The CutFlowService tool in the ATLAS experiment

We present the CutFlowSvc, a service for coherent organization of the reading and writing of bookkeeping information, in the context of the ATLAS experiment. It is a necessary tool because amount of data stored by the LHC experiments is typically so large, that running each individual analysis or performance measurement over the full dataset including all details is prohibitive. Hence, the data is stored consecutively in different data sets of smaller size, each skimmed from the previous, larger format. The end user therefore needs detailed information about the application and efficiency of each cut used for the reduction of the dataset in the steps before. The implementation of the CutFlowSvc within the ATLAS software framework Athena is summarized as well as its application for different ATLAS-specific data formats: the POOL format as well as flat data structures. Finally an overview of future developments is presented.