Description
The challenges of end user analysis at the LHC are the data input sets of varying formats and sizes (from a few rare special events to full data sets of 10^9 events), the number of analysis cycles needed to extract the physics signals (varying from one scheduled production to tens of cycles needed to optimize the algorithms) and last not least the large number of physicists performing analysis with their wide spectrum of computing skills.
Because of the large computing needs batch type analysis, both scheduled and individual user analysis, are performed on large computing grids. At the same time the need for fast development and testing cycles triggered an interest in interactive tools.
This talk describes the solutions implemented by the LHC experiments to cope with these analysis challenges.
Speaker
Andreas Morsch
(CERN and ALICE)
