Amazon AWS Certified Solutions Architect - Professional SAP-C02 Bundle

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.

Use a cluster placement group with supported enhanced-networking instances would address a different requirement: A high-frequency trading app needs low-latency east-west network inside one AZ. Cluster placement groups place compatible instances close for low-latency throughput.

This alternative, Select memory-optimized instances after measuring heap and allocation profiles, is appropriate when A large in-memory analytics service is page-faulting because its working set exceeds RAM. Memory-focused capacity reduces swapping and memory-pressure bottlenecks.

For the separate situation A throughput-sensitive EC2 fleet underutilizes network bandwidth due to its instance type, Compare supported enhanced networking, instance bandwidth and placement constraints could be useful. Network performance depends on instance capabilities as well as topology.

Use a cluster placement group with supported enhanced-networking instances would address a different requirement: A high-frequency trading app needs low-latency east-west network inside one AZ. Cluster placement groups place compatible instances close for low-latency throughput.

This alternative, Select memory-optimized instances after measuring heap and allocation profiles, is appropriate when A large in-memory analytics service is page-faulting because its working set exceeds RAM. Memory-focused capacity reduces swapping and memory-pressure bottlenecks.

For the separate situation A batch job performs high-volume numerical operations that benefit from SIMD instructions, Benchmark compute-optimized EC2 families against the job's actual throughput could be useful. Compute-optimized instances prioritize processor resources for CPU-bound applications.

For the separate situation A throughput-sensitive EC2 fleet underutilizes network bandwidth due to its instance type, Compare supported enhanced networking, instance bandwidth and placement constraints could be useful. Network performance depends on instance capabilities as well as topology.

Benchmark compute-optimized EC2 families against the job's actual throughput solves the related scenario A batch job performs high-volume numerical operations that benefit from SIMD instructions; compute-optimized instances prioritize processor resources for CPU-bound applications.

Although Choose supported accelerated-computing instances sized for model memory and throughput can be valid when A machine learning inference job needs GPU-accelerated deep-learning execution, it targets a different technical condition. GPU instances accelerate parallel model operations unavailable on general-purpose CPUs.

For the separate situation A batch job performs high-volume numerical operations that benefit from SIMD instructions, Benchmark compute-optimized EC2 families against the job's actual throughput could be useful. Compute-optimized instances prioritize processor resources for CPU-bound applications.

Although Use a cluster placement group with supported enhanced-networking instances can be valid when A high-frequency trading app needs low-latency east-west network inside one AZ, it targets a different technical condition. Cluster placement groups place compatible instances close for low-latency throughput.

Compare supported enhanced networking, instance bandwidth and placement constraints would address a different requirement: A throughput-sensitive EC2 fleet underutilizes network bandwidth due to its instance type. Network performance depends on instance capabilities as well as topology.

Although Use a cluster placement group with supported enhanced-networking instances can be valid when A high-frequency trading app needs low-latency east-west network inside one AZ, it targets a different technical condition. Cluster placement groups place compatible instances close for low-latency throughput.

Select memory-optimized instances after measuring heap and allocation profiles would address a different requirement: A large in-memory analytics service is page-faulting because its working set exceeds RAM. Memory-focused capacity reduces swapping and memory-pressure bottlenecks.

This alternative, Choose supported accelerated-computing instances sized for model memory and throughput, is appropriate when A machine learning inference job needs GPU-accelerated deep-learning execution. GPU instances accelerate parallel model operations unavailable on general-purpose CPUs.

Choose Amazon FSx for Lustre with an appropriate deployment and S3 integration would address a different requirement: A distributed HPC application needs very high throughput parallel file access and optional S3 data repository integration. Lustre is designed for parallel high-performance file workloads.

This alternative, Store the content in S3 with suitable key naming and retrieval patterns, is appropriate when An object archive stores images and videos accessed through REST APIs rather than mounted file-system semantics. Object storage scales independently of server volume and file share mounts.

Use Amazon EFS with appropriate performance and access controls solves the related scenario Hundreds of Linux application hosts require concurrent access to a managed shared POSIX file system; eFS provides elastic shared NFS storage for Linux workloads.

This alternative, Store the content in S3 with suitable key naming and retrieval patterns, is appropriate when An object archive stores images and videos accessed through REST APIs rather than mounted file-system semantics. Object storage scales independently of server volume and file share mounts.

For the separate situation An existing FSx for Windows File Server share has adequate storage but SMB clients saturate file throughput during daily processing, Increase FSx for Windows throughput capacity after measuring client I/O and workload limits could be useful. File-system throughput capacity is provisioned separately from total stored file capacity.

Although Choose Amazon FSx for Lustre with an appropriate deployment and S3 integration can be valid when A distributed HPC application needs very high throughput parallel file access and optional S3 data repository integration, it targets a different technical condition. Lustre is designed for parallel high-performance file workloads.

For the separate situation A relational database on EC2 needs consistently provisioned high random IOPS on attached block storage, Select EBS Provisioned IOPS SSD with performance sizing validated against workload needs could be useful. Provisioned IOPS EBS supplies independently configured block IOPS and throughput.

Although Use Amazon EFS with appropriate performance and access controls can be valid when Hundreds of Linux application hosts require concurrent access to a managed shared POSIX file system, it targets a different technical condition. EFS provides elastic shared NFS storage for Linux workloads.

Choose Amazon FSx for Lustre with an appropriate deployment and S3 integration would address a different requirement: A distributed HPC application needs very high throughput parallel file access and optional S3 data repository integration. Lustre is designed for parallel high-performance file workloads.

Use Amazon EFS with appropriate performance and access controls solves the related scenario Hundreds of Linux application hosts require concurrent access to a managed shared POSIX file system; eFS provides elastic shared NFS storage for Linux workloads.

Although Store the content in S3 with suitable key naming and retrieval patterns can be valid when An object archive stores images and videos accessed through REST APIs rather than mounted file-system semantics, it targets a different technical condition. Object storage scales independently of server volume and file share mounts.

This alternative, Select EBS Provisioned IOPS SSD with performance sizing validated against workload needs, is appropriate when A relational database on EC2 needs consistently provisioned high random IOPS on attached block storage. Provisioned IOPS EBS supplies independently configured block IOPS and throughput.

Although Choose Amazon FSx for Lustre with an appropriate deployment and S3 integration can be valid when A distributed HPC application needs very high throughput parallel file access and optional S3 data repository integration, it targets a different technical condition. Lustre is designed for parallel high-performance file workloads.

Select EBS Provisioned IOPS SSD with performance sizing validated against workload needs would address a different requirement: A relational database on EC2 needs consistently provisioned high random IOPS on attached block storage. Provisioned IOPS EBS supplies independently configured block IOPS and throughput.

This alternative, Increase FSx for Windows throughput capacity after measuring client I/O and workload limits, is appropriate when An existing FSx for Windows File Server share has adequate storage but SMB clients saturate file throughput during daily processing. File-system throughput capacity is provisioned separately from total stored file capacity.