Model-driven data warehousing for service-based business process monitoring

Today for compliance reasons as well as for optimization of business processes companies need to store runtime data of processes. However, creating and adapting an analytics infrastructure using data warehousing is a cumbersome task. In previous work we introduced the aPro architecture in order to enable companies with heterogeneous infrastructures to create near real-time monitoring and analytics infrastructures. Creation and deployment of monitoring infrastructure is provided as a fully automatic service. In this work we add a model-driven data warehouse to aPro, using the existing monitoring model to automatically configure and deploy a data warehouse. We show how a near real-time monitoring infrastructure can be utilized to simplify Extract-Transform-Load processes for filling data warehouses. We describe a prototypical implementation and evaluate it with a real-world process.