Automate and Optimize SAP Network Performance in a Multi-AZ deployment

This article provides guidance on automating and optimizing SAP network performance in a multi-Availability Zone (AZ) deployment on AWS. It highlights the importance of running performance-critical SAP batch jobs or transactions on application servers located in the same AZ as the database server to minimize network latency.

Specifically, the article covers:

• SAP's recommendation on network latency between application and database servers
• Impact of inter-AZ and intra-AZ network latency on performance-critical SAP workloads
• A solution to automatically update SAP's load-balancing mechanism (logon groups, batch server groups, and RFC server groups) to point to the application servers in the same AZ as the database server, even after a database failover
• Implementation details of the solution, including code snippets for detecting database failure, modifying logon/RFC server groups, and changing batch server groups
• Conclusion emphasizing the benefits of the solution for high availability and optimal performance of SAP workloads on AWS