Enterprise environments have isolated teams responsible separately for database management and the management of underlying networkattached server-storage infrastructure (referred to as Storage Area Networks or SANs). Our vision is to develop an innovative management framework to simplify administrative tasks that require an understanding of both database and SAN details. As a concrete instance, this paper describes problem diagnosis of a database query slowdown. To address the diagnosis challenges including noisy monitoring data, we have implemented DIADS, which uses existing monitoring tools to generate Annotated Plan Graphs (APGs). Using an innovative workflow that combines domain-specific knowledge with machine-learning, DIADS was successfully applied on a real-world testbed to pin-point complex combination of events across both the database and SAN.