Amazon AWS Certified Solutions Architect - Professional SAP-C02 Bundle

For the separate situation A global media site retrieves the same files from origin for most requests despite a CloudFront distribution, Inspect cache-key headers, cookies and TTL policies to improve cacheability could be useful. Excessive cache-key variation fragments edge cache hit rates.

Review TTL, key distribution and read-through cache population solves the related scenario An ElastiCache deployment shows a high miss rate and database traffic did not fall; cache effectiveness depends on repeated keys and suitable expiration rules.

Although Add explicit invalidation or short TTL for critical mutable inventory fields can be valid when A product API returns stale inventory levels after cache entries are updated elsewhere, it targets a different technical condition. Cache-coherence controls bound staleness when the underlying source changes.

Although Configure CloudFront cache policies that forward necessary context and bypass unsafe caching can be valid when A server-rendered site requires personalized dynamic pages that must never be served from a shared stale cache, it targets a different technical condition. Correct cache keys and TTLs prevent one user's personalized output reaching another.

Enable supported HTTP compression at the edge or origin and benchmark transfer would address a different requirement: A legacy application returns large responses with redundant uncompressed text. Response compression reduces transmitted bytes and download time for compressible payloads.

For the separate situation An ALB-heavy app has slow requests due to mismatched backend target health and drain settings, Analyze ALB target-response-time and unhealthy-host metrics before modifying target capacity could be useful. Load balancer metrics separate frontend latency from backend processing failures.

Cache versioned static assets through CloudFront edge locations would address a different requirement: A global browser app repeatedly fetches static JavaScript from one Region with high RTT. Edge delivery reduces geographic round trips and origin load.

This alternative, Enable supported HTTP compression at the edge or origin and benchmark transfer, is appropriate when A legacy application returns large responses with redundant uncompressed text. Response compression reduces transmitted bytes and download time for compressible payloads.

For the separate situation An ALB-heavy app has slow requests due to mismatched backend target health and drain settings, Analyze ALB target-response-time and unhealthy-host metrics before modifying target capacity could be useful. Load balancer metrics separate frontend latency from backend processing failures.

This alternative, Evaluate AWS Global Accelerator rather than relying only on DNS-based regional routing, is appropriate when A gaming service requires stable anycast IPs and low-latency TCP sessions to healthy regional backends. Anycast ingress routes flows over AWS global network toward healthy endpoints.

For the separate situation A server-rendered site requires personalized dynamic pages that must never be served from a shared stale cache, Configure CloudFront cache policies that forward necessary context and bypass unsafe caching could be useful. Correct cache keys and TTLs prevent one user's personalized output reaching another.

Although Cache versioned static assets through CloudFront edge locations can be valid when A global browser app repeatedly fetches static JavaScript from one Region with high RTT, it targets a different technical condition. Edge delivery reduces geographic round trips and origin load.

For the separate situation A server-rendered site requires personalized dynamic pages that must never be served from a shared stale cache, Configure CloudFront cache policies that forward necessary context and bypass unsafe caching could be useful. Correct cache keys and TTLs prevent one user's personalized output reaching another.

Cache versioned static assets through CloudFront edge locations solves the related scenario A global browser app repeatedly fetches static JavaScript from one Region with high RTT; edge delivery reduces geographic round trips and origin load.

Although Enable supported HTTP compression at the edge or origin and benchmark transfer can be valid when A legacy application returns large responses with redundant uncompressed text, it targets a different technical condition. Response compression reduces transmitted bytes and download time for compressible payloads.

Evaluate AWS Global Accelerator rather than relying only on DNS-based regional routing solves the related scenario A gaming service requires stable anycast IPs and low-latency TCP sessions to healthy regional backends; anycast ingress routes flows over AWS global network toward healthy endpoints.

Although Enable supported HTTP compression at the edge or origin and benchmark transfer can be valid when A legacy application returns large responses with redundant uncompressed text, it targets a different technical condition. Response compression reduces transmitted bytes and download time for compressible payloads.

Analyze ALB target-response-time and unhealthy-host metrics before modifying target capacity would address a different requirement: An ALB-heavy app has slow requests due to mismatched backend target health and drain settings. Load balancer metrics separate frontend latency from backend processing failures.

