Abstract: In this presentation, an update on a turbulence database system is provided, together with a sample application to a transitional boundary layer flow focusing on scaling properties of turbulent spots during boundary layer transition. The Johns Hopkins Turbulence Databases (JHTDB) exposes large-scale turbulent data to the research community while at the same time providing easy-to-use client interfaces based on Web Services that act as “virtual flow sensors” that can be placed in the turbulent flows. This approach has greatly facilitated retrieving and interacting with the data. At present JHTDB contains over 1/2 Petabyte of data from various turbulent flow simulations. The data have been used in over 150 peer-reviewed journal publications on turbulence from authors world-wide. We also present an application of the transitional (by-pass) boundary layer dataset contained in JHTDB. Specifically, we develop a new approach to detect the interface that separates the turbulent boundary layer from the laminar or outer regions of the flow using machine learning. A self-organized map based clustering method is shown to enable determination of the interface without having to prescribe arbitrary threshold values as traditional interface detection methods require. Scaling properties of the interface are studied and links to fractal properties of turbulent non-turbulent interfaces in high Reynolds number flows are established. This work has been performed with Drs. Zhao Wu and Tamer Zaki, while the database (supported by the NSF) has resulted from a long-term collaboration with the JHTDB team.