Amazon AWS Certified Solutions Architect - Associate SAA-C03 Bundle

The option Move analytical queries to suitable replicas or warehouse as warranted addresses a separate condition: analytics scans repeatedly slow the OLTP RDS primary during peak hours. Read workload separation reduces transactional contention. The actual requirement is a large legacy RDS instance has more storage and CPU than the dataset needs. Recommended corrective action: Measure query demand and migrate to a smaller suitable instance. Database right sizing aligns resources with workload.

Apply snapshot retention governance and delete approved obsolete copies would help if retired snapshots from an old RDS database remain indefinitely. Unneeded snapshots keep incurring storage cost. That is not the symptom described here: a large legacy RDS instance has more storage and CPU than the dataset needs. The matching response is Measure query demand and migrate to a smaller suitable instance, because database right sizing aligns resources with workload.

For the alternative failure a read replica exists solely for a report run once monthly, the control Start/stop or choose a cost-appropriate analytic extract instead where supported is relevant: infrequent reports need not require continuous reader capacity. Here the evidence instead is a large legacy RDS instance has more storage and CPU than the dataset needs. The selected technical mechanism is Measure query demand and migrate to a smaller suitable instance: database right sizing aligns resources with workload.

Use DynamoDB with an appropriate partition-key schema would help if a shopping-cart state model uses mostly primary-key reads and writes. DynamoDB suits high-scale key-value access. That is not the symptom described here: an app needs strictly relational transactions, joins and foreign-key logic. The matching response is Use a supported RDS/Aurora relational engine, because relational engines support transactional relational workloads.

Use Redshift or a suitable analytical warehouse solves a different operational problem where long historical SQL aggregations dominate queries over transactional row operations. Columnar analytics systems are designed for large scans. It does not directly address an app needs strictly relational transactions, joins and foreign-key logic. The case calls for Use a supported RDS/Aurora relational engine, since relational engines support transactional relational workloads.

The competing option Use Amazon OpenSearch Service when search is primary would be useful when event search requires relevance ranking and full-text filtering. Search indexes serve term and text relevance. The present issue is an app needs strictly relational transactions, joins and foreign-key logic; it is more directly addressed by Use a supported RDS/Aurora relational engine. Relational engines support transactional relational workloads.

Use Amazon OpenSearch Service when search is primary solves a different operational problem where event search requires relevance ranking and full-text filtering. Search indexes serve term and text relevance. It does not directly address a shopping-cart state model uses mostly primary-key reads and writes. The case calls for Use DynamoDB with an appropriate partition-key schema, since dynamoDB suits high-scale key-value access.

The competing option Evaluate Amazon Neptune for the graph workload would be useful when a graph traversal finds friends-of-friends across millions of relationships. Graph databases optimize connected relationships. The present issue is a shopping-cart state model uses mostly primary-key reads and writes; it is more directly addressed by Use DynamoDB with an appropriate partition-key schema. DynamoDB suits high-scale key-value access.

The option Use Redshift or a suitable analytical warehouse addresses a separate condition: long historical SQL aggregations dominate queries over transactional row operations. Columnar analytics systems are designed for large scans. The actual requirement is a shopping-cart state model uses mostly primary-key reads and writes. Recommended corrective action: Use DynamoDB with an appropriate partition-key schema. DynamoDB suits high-scale key-value access.

Use Redshift or a suitable analytical warehouse solves a different operational problem where long historical SQL aggregations dominate queries over transactional row operations. Columnar analytics systems are designed for large scans. It does not directly address a graph traversal finds friends-of-friends across millions of relationships. The case calls for Evaluate Amazon Neptune for the graph workload, since graph databases optimize connected relationships.

The option Use DynamoDB with an appropriate partition-key schema addresses a separate condition: a shopping-cart state model uses mostly primary-key reads and writes. DynamoDB suits high-scale key-value access. The actual requirement is a graph traversal finds friends-of-friends across millions of relationships. Recommended corrective action: Evaluate Amazon Neptune for the graph workload. Graph databases optimize connected relationships.