Profile application metadata operations and parallelize I/O before increasing throughput would address a different requirement: An EFS-backed workload has a small-files latency problem and expects EFS throughput settings alone to fix it. Network file-system metadata latency may dominate workloads with many tiny files.

For the separate situation An OpenSearch cluster has one overloaded shard while the rest are idle, Rebalance shard routing and choose a healthier index-sharding and ingestion design could be useful. Skewed shards cause hotspot latency regardless of aggregate cluster size.

Evaluate Amazon Timestream for InfluxDB 3 Enterprise with workload-tested retention and cardinality solves the related scenario A new observability platform needs complex high-cardinality historical time-series analytics with InfluxDB 3 query semantics; influxDB 3 uses columnar storage and supports high-cardinality analytic queries; retired LiveAnalytics is not available to new customers.

This alternative, Rebalance shard routing and choose a healthier index-sharding and ingestion design, is appropriate when An OpenSearch cluster has one overloaded shard while the rest are idle. Skewed shards cause hotspot latency regardless of aggregate cluster size.

For the separate situation A new observability platform needs complex high-cardinality historical time-series analytics with InfluxDB 3 query semantics, Evaluate Amazon Timestream for InfluxDB 3 Enterprise with workload-tested retention and cardinality could be useful. InfluxDB 3 uses columnar storage and supports high-cardinality analytic queries; retired LiveAnalytics is not available to new customers.

Move heavy historical analytics into Redshift or a suitable columnar analytics design solves the related scenario An analytical SQL report scans large rows in a relational OLTP database each hour; columnar analytics engines optimize scans and aggregations while preserving OLTP capacity.

For the separate situation An OpenSearch cluster has one overloaded shard while the rest are idle, Rebalance shard routing and choose a healthier index-sharding and ingestion design could be useful. Skewed shards cause hotspot latency regardless of aggregate cluster size.

Although Move to a supported gp3 or io2 configuration with deliberately provisioned IOPS and throughput can be valid when An EC2-hosted database is limited by baseline EBS IOPS under peak transactional writes, it targets a different technical condition. Tunable SSD volume performance addresses block I/O limits independent of capacity.

Profile application metadata operations and parallelize I/O before increasing throughput would address a different requirement: An EFS-backed workload has a small-files latency problem and expects EFS throughput settings alone to fix it. Network file-system metadata latency may dominate workloads with many tiny files.

Although Move to a supported gp3 or io2 configuration with deliberately provisioned IOPS and throughput can be valid when An EC2-hosted database is limited by baseline EBS IOPS under peak transactional writes, it targets a different technical condition. Tunable SSD volume performance addresses block I/O limits independent of capacity.

Profile application metadata operations and parallelize I/O before increasing throughput would address a different requirement: An EFS-backed workload has a small-files latency problem and expects EFS throughput settings alone to fix it. Network file-system metadata latency may dominate workloads with many tiny files.

This alternative, Evaluate Amazon Timestream for InfluxDB 3 Enterprise with workload-tested retention and cardinality, is appropriate when A new observability platform needs complex high-cardinality historical time-series analytics with InfluxDB 3 query semantics. InfluxDB 3 uses columnar storage and supports high-cardinality analytic queries; retired LiveAnalytics is not available to new customers.

Although Move heavy historical analytics into Redshift or a suitable columnar analytics design can be valid when An analytical SQL report scans large rows in a relational OLTP database each hour, it targets a different technical condition. Columnar analytics engines optimize scans and aggregations while preserving OLTP capacity.

Rebalance shard routing and choose a healthier index-sharding and ingestion design would address a different requirement: An OpenSearch cluster has one overloaded shard while the rest are idle. Skewed shards cause hotspot latency regardless of aggregate cluster size.

This alternative, Move to a supported gp3 or io2 configuration with deliberately provisioned IOPS and throughput, is appropriate when An EC2-hosted database is limited by baseline EBS IOPS under peak transactional writes. Tunable SSD volume performance addresses block I/O limits independent of capacity.