Amazon AWS Certified Solutions Architect - Professional SAP-C02 Bundle

Enable an appropriate database Multi-AZ deployment and test failover would address a different requirement: A busy database suffers one-AZ outages that require manual restore from nightly snapshots. Multi-AZ synchronous standby provides operational recovery faster than backup-only restore.

This alternative, Introduce an SQS or Kinesis buffering tier with backpressure-aware consumers, is appropriate when A core ingestion pipeline must buffer high-frequency writes during temporary database throttling. Buffering separates producer availability from short periods of downstream saturation.

For the separate situation An ALB serves three AZs but all Auto Scaling instances run in one AZ, Spread Auto Scaling capacity and target health across the supported load-balancer AZs could be useful. Multi-AZ frontend alone does not protect a single-AZ application fleet.

For the separate situation An Aurora cluster handles heavy read-only reports that compete with transaction writes, Add Aurora readers and route reports to the reader endpoint could be useful. Read replicas provide separate read capacity without changing the writer workload.

Although Optimize underlying database access and provision read capacity instead of accepting stale caches can be valid when A CPU-heavy API sends uncacheable account balance reads and cache staleness is forbidden, it targets a different technical condition. Strict freshness requirements can rule out an eventually consistent cache path.

Put Amazon CloudFront in front of the S3 origin with appropriate caching rules would address a different requirement: A globally distributed user base needs fast immutable image delivery without retrieving from S3 in one Region each time. Edge caches reduce origin reads and improve global response latency.

Optimize underlying database access and provision read capacity instead of accepting stale caches solves the related scenario A CPU-heavy API sends uncacheable account balance reads and cache staleness is forbidden; strict freshness requirements can rule out an eventually consistent cache path.

Although Add an ElastiCache read-through cache with coherent invalidation and TTL policies can be valid when A relational product catalog receives thousands of identical reads per second and database CPU is saturated, it targets a different technical condition. In-memory caching offloads repeated reads while controlling staleness.

Evaluate DynamoDB Accelerator for suitable eventually consistent read workloads would address a different requirement: A data API must accelerate repeated queries against DynamoDB with minimal cache management. DAX provides a DynamoDB-compatible in-memory read cache for supported access patterns.

Put Amazon CloudFront in front of the S3 origin with appropriate caching rules would address a different requirement: A globally distributed user base needs fast immutable image delivery without retrieving from S3 in one Region each time. Edge caches reduce origin reads and improve global response latency.

This alternative, Evaluate DynamoDB Accelerator for suitable eventually consistent read workloads, is appropriate when A data API must accelerate repeated queries against DynamoDB with minimal cache management. DAX provides a DynamoDB-compatible in-memory read cache for supported access patterns.

For the separate situation A CPU-heavy API sends uncacheable account balance reads and cache staleness is forbidden, Optimize underlying database access and provision read capacity instead of accepting stale caches could be useful. Strict freshness requirements can rule out an eventually consistent cache path.

This alternative, Put Amazon CloudFront in front of the S3 origin with appropriate caching rules, is appropriate when A globally distributed user base needs fast immutable image delivery without retrieving from S3 in one Region each time. Edge caches reduce origin reads and improve global response latency.

For the separate situation An Aurora cluster handles heavy read-only reports that compete with transaction writes, Add Aurora readers and route reports to the reader endpoint could be useful. Read replicas provide separate read capacity without changing the writer workload.

Add an ElastiCache read-through cache with coherent invalidation and TTL policies solves the related scenario A relational product catalog receives thousands of identical reads per second and database CPU is saturated; in-memory caching offloads repeated reads while controlling staleness.

This alternative, Evaluate DynamoDB Accelerator for suitable eventually consistent read workloads, is appropriate when A data API must accelerate repeated queries against DynamoDB with minimal cache management. DAX provides a DynamoDB-compatible in-memory read cache for supported access patterns.

