Hosted by: Alessandro Tricoli

The LHC experiments require huge, ever-increasing volumes of data to explore the frontiers of particle physics, and the grid provides the infrastructure to meet this challenge. Today, even medium sized experiments produce petabytes of data that need to be analysed by international collaborations and the relationship between particle physicists and their data has had to evolve to keep pace. Calibrating and systematically understanding detectors requires detailed information, deep learning algorithms promise to allow us to fully exploit our experiments, and on the other hand we want to analyse our data quickly and be the first to publish. This talk will first describe how, shortly after the Higgs discovery, ATLAS realised that the way that analysis was done had to change, and will briefly illustrate what those changes entailed. Next, the NA62 experiment will be described. Recording a billion events per day, NA62 aims to measure the very rare decay of a charged kaon to a charged pion and two neutrinos. The collaboration had to quickly implement petabyte scale data processing infrastructure to perform the sub-nanosecond calibrations needed to challenge the 10% precision of the theory prediction.