Design of SAP Workloads on AWS practice questions

Deploy the SAP application on a single EC2 instance and use S3 for database backups with cross-region replication.

Why wrong: Single instance is a single point of failure; S3 is not suitable for database replication.

Deploy a single EC2 instance for the SAP application and database in one Availability Zone, and take hourly snapshots of the EBS volumes.

Why wrong: Single AZ does not provide high availability; hourly snapshots do not meet RPO of 15 minutes.

Use two EC2 instances in an Auto Scaling group with instance store volumes for the database, and store application files on S3.

Why wrong: Instance store is ephemeral; S3 cannot serve as a shared file system for SAP.

Migrate from gp2 to io2 EBS volumes with the same size.

Why wrong: io2 is more expensive per GB; gp2 size increase is more cost-effective.

Modify the volume to use Provisioned IOPS (io1) with a higher IOPS value.

Why wrong: Provisioned IOPS incurs additional cost; increasing gp2 size is cheaper.

Add an additional EBS volume and stripe the volumes using LVM.

Why wrong: Striping adds complexity and cost; increasing volume size is simpler and cheaper.

Use a single io2 Block Express EBS volume of the required size.

Why wrong: Single volume may not provide enough IOPS for 1.9 TB HANA; striping multiple volumes is recommended.

Use multiple gp3 EBS volumes striped together with LVM.

Why wrong: gp3 has lower max IOPS per volume than io2; for large HANA, io2 is preferred.

Use NVMe instance store volumes for the HANA data and log areas.

Why wrong: Instance store is ephemeral; data is lost on instance stop/termination.

Use EBS snapshots of the root and data volumes taken every 5 minutes.

Why wrong: EBS snapshots alone are not crash-consistent for HANA; they require application consistency.

Set up HANA system replication across two EC2 instances in different Availability Zones with manual failover.

Why wrong: System replication requires managing secondary instance and manual failover increases RTO.

Schedule manual HANA backups to S3 using cron scripts and hdbsql commands.

Why wrong: Manual scheduling increases operational overhead and risk of missed backups.

Place the application servers in one Availability Zone and the database in a different Availability Zone.

Why wrong: Cross-AZ traffic adds latency compared to same AZ.

Place the application servers in one VPC and the database in a different VPC connected via VPC peering.

Why wrong: VPC peering adds slight latency; same VPC is better.

Place the application servers in one AWS Region and the database in another Region.

Why wrong: Cross-Region latency is too high for SAP database communication.

An Application Load Balancer to distribute traffic between ASCS and ERS instances.

Why wrong: ASCS and ERS use a virtual IP, not a load balancer.

A read replica of the SAP HANA database to offload application traffic.

Why wrong: Read replicas are for databases, not for ASCS/ERS.

A secondary Windows Server Failover Cluster in a different Availability Zone.

Why wrong: Linux Pacemaker is typically used on AWS for SAP HA.

SAP HANA supports RAID 0 across multiple io2 volumes without additional software.

Why wrong: HANA requires LVM or similar for striping; RAID 0 is not directly supported.

io2 Block Express volumes can be attached to multiple EC2 instances simultaneously using EBS Multi-Attach.

Why wrong: io2 Block Express does not support Multi-Attach.

Change the volume type to gp3 with higher throughput.

Why wrong: gp3 is not optimized for low latency.

Move to a larger instance type with more network bandwidth.

Why wrong: Instance size does not directly reduce EBS latency.

Configure RAID 0 across multiple volumes.

Why wrong: RAID0 improves throughput, not latency.

Single AZ with HANA System Replication to a standby in the same AZ.

Why wrong: Does not protect against AZ failure.

Single AZ with daily backups to Amazon S3.

Why wrong: Backup & Restore has high RTO, not automatic.

Multi-AZ with HANA System Replication in async mode and manual failover.

Why wrong: Manual failover is not automatic.

The ALB health check interval is too short.

Why wrong: Health check interval doesn't cause uneven distribution.

Cross-zone load balancing is enabled.

Why wrong: Cross-zone balancing helps distribution.

The ALB is using the least outstanding requests routing algorithm.

