AWS Asset Planning & Workload Migration Study Guide

This guide covers the essential components of Domain 4: Accelerate Workload Migration and Modernization for the AWS Certified Solutions Architect - Professional (SAP-C02) exam, focusing specifically on asset planning and assessment.

Learning Objectives

By the end of this module, you will be able to:

• Perform a comprehensive Portfolio Assessment using AWS-native tools.
• Evaluate and categorize applications according to the 7Rs migration strategies.
• Formulate a Wave Plan for migration based on workload interdependencies.
• Calculate and justify Total Cost of Ownership (TCO) for cloud vs. on-premises environments.

Key Terms & Glossary

• Total Cost of Ownership (TCO): A financial estimate intended to help buyers determine the direct and indirect costs of a product or system (on-premises hardware vs. AWS consumption).
• Wave Planning: The process of grouping migration candidates into logical "waves" based on complexity, risk, and dependencies.
• Migration Hub: A central AWS service used to track the progress of application migrations across multiple AWS and partner tools.
• Application Discovery Service (ADS): An AWS tool that collects usage and configuration data from on-premises servers to help plan migrations.

The "Big Idea"

Asset planning is the foundational stage of the migration journey. It transforms a messy list of IT assets into a strategic roadmap. Without proper asset planning, migrations often stall due to "hidden dependencies" (e.g., a database shared by three apps you didn't know about) or "cost shocks" where the TCO was poorly estimated. Success in this domain relies on moving from an Infrastructure-centric view to an Application-centric view.

Formula / Concept Box

Hierarchical Outline

1. Portfolio Assessment & Discovery
   • Inventory Collection: Using AWS Application Discovery Service (Agentless or Agent-based).
   • Dependency Mapping: Identifying network connections between servers.
2. The 7Rs Strategy Selection
   • Retire: Decommissioning apps no longer needed.
   • Retain: Keeping apps on-premise (e.g., legacy mainframe).
   • Rehost: "Lift and shift" to EC2.
   • Relocate: Moving VMware workloads to VMC on AWS.
   • Repurchase: Moving to a SaaS model (e.g., Salesforce).
   • Replatform: "Lift and reshape" (e.g., Move DB to RDS without code changes).
   • Refactor/Architect: Full modernization using cloud-native features (Serverless, Containers).
3. Migration Execution Planning
   • Prioritization: Ranking apps by business impact and technical ease.
   • Data Migration: Selecting tools (DataSync, Snowball, Transfer Family).

Visual Anchors

Migration Decision Workflow

Effort vs. Business Value of Migration

\begin{tikzpicture}[scale=1.2] % Axes \draw[->] (0,0) -- (6,0) node[right] {Technical Effort}; \draw[->] (0,0) -- (0,5) node[above] {Cloud Benefit};

\end{tikzpicture}

Definition-Example Pairs

• Replatforming: Changing the underlying platform of an application to leverage cloud features without changing core code.
  • Example: Moving a self-managed Oracle database on a Windows server to Amazon RDS for Oracle to reduce patching overhead.
• Retaining: Keeping an application in its current environment, often due to compliance or high latency requirements.
  • Example: A high-frequency trading application that requires sub-millisecond local fiber access to a specific physical exchange.
• Repurchasing: Moving to a different product, typically a SaaS platform.
  • Example: Moving from a self-hosted Jenkins server to using GitHub Actions or AWS CodeBuild.

Worked Examples

Scenario: The Legacy Web App

Scenario: A company has a 10-year-old Java web app running on an old version of Tomcat. It connects to a MySQL database. Management wants to move to AWS in 4 weeks to close a data center.

Step 1: Assessment

• Use AWS Migration Hub to track the project.
• Run Application Discovery Service to find dependencies.

Step 2: Strategy Selection

• Refactor? No (Too slow, 4-week deadline).
• Choice: Replatform. Move the web tier to an Elastic Beanstalk environment and the database to Amazon RDS. This meets the deadline while removing the burden of OS/DB patching.

Step 3: Execution

• Use AWS DataSync to move static assets and a database backup to AWS.

Checkpoint Questions

1. Which of the 7Rs offers the fastest migration path for a large-scale data center evacuation?
2. What is the primary difference between the "Agentless" and "Agent-based" versions of the AWS Application Discovery Service?
3. Why is "Retire" considered a valid migration strategy in the assessment phase?
4. How does AWS Migration Hub assist in a multi-partner tool environment?

[!TIP] Answers: 1. Rehost (Lift & Shift). 2. Agentless is for VMware environments (vCenter); Agent-based is for physical servers or other hypervisors. 3. It reduces the scope of migration and saves costs by identifying unused assets. 4. It provides a single dashboard to track progress from various migration tools.

Muddy Points & Cross-Refs

• Replatform vs. Refactor: This is the most common confusion. If you are changing the way the code works (e.g., breaking a monolith into microservices), it is Refactor. If you are just changing the managed service it runs on (e.g., putting a container in Fargate instead of EC2), it is usually Replatform.
• Cross-Reference: For data movement specifics, see the Data Migration Tools module (covering Snowball Edge vs. Snowcone).

Comparison Tables

Comparing Assessment Discovery Methods