Use a supported RDS/Aurora relational engine would help if an app needs strictly relational transactions, joins and foreign-key logic. Relational engines support transactional relational workloads. That is not the symptom described here: a graph traversal finds friends-of-friends across millions of relationships. The matching response is Evaluate Amazon Neptune for the graph workload, because graph databases optimize connected relationships.

The competing option Use a supported RDS/Aurora relational engine would be useful when an app needs strictly relational transactions, joins and foreign-key logic. Relational engines support transactional relational workloads. The present issue is event search requires relevance ranking and full-text filtering; it is more directly addressed by Use Amazon OpenSearch Service when search is primary. Search indexes serve term and text relevance.

The option Use Redshift or a suitable analytical warehouse addresses a separate condition: long historical SQL aggregations dominate queries over transactional row operations. Columnar analytics systems are designed for large scans. The actual requirement is event search requires relevance ranking and full-text filtering. Recommended corrective action: Use Amazon OpenSearch Service when search is primary. Search indexes serve term and text relevance.

Use DynamoDB with an appropriate partition-key schema would help if a shopping-cart state model uses mostly primary-key reads and writes. DynamoDB suits high-scale key-value access. That is not the symptom described here: event search requires relevance ranking and full-text filtering. The matching response is Use Amazon OpenSearch Service when search is primary, because search indexes serve term and text relevance.

Use a supported RDS/Aurora relational engine would help if an app needs strictly relational transactions, joins and foreign-key logic. Relational engines support transactional relational workloads. That is not the symptom described here: long historical SQL aggregations dominate queries over transactional row operations. The matching response is Use Redshift or a suitable analytical warehouse, because columnar analytics systems are designed for large scans.

For the alternative failure a shopping-cart state model uses mostly primary-key reads and writes, the control Use DynamoDB with an appropriate partition-key schema is relevant: dynamoDB suits high-scale key-value access. Here the evidence instead is long historical SQL aggregations dominate queries over transactional row operations. The selected technical mechanism is Use Redshift or a suitable analytical warehouse: columnar analytics systems are designed for large scans.

Use Amazon OpenSearch Service when search is primary solves a different operational problem where event search requires relevance ranking and full-text filtering. Search indexes serve term and text relevance. It does not directly address long historical SQL aggregations dominate queries over transactional row operations. The case calls for Use Redshift or a suitable analytical warehouse, since columnar analytics systems are designed for large scans.

Add read replicas only if sustained reporting load merits their cost would help if analysts need separate read scaling from the production writer. Replicas increase serving capacity but incur separate resources. That is not the symptom described here: a business-critical database must fail over automatically in one region. The matching response is Use RDS Multi-AZ despite added standby cost, because availability requirements justify redundant database capacity.

For the alternative failure a low-value internal test database does not need a full-time Multi-AZ standby, the control Use a suitable Single-AZ test instance with tested backups is relevant: relaxed availability requirements permit lower-cost deployment. Here the evidence instead is a business-critical database must fail over automatically in one region. The selected technical mechanism is Use RDS Multi-AZ despite added standby cost: availability requirements justify redundant database capacity.

The competing option Enforce backup-retention policy and lifecycle cleanup would be useful when teams retain full database snapshots far beyond the required restore window. Retention controls prevent unnecessary snapshot charges. The present issue is a business-critical database must fail over automatically in one region; it is more directly addressed by Use RDS Multi-AZ despite added standby cost. Availability requirements justify redundant database capacity.

For the alternative failure the DB has a large provisioned IOPS tier but observed IOPS is low, the control Evaluate gp3 or reduced provisioned performance after load testing is relevant: overprovisioned IOPS increases storage cost. Here the evidence instead is a low-value internal test database does not need a full-time Multi-AZ standby. The selected technical mechanism is Use a suitable Single-AZ test instance with tested backups: relaxed availability requirements permit lower-cost deployment.

