SAP HANA Cloud: Data Management for Modern Enterprise Applications

Cloud Computing environments are a fantastic foundation for unprecedented opportunities in the context of data management. Scalability and availability can be achieved by consequently disaggregating compute and storage and leveraging different data centers around the globe. Although conceptually infinitely available resources promise an easy game for database systems, leveraging those opportunities on the one side and providing enterprise-scale data management solutions for a wide range of extremely challenging business applications on the other side is by far not trivial. Within this paper we share the fundamental principles of SAP HANA Cloud---the data management backbone for all SAP applications---by providing insights into design criteria and technological centerpieces of SAP HANA Cloud as a result of a stringent evolutionary shift from a purely on-premise, in-memory database engine to an on-Cloud, memory-storage hierarchy-aware data management platform. We will motivate the transition and the decision of certain developments by sharing key characteristics of some key applications run by SAP, which require enterprise qualities like availability, cloud-capabilities like resource elasticity, while managing overall costs and the flexibility to adapt the underlying data management system to their specific needs. The overall goal of the paper is to provide insights into the ''story of SAP HANA Cloud'' by sharing challenges and lessons learned within the evolution towards a composable cloud-based data platform, thus strongly arguing for a ''one size fits all'' solution, if done right.