The physics program in the ATLAS experiment at the Large Hadron Collider has achieved high-quality and competitive physics results since its inception, based on large and accurate datasets of simulated Monte Carlo events. Generating these billions of Monte Carlo using the Geant4 framework requires significant computing resources. As such the ATLAS CPU is limited and taking into account the expected increase of the average number of proton-proton collisions during the high luminosity LHC phase, a very fast detector simulation is critical when high accuracy is not needed. In order to solve this problem and to meet the challenge of simulated sample statistics requirements, fast alternatives approaches have been developed by ATLAS to replace the algorithms currently used in the standard MC production chain. This document provides a description of the Fast simulation chain tools and also presents the latest physics performance studies: Fast tracking detector simulation, Fast digitization and Pile-up simulation.