Enforce backup-retention policy and lifecycle cleanup solves a different operational problem where teams retain full database snapshots far beyond the required restore window. Retention controls prevent unnecessary snapshot charges. It does not directly address a low-value internal test database does not need a full-time Multi-AZ standby. The case calls for Use a suitable Single-AZ test instance with tested backups, since relaxed availability requirements permit lower-cost deployment.

The option Add read replicas only if sustained reporting load merits their cost addresses a separate condition: analysts need separate read scaling from the production writer. Replicas increase serving capacity but incur separate resources. The actual requirement is a low-value internal test database does not need a full-time Multi-AZ standby. Recommended corrective action: Use a suitable Single-AZ test instance with tested backups. Relaxed availability requirements permit lower-cost deployment.

Enforce backup-retention policy and lifecycle cleanup solves a different operational problem where teams retain full database snapshots far beyond the required restore window. Retention controls prevent unnecessary snapshot charges. It does not directly address analysts need separate read scaling from the production writer. The case calls for Add read replicas only if sustained reporting load merits their cost, since replicas increase serving capacity but incur separate resources.

The option Use a suitable Single-AZ test instance with tested backups addresses a separate condition: a low-value internal test database does not need a full-time Multi-AZ standby. Relaxed availability requirements permit lower-cost deployment. The actual requirement is analysts need separate read scaling from the production writer. Recommended corrective action: Add read replicas only if sustained reporting load merits their cost. Replicas increase serving capacity but incur separate resources.

Evaluate gp3 or reduced provisioned performance after load testing would help if the DB has a large provisioned IOPS tier but observed IOPS is low. Overprovisioned IOPS increases storage cost. That is not the symptom described here: analysts need separate read scaling from the production writer. The matching response is Add read replicas only if sustained reporting load merits their cost, because replicas increase serving capacity but incur separate resources.

The competing option Use RDS Multi-AZ despite added standby cost would be useful when a business-critical database must fail over automatically in one region. Availability requirements justify redundant database capacity. The present issue is the DB has a large provisioned IOPS tier but observed IOPS is low; it is more directly addressed by Evaluate gp3 or reduced provisioned performance after load testing. Overprovisioned IOPS increases storage cost.

Add read replicas only if sustained reporting load merits their cost would help if analysts need separate read scaling from the production writer. Replicas increase serving capacity but incur separate resources. That is not the symptom described here: the DB has a large provisioned IOPS tier but observed IOPS is low. The matching response is Evaluate gp3 or reduced provisioned performance after load testing, because overprovisioned IOPS increases storage cost.

For the alternative failure a low-value internal test database does not need a full-time Multi-AZ standby, the control Use a suitable Single-AZ test instance with tested backups is relevant: relaxed availability requirements permit lower-cost deployment. Here the evidence instead is the DB has a large provisioned IOPS tier but observed IOPS is low. The selected technical mechanism is Evaluate gp3 or reduced provisioned performance after load testing: overprovisioned IOPS increases storage cost.

Evaluate gp3 or reduced provisioned performance after load testing would help if the DB has a large provisioned IOPS tier but observed IOPS is low. Overprovisioned IOPS increases storage cost. That is not the symptom described here: teams retain full database snapshots far beyond the required restore window. The matching response is Enforce backup-retention policy and lifecycle cleanup, because retention controls prevent unnecessary snapshot charges.

For the alternative failure a low-value internal test database does not need a full-time Multi-AZ standby, the control Use a suitable Single-AZ test instance with tested backups is relevant: relaxed availability requirements permit lower-cost deployment. Here the evidence instead is teams retain full database snapshots far beyond the required restore window. The selected technical mechanism is Enforce backup-retention policy and lifecycle cleanup: retention controls prevent unnecessary snapshot charges.

Add read replicas only if sustained reporting load merits their cost solves a different operational problem where analysts need separate read scaling from the production writer. Replicas increase serving capacity but incur separate resources. It does not directly address teams retain full database snapshots far beyond the required restore window. The case calls for Enforce backup-retention policy and lifecycle cleanup, since retention controls prevent unnecessary snapshot charges.