Amazon AWS Certified Solutions Architect - Professional SAP-C02 Bundle

For the separate situation A company wants all new services to synchronously call one centralized orchestrator for simple notifications, Prefer SNS fanout or EventBridge pub-sub for independent subscriber notifications could be useful. Asynchronous fanout avoids unnecessary runtime coupling through a central synchronous service.

Introduce an SQS queue and idempotent inventory consumers solves the related scenario A monolith holds incoming orders synchronously while a slow inventory system responds; queued work absorbs dependency latency and separates acceptance from processing.

Although Assign data ownership and introduce a versioned API or events between the services can be valid when A migration extracts customer profiles into a new service but shared database tables cause accidental cross-service writes, it targets a different technical condition. Service isolation requires explicit state ownership rather than shared mutable tables.

Deploy MSK with network, identity and broker compatibility planning would address a different requirement: A modernized service must consume Kafka events using its unchanged Kafka producers and consumers. Managed Kafka can preserve existing event interfaces while reducing cluster operations.

This alternative, Export historic telemetry into partitioned Parquet datasets on S3 and query with Athena, is appropriate when A legacy relational telemetry store is retained indefinitely although compliance auditors query historical sensor data only once a month. Columnar object-lake analytics separates low-frequency history queries from continuously provisioned OLTP databases.

For the separate situation A microservice needs application-level read caching without maintaining a self-managed Redis server, Use ElastiCache with appropriate replication and cache key governance could be useful. Managed caches reduce maintenance while providing low-latency hot-data access.

Consider DynamoDB for the key-value portion with explicit access-pattern design would address a different requirement: A legacy job performs frequent exact-key lookups but spends most time waiting on a relational server. Purpose-built key-value storage avoids unnecessary relational query overhead for point lookups.

This alternative, Add OpenSearch indexing for suitable search queries while retaining the system of record, is appropriate when An analytics platform's ad-hoc searches are much faster with an inverted index than relational wildcard scans. Dedicated search indices offload text search from transactional storage.

For the separate situation A modernized service must consume Kafka events using its unchanged Kafka producers and consumers, Deploy MSK with network, identity and broker compatibility planning could be useful. Managed Kafka can preserve existing event interfaces while reducing cluster operations.

This alternative, Add OpenSearch indexing for suitable search queries while retaining the system of record, is appropriate when An analytics platform's ad-hoc searches are much faster with an inverted index than relational wildcard scans. Dedicated search indices offload text search from transactional storage.

Deploy MSK with network, identity and broker compatibility planning solves the related scenario A modernized service must consume Kafka events using its unchanged Kafka producers and consumers; managed Kafka can preserve existing event interfaces while reducing cluster operations.

Although Export historic telemetry into partitioned Parquet datasets on S3 and query with Athena can be valid when A legacy relational telemetry store is retained indefinitely although compliance auditors query historical sensor data only once a month, it targets a different technical condition. Columnar object-lake analytics separates low-frequency history queries from continuously provisioned OLTP databases.

For the separate situation A microservice needs application-level read caching without maintaining a self-managed Redis server, Use ElastiCache with appropriate replication and cache key governance could be useful. Managed caches reduce maintenance while providing low-latency hot-data access.

Consider DynamoDB for the key-value portion with explicit access-pattern design solves the related scenario A legacy job performs frequent exact-key lookups but spends most time waiting on a relational server; purpose-built key-value storage avoids unnecessary relational query overhead for point lookups.

Although Export historic telemetry into partitioned Parquet datasets on S3 and query with Athena can be valid when A legacy relational telemetry store is retained indefinitely although compliance auditors query historical sensor data only once a month, it targets a different technical condition. Columnar object-lake analytics separates low-frequency history queries from continuously provisioned OLTP databases.

Although Consider DynamoDB for the key-value portion with explicit access-pattern design can be valid when A legacy job performs frequent exact-key lookups but spends most time waiting on a relational server, it targets a different technical condition. Purpose-built key-value storage avoids unnecessary relational query overhead for point lookups.