Why wrong: That algorithm balances based on pending requests, not uneven.

Take hourly backups to S3 and restore in us-west-2 when needed.

Why wrong: Restore time exceeds RTO.

Use HANA System Replication to us-west-2 with manual failover.

Why wrong: Manual failover increases RTO and operational overhead.

Use HANA System Replication in async mode to us-west-2.

Why wrong: Async mode may not meet RPO of 15 minutes.

r5.24xlarge

Why wrong: r5 is smaller than x1e for scale-up.

i3.16xlarge

Why wrong: i3 is storage-optimized.

c5.18xlarge

Why wrong: c5 is compute-optimized, not memory-optimized.

Disable encryption in transit between application and database tiers.

Why wrong: Encryption in transit should be enforced.

Allow SSH access from the internet for administration.

Why wrong: SSH from internet is not secure; use bastion host.

Store SAP backup files in a public S3 bucket for easy access.

Why wrong: Public buckets are not secure.

Back up HANA data directly to Amazon S3 using standard tools.

Why wrong: HANA does not directly support S3 backup without Backint.

Replicate backups to another AWS Region using S3 Cross-Region Replication.

Why wrong: This is for disaster recovery, not backup strategy.

Use AWS Database Migration Service (DMS) with continuous replication from the primary HANA database to a target in us-west-2.

Why wrong: DMS does not support HANA as a source for continuous replication.

Use AWS Backup with continuous backups and cross-region copy to achieve RPO of 15 minutes.

Why wrong: AWS Backup for HANA does not support continuous backups; it supports periodic backups only.

Use AWS Backup to take hourly backups of the HANA database and restore in us-west-2 during a disaster.

Why wrong: Hourly backups result in RPO of up to 1 hour, exceeding 15 minutes.

The instance type of the new ASCS instance is different from the original, causing incompatibility with the SAP kernel.

Why wrong: Instance type change does not affect quorum device availability.

The ASCS instance does not have a public IP address, preventing communication with the quorum device.

Why wrong: Quorum communication can occur over private IP; public IP is not required.

The Security Group associated with the ASCS instance does not allow inbound traffic on the quorum port.

Why wrong: Quorum is typically based on shared storage or network connectivity, not a specific port.

Export the HANA database to flat files using SAP HANA Studio, upload the files to Amazon S3, and import them into an EC2 HANA instance.

Why wrong: Export/import is time-consuming and requires significant downtime.

Use AWS Database Migration Service (DMS) with ongoing replication to migrate the database to Amazon RDS for SAP HANA.

Why wrong: DMS does not support SAP HANA as a source or target.

Use AWS Backup to take a full backup of the on-premises HANA database and restore it to an EC2 instance in AWS.

Why wrong: Backup and restore involves longer downtime due to backup and restore times.

Create a VPC peering connection between the subnets of the application servers and the database.

Why wrong: VPC peering is for cross-VPC connectivity; within the same VPC, it is not applicable.

Change the HANA database instance type to a compute-optimized instance such as c5.9xlarge.

Why wrong: Instance type change does not directly reduce network latency.

Enable Enhanced Networking on the application servers by attaching an Elastic Network Adapter (ENA).

Why wrong: Enhanced Networking is already enabled on current generation instances; this may not reduce latency further.

HANA System Replication is not configured correctly for automatic failover.

Why wrong: The scenario states HSR is configured; the issue is more likely with application connectivity.

The Amazon EFS file system is not available in the secondary Availability Zone.

Why wrong: EFS is a regional service accessible from all AZs in the region.

The security group of the application servers does not allow outbound traffic to the secondary HANA instance.

Why wrong: Security groups can be updated automatically, but this is less likely if proper design is in place.

Use EBS gp3 volumes for the HANA data volume.

Why wrong: gp3 is general purpose; io2 is better for HANA.

Use EC2 instance store volumes for the HANA data volume.

Why wrong: Instance store is ephemeral and not suitable for persistent database data.

Store HANA data on Amazon S3 with S3 Express One Zone.

Why wrong: S3 is object storage and not suitable for HANA data volumes.