Exam AWS Certified Solutions Architect - Professional SAP-C02 topic 1 question 235 discussion

A. Ensure the HPC cluster is launched within a single Availability Zone: This choice ensures that the EC2 instances in the cluster have low network latency and high bandwidth, as they are located within the same data center. C. Select EC2 instance types with an Elastic Fabric Adapter (EFA) enabled: EFA is a network interface that provides low-latency, high-bandwidth communication between EC2 instances. By selecting instance types with EFA enabled, the cluster can benefit from improved inter-instance communication. F. Replace Amazon EFS with Amazon FSx for Lustre: Amazon FSx for Lustre is a high-performance file system optimized for HPC workloads. By using FSx for Lustre instead of Amazon EFS, the cluster can achieve better performance for the large number of shared files generated by the workload. And what about a cluster placement group?

C. Select EC2 instance types with an Elastic Fabric Adapter (EFA) enabled. D. Ensure the cluster is launched across multiple Availability Zones. F. Replace Amazon EFS with Amazon FSx for Lustre.

Going to ACF ,looks logical

A, C, F

ACF for performance

@MODERATOR - Please remove my previous comment. I agree with ACF. Thank you MoussaNoussa for clarifying

ACF is the correct answer

A, C and F

A - Single AZ is better than multi AZ for performance C - Use EFA. https://aws.amazon.com/hpc/efa/ - It tells you that's HPC is a use case. F - Use FSx for Lustre - https://aws.amazon.com/fsx/lustre/. HPC is a use case.

A, C and F

ACF is the correct answer

a-c-f...a-c-f...a-c-f To achieve maximum performance from the HPC cluster, the following design choices should be made: A. Ensure the HPC cluster is launched within a single Availability Zone: This choice ensures that the EC2 instances in the cluster have low network latency and high bandwidth, as they are located within the same data center. C. Select EC2 instance types with an Elastic Fabric Adapter (EFA) enabled: EFA is a network interface that provides low-latency, high-bandwidth communication between EC2 instances. By selecting instance types with EFA enabled, the cluster can benefit from improved inter-instance communication. F. Replace Amazon EFS with Amazon FSx for Lustre: Amazon FSx for Lustre is a high-performance file system optimized for HPC workloads. By using FSx for Lustre instead of Amazon EFS, the cluster can achieve better performance for the large number of shared files generated by the workload.

B: more interface does not mean faster. so B is not a good choice. E: RAID? is often not recommended on Cloud Platform, aws has already raid the drive for you underlay. A: HPC recommended to use multiregion. so CDF

ACF is the right answer

CDF are correct C - EFA provides low-latency and high-bandwidth communication between EC2 instances. It can optimize the network performance of the HPC cluster. D - Launching the HPC cluster across multiple Availability Zones allows you to distribute the workload and resources, reducing the chances of a single point of failure and increasing overall performance. F - FSx for Lustre is a high-performance file system optimized for HPC workloads.