Export historic telemetry into partitioned Parquet datasets on S3 and query with Athena would address a different requirement: A legacy relational telemetry store is retained indefinitely although compliance auditors query historical sensor data only once a month. Columnar object-lake analytics separates low-frequency history queries from continuously provisioned OLTP databases.

This alternative, Add OpenSearch indexing for suitable search queries while retaining the system of record, is appropriate when An analytics platform's ad-hoc searches are much faster with an inverted index than relational wildcard scans. Dedicated search indices offload text search from transactional storage.

Run load and concurrency tests and bound serverless concurrency before cutover would address a different requirement: An organization moves a batch workflow from servers to Lambda without validating downstream connection limits. Serverless scalability can overwhelm downstream services unless controlled.

This alternative, Build versioned IaC, observability and progressive deployment before promoting the service, is appropriate when A new container service works in development but production lacks a reproducible infrastructure baseline. Repeatable environments and measured rollout are prerequisites for safe modernization.

For the separate situation A legacy application remains available while users are gradually shifted to a replacement service, Use feature flags or weighted routing with compatible data contracts and monitored rollback could be useful. Incremental routing allows controlled validation while old and new services coexist.

This alternative, Build versioned IaC, observability and progressive deployment before promoting the service, is appropriate when A new container service works in development but production lacks a reproducible infrastructure baseline. Repeatable environments and measured rollout are prerequisites for safe modernization.

For the separate situation An application-team proposal removes all audit trails to reduce migration overhead, Preserve CloudTrail, application audit and retention requirements in the modernized platform could be useful. Modernization cannot discard accountability and compliance obligations.

Although Apply the strangler-fig pattern with incremental routing to independently tested services can be valid when A large rewrite would take two years and risks preventing a business-critical release, it targets a different technical condition. Incremental extraction delivers value and retains a rollback path per capability.

For the separate situation An application-team proposal removes all audit trails to reduce migration overhead, Preserve CloudTrail, application audit and retention requirements in the modernized platform could be useful. Modernization cannot discard accountability and compliance obligations.

Apply the strangler-fig pattern with incremental routing to independently tested services solves the related scenario A large rewrite would take two years and risks preventing a business-critical release; incremental extraction delivers value and retains a rollback path per capability.

Use feature flags or weighted routing with compatible data contracts and monitored rollback would address a different requirement: A legacy application remains available while users are gradually shifted to a replacement service. Incremental routing allows controlled validation while old and new services coexist.

Run load and concurrency tests and bound serverless concurrency before cutover solves the related scenario An organization moves a batch workflow from servers to Lambda without validating downstream connection limits; serverless scalability can overwhelm downstream services unless controlled.

Although Build versioned IaC, observability and progressive deployment before promoting the service can be valid when A new container service works in development but production lacks a reproducible infrastructure baseline, it targets a different technical condition. Repeatable environments and measured rollout are prerequisites for safe modernization.

This alternative, Preserve CloudTrail, application audit and retention requirements in the modernized platform, is appropriate when An application-team proposal removes all audit trails to reduce migration overhead. Modernization cannot discard accountability and compliance obligations.

Although Run load and concurrency tests and bound serverless concurrency before cutover can be valid when An organization moves a batch workflow from servers to Lambda without validating downstream connection limits, it targets a different technical condition. Serverless scalability can overwhelm downstream services unless controlled.

This alternative, Build versioned IaC, observability and progressive deployment before promoting the service, is appropriate when A new container service works in development but production lacks a reproducible infrastructure baseline. Repeatable environments and measured rollout are prerequisites for safe modernization.

For the separate situation A large rewrite would take two years and risks preventing a business-critical release, Apply the strangler-fig pattern with incremental routing to independently tested services could be useful. Incremental extraction delivers value and retains a rollback path per capability.