A company is planning to migrate its SAP Business Warehouse (SAP BW) 7.5 system on SAP HANA from on premises to AWS. The production database is 4 TB in size and has a scale-out architecture that consi

Question

Options

• AImplement SAP HANA scale-out Amazon EC2 instances with default tenancy.
• BImplement SAP HANA scale-out Amazon EC2 instances with Capacity Reservations in a cluster
• CImplement SAP HANA scale-out Amazon EC2 instances in a spread placement group.
• DImplement SAP HANA scale-out Amazon EC2 instances in a partition placement group.

• A
  6% (1)
• B
  78% (14)
• C
  6% (1)
• D
  11% (2)

Explanation

SAP HANA scale-out relies heavily on fast, low-latency inter-node communication (inter-node joins, data redistribution, etc.). An EC2 cluster placement group places instances physically close together on high-bandwidth, low-latency 10/25/100 Gbps networking within a single Availability Zone, maximizing throughput between the three HANA nodes. Adding Capacity Reservations ensures the specific instance types needed for the large-memory SAP HANA nodes are actually available when required. A spread placement group (C) deliberately separates instances across distinct hardware for fault isolation, which reduces inter-node network performance. A partition placement group (D) is designed for large distributed workloads like HDFS and Cassandra, not for the tightly-coupled communication pattern of SAP HANA scale-out. Default tenancy without a placement group (A) provides no throughput optimization.

No community discussion yet for this question.