Stark broadening theories and calculations have been extensively developed for about 50 years. The theory can now be considered as mature for many applications, especially for accurate spectroscopic diagnostics and modelling. In astrophysics, with the increasing sensitivity of observations and spectral resolution, in all domains of wavelengths from far UV to infrared, it has become possible to develop realistic models of interiors and atmospheres of stars and interpret their evolution and the creation of elements through nuclear reactions. For hot stars, especially white dwarfs, Stark broadening is the dominant collisional line broadening process. This requires the knowledge of numerous profiles, especially for trace elements, which are used as useful probes for modern spectroscopic diagnostics. Hence, calculations based on a simple but enough accurate and fast method, are necessary for obtaining numerous results. Ab initio calculations are a growing domain of development. Nowadays, the access to such data via an on line database becomes crucial. This is the object of STARK-B, which is a collaborative project between the Paris Observatory and the Astronomical Observatory of Belgrade. It is a database of calculated widths and shifts of isolated lines of atoms and ions due to electron and ion collisions. It is devoted to modelling and spectroscopic diagnostics of stellar atmospheres and envelopes. In addition, it is relevant to laboratory plasmas, laser equipments and technological plasmas. It is a part of VAMDC (Virtual Atomic and Molecular Data Centre), which is an European Union funded collaboration between groups involved in the generation and use of atomic and molecular data.