4 of the 7 questions we have for this objective, shown with the answer and the reasoning.
Select existing workloads and processes for potential migration
Ombra Foods created its first AWS Organizations organization in March 2026 and plans to move 480 servers running on VMware and Hyper-V to Amazon EC2. Before any migration, the team must collect server inventory from several sources, group servers into applications, identify dependencies between them, and produce prioritized migration waves. The team wants an AWS service that covers discovery and planning and can continue into rehosting.
Which AWS service should the solutions architect use?
- AAWS Migration Hub with AWS Application Discovery Service agentsMigration Hub and Application Discovery Service stopped accepting new customers on November 7, 2025. An organization created in 2026 cannot start using them and is directed to AWS Transform.
- BAWS Transform, using its migration capabilities for discovery and migration planningAWS Transform collects and analyzes server inventory from multiple sources, groups applications, identifies dependencies, and builds prioritized waves, and it continues into landing zone, network, and rehost steps to EC2.
- CAWS Compute OptimizerCompute Optimizer recommends sizes for resources that already run in AWS. It does not discover on-premises servers or plan migration waves.
- DAWS Trusted AdvisorTrusted Advisor checks existing AWS accounts for cost, security, fault tolerance, and quota issues. It has no on-premises discovery or wave planning.
Why: An application migration assessment establishes what exists, how it is connected, and in what order it should move. For years that role was filled by Migration Hub and Application Discovery Service, and many exam resources still name them, but both closed to new customers in November 2025 and AWS recommends AWS Transform, which covers discovery, dependency analysis, wave planning, and execution in one experience. Existing customers of the older services can still complete their projects, so check the account's history before choosing. Compute Optimizer and Trusted Advisor operate only on resources already in AWS.
Select existing workloads and processes for potential migration
Ferrous Manufacturing runs 340 VMs on VMware vSphere in a data center whose lease ends in 10 weeks. The operations team relies on vCenter, existing VMware backup tooling, and VMware-specific runbooks, and it does not want to change operating models during the move. Application owners have not had time to assess their workloads, and any modernization will be planned after the exit.
Which migration strategy best meets these requirements?
- ARelocate the VMs to a VMware environment on AWS, such as Amazon Elastic VMware Service, by using VMware migration tooling.Relocate moves VMs at the hypervisor level into a VMware environment on AWS, so the team keeps vCenter, tooling, and runbooks while meeting the 10-week deadline without touching each workload.
- BRehost each VM to Amazon EC2 with AWS Transform MGN (formerly AWS Application Migration Service).Rehosting to EC2 is fast, but it changes the operating model from vSphere to EC2, which the team explicitly wants to avoid during the move.
- CReplatform the VMs into containers on Amazon ECS.Containerizing 340 unassessed workloads in 10 weeks is unrealistic, and it is a modernization step that the company has deferred until after the exit.
- DRetain the VMs on premises and negotiate a lease extension.Retain keeps workloads in place, which does not meet the requirement to vacate the data center when the lease ends.
Why: Relocate is the strategy for moving a large number of servers from an on-premises virtualization platform to the cloud version of the same platform without buying hardware or rewriting applications. Its value is speed combined with an unchanged operating model, which is exactly what a hard data center deadline and a VMware-centric operations team call for. Rehost to EC2 is the right choice when the organization is willing to adopt native EC2 operations; here the stated constraint rules it out for now.
Select existing workloads and processes for potential migration
Marisol Dental has a fully defined target design for a small migration: 12 EC2 instances of known types, two Amazon RDS for PostgreSQL instances, 8 TB of Amazon S3 storage, and an Application Load Balancer in one Region. The finance manager wants a shareable monthly and annual cost estimate for this design, comparing On-Demand with a 1-year Savings Plan, to place next to the on-premises costs in a TCO spreadsheet.
Which tool should the solutions architect use?
- AAWS Cost ExplorerCost Explorer analyzes and forecasts spend that is already occurring in AWS. It cannot price a design that has not been deployed.
- BAWS Pricing CalculatorPricing Calculator builds estimates for planned resources, supports purchase options such as Savings Plans, and produces a shareable estimate that finance can compare with on-premises costs.
- CAWS BudgetsBudgets tracks actual or forecast spend against a threshold and sends alerts. It does not produce an estimate for a future design.
- DAWS Compute OptimizerCompute Optimizer recommends rightsizing for resources that already run in AWS based on their utilization history.
Why: A TCO comparison needs two inputs: the full on-premises cost, including hardware refresh, facilities, power, licenses, and staff time, and a credible AWS cost. When the AWS architecture is already defined, AWS Pricing Calculator is the right tool for the AWS side because it prices specific services and purchase options and generates a shareable estimate. For large estates where sizes are not yet known, Migration Evaluator derives sizes from measured utilization instead. Cost Explorer, Budgets, and Compute Optimizer all depend on usage that already exists.
Select existing workloads and processes for potential migration
Calloway Bank is classifying its application portfolio before a data center exit. The assessment found: an internal reporting tool with no logins in 14 months; a trading system that must stay within 0.5 ms of an exchange matching engine located in the same colocation facility; a customer portal on Java 8 and WebLogic whose product team wants independent deployments and elastic scaling for a new mobile channel; and a document archive on Windows Server whose owner wants the cheapest fast move with no code changes.
Which three strategy assignments are appropriate? Each correct answer presents a complete classification.
- ARetire the internal reporting tool.An application with no usage for over a year is a candidate for decommissioning after owner confirmation, which removes its cost and migration effort entirely.
- BRetain the trading system on premises.The sub-millisecond latency requirement to an exchange in the same facility is a business constraint that keeps this workload where it is for now.
- CRehost the customer portal to Amazon EC2 without changes.Rehosting would move the portal quickly but keep the monolith, so it would not deliver the independent deployments and elastic scaling the product team requires.
- DRefactor the customer portal into services on containers or serverless components.Refactor is justified when there is a strong business driver for new capabilities such as independent deployments and elasticity, which the new mobile channel provides.
- ERepurchase the document archive as a SaaS content platform.Moving to a new product requires data migration, licensing, and user change, which is not the cheapest fast move with no changes. Rehosting the Windows server fits that requirement.
Why: The seven strategies are chosen per application from evidence gathered in the assessment, not applied uniformly. Lack of use points to retire, hard latency or dependency constraints point to retain, and a strong business case for agility points to refactor despite its cost. The archive's owner wants speed and no change, which points to rehost rather than repurchase. The trap is to rehost everything for speed, which ignores the stated business drivers for individual applications.