For the separate situation A relational product catalog receives thousands of identical reads per second and database CPU is saturated, Add an ElastiCache read-through cache with coherent invalidation and TTL policies could be useful. In-memory caching offloads repeated reads while controlling staleness.

Although Put Amazon CloudFront in front of the S3 origin with appropriate caching rules can be valid when A globally distributed user base needs fast immutable image delivery without retrieving from S3 in one Region each time, it targets a different technical condition. Edge caches reduce origin reads and improve global response latency.

Index events in Amazon OpenSearch Service with suitable sharding and lifecycle solves the related scenario A log analytics service needs near-real-time full-text search over application events; openSearch supports search-oriented indexes and aggregations.

Although Use Amazon Aurora or RDS relational databases with an appropriate engine can be valid when An application requires complex SQL joins and ACID transactions over structured entities, it targets a different technical condition. Relational databases natively support SQL joins and transaction semantics.

This alternative, Evaluate Amazon Neptune for graph traversal workload requirements, is appropriate when A social graph must repeatedly traverse variable-depth connections between entities. A graph model supports relationship traversals beyond simple key lookups.

Although Choose DynamoDB with well-distributed partition keys and suitable capacity mode can be valid when A high-throughput key-value service needs predictable single-digit-millisecond access at large scale, it targets a different technical condition. DynamoDB is designed for scalable key-value access patterns without relational joins.

Evaluate Amazon Neptune for graph traversal workload requirements would address a different requirement: A social graph must repeatedly traverse variable-depth connections between entities. A graph model supports relationship traversals beyond simple key lookups.

For the separate situation A new industrial monitoring system needs real-time measurements queried by timestamp through native InfluxDB APIs, Use Amazon Timestream for InfluxDB with appropriate tag and retention design could be useful. Managed InfluxDB supports time-series ingestion and query patterns for new customers, unlike closed-to-new-customer LiveAnalytics.

This alternative, Index events in Amazon OpenSearch Service with suitable sharding and lifecycle, is appropriate when A log analytics service needs near-real-time full-text search over application events. OpenSearch supports search-oriented indexes and aggregations.

For the separate situation A new industrial monitoring system needs real-time measurements queried by timestamp through native InfluxDB APIs, Use Amazon Timestream for InfluxDB with appropriate tag and retention design could be useful. Managed InfluxDB supports time-series ingestion and query patterns for new customers, unlike closed-to-new-customer LiveAnalytics.

Choose DynamoDB with well-distributed partition keys and suitable capacity mode solves the related scenario A high-throughput key-value service needs predictable single-digit-millisecond access at large scale; dynamoDB is designed for scalable key-value access patterns without relational joins.

For the separate situation A new industrial monitoring system needs real-time measurements queried by timestamp through native InfluxDB APIs, Use Amazon Timestream for InfluxDB with appropriate tag and retention design could be useful. Managed InfluxDB supports time-series ingestion and query patterns for new customers, unlike closed-to-new-customer LiveAnalytics.

Use Amazon Aurora or RDS relational databases with an appropriate engine solves the related scenario An application requires complex SQL joins and ACID transactions over structured entities; relational databases natively support SQL joins and transaction semantics.

Although Evaluate Amazon Neptune for graph traversal workload requirements can be valid when A social graph must repeatedly traverse variable-depth connections between entities, it targets a different technical condition. A graph model supports relationship traversals beyond simple key lookups.

For the separate situation A high-throughput key-value service needs predictable single-digit-millisecond access at large scale, Choose DynamoDB with well-distributed partition keys and suitable capacity mode could be useful. DynamoDB is designed for scalable key-value access patterns without relational joins.

Use Amazon Aurora or RDS relational databases with an appropriate engine solves the related scenario An application requires complex SQL joins and ACID transactions over structured entities; relational databases natively support SQL joins and transaction semantics.

Evaluate Amazon Neptune for graph traversal workload requirements would address a different requirement: A social graph must repeatedly traverse variable-depth connections between entities. A graph model supports relationship traversals beyond simple key lookups.