4.1 - Planning and assessing migrations (Discovery, Migration Hub, 7 Rs)
AWS Solutions Architect Professional objective 4.1 covers planning and assessing a migration before it starts. AWS Application Discovery Service inventories on-premises servers and maps dependencies, using the Discovery Agent for deep process and network detail per host or the Agentless Collector in vCenter for hypervisor-level inventory. AWS Migration Hub tracks migration status centrally and adds Strategy Recommendations, while Migration Evaluator (formerly TSO Logic) projects AWS run cost and right-sizes to justify the business case. You classify each application into a 7 Rs strategy - rehost with AWS Application Migration Service when speed of exit drives, repurchase a commodity tool as SaaS, refactor an app needing cloud-native scaling - and you prepare the destination by setting up a Control Tower landing zone with preventive SCP guardrails. Direct Connect supports an interim hybrid state where a web tier runs in AWS while the database stays on-premises. Expect scenarios about inventory and dependency mapping, a TCO business case, or picking a 7 Rs strategy, and ask which service or approach fits.
Inventory servers and map dependencies = Application Discovery Service (Agent for deep detail, Agentless in vCenter). TCO and projected AWS cost for the business case = Migration Evaluator. Speed of data-center exit, no code change = rehost with MGN.
Practice questions
1. Before migrating hundreds of on-premises servers, a company needs data on utilization, dependencies between servers, and right-sizing recommendations to plan waves. Which service provides this discovery?
- AWS Trusted Advisor checks run on the existing AWS account
- Amazon Inspector vulnerability scans of the running workloads
- AWS Application Discovery Service (with Migration Hub) (correct answer)
- AWS Config aggregators collecting resource configuration
Application Discovery Service collects on-premises utilization and network dependency data, feeding Migration Hub to plan and track migration waves and right-sizing. Trusted Advisor checks existing AWS accounts, Inspector scans for vulnerabilities, and Config tracks AWS resource configuration, not on-premises discovery.
2. A company has 300 servers and a hard 6-month deadline to exit its data center. Most apps are stable and rarely change. Leadership asks whether to refactor everything to serverless during the move. What should the architect advise?
- Refactor all 300 servers to Lambda and containers before any of them can move
- Delay the deadline indefinitely until every application has been fully rewritten
- Rehost most servers now to hit the deadline, then modernize selected apps afterward (correct answer)
- Retire the data center only after each workload passes a full microservices redesign
With a hard deadline and mostly stable apps, rehosting (lift-and-shift) first meets the timeline and lowest risk, and modernization is done later where it pays off. Refactoring everything up front, delaying indefinitely, or gating exit on full redesigns all blow the deadline for little immediate benefit.
3. Before migrating a data center, an architect must inventory on-premises servers and map how applications depend on each other so migration waves can be planned. Which AWS service is purpose-built for this?
- AWS Config recording resource configuration changes
- AWS Application Discovery Service collecting inventory and dependencies (correct answer)
- Amazon Inspector continuously scanning running workloads for software vulnerabilities and exposures
- AWS Trusted Advisor reviewing account best practices
Application Discovery Service gathers server specs, utilization, and network dependencies on-premises so you can group apps into migration waves. Config tracks AWS resources (not on-prem discovery), Inspector finds vulnerabilities, and Trusted Advisor reviews existing AWS accounts.
4. During portfolio assessment, discovery data shows several servers with near-zero CPU, no active users, and no dependencies for the past year. What is the recommended disposition?
- Rehost them to EC2 so nothing is lost during the migration
- Relocate them to VMware Cloud on AWS to preserve the hypervisor
- Retire them instead of migrating unused workloads (correct answer)
- Replatform them onto managed services to lower running cost
Servers with no usage, users, or dependencies should be retired, removing cost and migration effort. Rehosting, relocating, or replatforming an idle workload just carries dead weight into AWS and keeps paying for it.
5. A migration program needs a single pane of glass to track application migration status across tools and accounts, plus a projected AWS cost estimate to justify the business case. Which TWO services provide these? (Choose TWO.)
- AWS Migration Hub to track migration status centrally (correct answer)
- AWS Cost Explorer forecasting future spend on resources that are already running in the account
- Migration Evaluator (TSO Logic) to project AWS cost and right-size (correct answer)
- AWS Budgets alerting when current spend exceeds a threshold
Migration Hub aggregates status from MGN, DMS, and partner tools into one view, while Migration Evaluator builds a data-driven cost projection and right-sizing to justify the move. Cost Explorer and Budgets analyze existing AWS spend, not pre-migration business cases.
6. When sequencing a portfolio migration, which type of application is the best candidate for the first wave to build momentum and reduce risk?
- The revenue-critical system with the most upstream and downstream dependencies
- A tightly coupled application that shares storage with many others
- A regulated workload awaiting a compliance sign-off from auditors
- A low-risk application with few dependencies and a clear owner (correct answer)
Starting with a low-risk, loosely coupled app that has a clear owner lets the team learn the tooling and prove the process with minimal blast radius. The most-connected, tightly coupled, or compliance-blocked apps carry high risk and are better sequenced into later waves.