Pass Your Amazon AWS Certified Solutions Architect - Professional SAP-C02 Exam Easy!

Forcing dormant data into Frequent Access increases storage cost and does not use the eligible optional archival tiers.

Archive Instant Access provides millisecond access and is an automatic tier, whereas optional archival tiers require explicit configuration and asynchronous restore.

Unrequested objects may still be within the mandatory retention period; inactivity is not authority to delete them.

Capacity sizing alone cannot identify interapplication dependencies or safe cutover groups, and misses license and migration labor in a defensible TCO.

A default strategy without contract and dependency evidence cannot establish feasible waves or a defensible comparative TCO.

Dependency discovery must influence wave membership and sequence, and source/target cost and eligibility still matter; sorting by age does not satisfy the combined requirements.

Dedicated Hosts can be appropriate for some BYOL contracts but do not themselves confer vendor rights or certify the deployed OS/ERP configuration.

Successful boot does not establish contractual cloud rights or support certification; validate these before selecting the target strategy.

Technical boot success is not evidence of license portability, certified deployment requirements or vendor support entitlement.

The regulated dataset and its backups/processing must satisfy residency rules, not merely provide fast reads from an approved front end.

Residency requirements apply to copies and processing, not just the primary object store, and must govern analytics placement.

Residency controls must apply to replicas and derived processing data too; an unapproved cross-Region copy may violate the stated jurisdiction boundary.

VM imaging does not reproduce physical serial hardware integration, and the unsupported OS and vendor constraints may prevent safe rehosting.

CPU compatibility is insufficient to preserve the specialized device integration or vendor/platform support required for ordinary rehosting.

Replatform requires supported compatibility and access to physical dependencies; renaming the target OS does not establish it will function.

Discovered shared services are cutover prerequisites. Leaving them unreachable strands the earlier migrated clients even if their compute launches succeed.

Moving every dependency at once magnifies correlated risk and cannot replace evidence-based readiness, transitional connectivity and sequenced cutovers.

Region preference alone cannot make shared prerequisites available; cutover order must follow real service dependencies.

EC2 compute pricing alone omits material license and transfer/people costs as well as target-specific ongoing operations, so it cannot support a comparative lifecycle TCO.

Incomplete TCO still hides material on-premises/cloud cost drivers and transition overhead; complete comparable modeled costs are essential.

Selective operating expenses and interruptible rates do not constitute realistic workload-level TCO under its support/availability and transition obligations.

An EC2 boot test establishes only operating-system startup, not the application's ability to acquire its required hardware-backed license. Without an end-to-end license handshake, a live cutover may leave a technically running but unusable finance application.

Retirement requires business authorization and evidence of no remaining needed function. An unresolved hardware-license interface does not establish that a finance system can be removed without affecting users or contractual obligations.

Disk portability and application license behavior are separate properties; an S3 copy preserves bytes but does not supply the on-premises dongle protocol, network reachability, or documented vendor support. A real end-to-end execution trial is still required.

Hypervisor inventory may support sizing and some traffic measurements, but a static VM export does not identify the source operating-system processes calling the shared licensing database. The scenario specifically requires process-level evidence on physical servers.

VPC Flow Logs can show AWS network-flow metadata after deployment but not which on-premises process generated a connection or its execution profile. The requested process-to-dependency mapping needs source-host collection before migration.

DNS names can help with endpoint identification, but DNS query logs do not show which operating-system process owns a persistent TCP session or its runtime consumption profile. The requested dependency data requires server-side discovery.

A highly critical always-on system with many dependencies is a poor place to validate an immature migration process when a lower-risk representative candidate exists. It can be scheduled after the migration factory and dependent infrastructure have been proven.

The destination's residency requirements are a pre-migration legal and architectural gate, not an issue to discover after protected data moves. This candidate remains unsuitable while the permitted location remains unresolved.

Bundling dissimilar criticality and dependencies into one unproven wave makes root-cause isolation and rollback much harder. A rationalized application portfolio, not simultaneous scheduling, establishes a sound first wave.

Broad EC2 launch permission imposes no independent contract-aware constraint on the consumption of licenses. Discovering a violation on an invoice after a prohibited instance launches does not meet the preventive-control requirement.

A tag policy can standardize supported tag names and values, but it does not meter socket entitlements or apply licensing-specific hard limits. It is useful inventory hygiene, not a substitute for the License Manager counting rules.

A detective AWS Config rule may flag missing tags after resource creation, but it does not meter license sockets or block deployment by enforcing a self-managed license limit. License Manager counting and enforcement are needed for the stated preventive check.