of Redundancy in a cloud environment.

Module 2: Scaling Strategies

• Distinguish between Vertical Scaling (scaling up) and Horizontal Scaling (scaling out).
• Identify use cases for manual vs. automatic scaling.

Module 3: Service Level Agreements (SLAs)

• Describe how cloud providers guarantee uptime as a percentage (e.g., 99.9%).
• Learn how to calculate potential downtime based on different SLA tiers.

Module 4: Reliability & Predictability

• Explain how cloud infrastructure handles failures without data loss (Reliability).
• Understand how Predictability applies to both performance (consistent latency) and cost (budget tracking).

Module 5: Automation & Autoscale

• Describe the function of Autoscale in managing costs and performance dynamically.
• Identify tools like Application Insights for proactive monitoring.

Success Metrics

To demonstrate mastery of this curriculum, the learner must be able to:

1. Differentiate Scaling: Explain why a web application experiencing a traffic spike would benefit more from horizontal scaling than vertical scaling.
2. Calculate Uptime: Given a 99.9% SLA, identify that the maximum allowable downtime per year is approximately 8.77 hours.
3. Identify HA Components: List at least three infrastructure components (e.g., Region Pairs, Availability Zones) that contribute to high availability.
4. Scenario Analysis: Match a business problem (e.g., "Our server crashes every Friday at 5 PM") to the correct cloud solution (Autoscale).

Real-World Application

In the professional landscape, high availability and scalability are not just technical features—they are business requirements.

Visualizing High Availability

The following diagram illustrates a standard High Availability architecture where traffic is distributed across multiple instances to prevent a single point of failure.

\begin{tikzpicture}[node distance=2cm, every node/.style={draw, fill=blue!10, rounded corners, minimum width=2.5cm, minimum height=1cm, align=center}] % Nodes \node (user) [fill=green!10] {User Traffic}; \node (lb) [below of=user] {Load Balancer}; \node (vm1) [below left of=lb, xshift=-1cm] {Web Server A}; \node (vm2) [below right of=lb, xshift=1cm] {Web Server B}; \node (db) [below of=lb, yshift=-2.5cm, fill=orange!10] {Redundant Database};

% Connections \draw [->, thick] (user) -- (lb); \draw [->, thick] (lb) -- (vm1); \draw [->, thick] (lb) -- (vm2); \draw [->, thick] (vm1) -- (db); \draw [->, thick] (vm2) -- (db); \end{tikzpicture}

Case Studies:

• E-commerce (Black Friday): During massive traffic spikes, a company uses Scalability to add 100 temporary servers automatically, then deletes them when the sale ends to save money.
• Banking Systems: A bank requires High Availability across different geographic regions. If a data center in London loses power, the system immediately fails over to a data center in Dublin, ensuring customers can still access their funds.
• SaaS Startups: Utilizing Manageability tools like Azure Advisor or Application Insights allows a small team to monitor thousands of users without needing a massive on-premise IT department.

[!IMPORTANT] High availability focuses on uptime (staying online), while scalability focuses on capacity (handling the load). A system can be highly available but fail to scale, leading to slow performance for users.