A Probabilistic Analysis of Data Popularity in ATLAS Data Caching

One of the most important aspects in any distribution system is efficient data replication over storage / computing centers, that guarantees high data availability and low cost of resources utilization. In this paper we propose a data distribution scheme for the production and distributed analysis system PanDA at the ATLAS experiment. Our proposed scheme is based on an investigation of data usage. Thus, the paper is focused on the main concepts of data popularity at the PanDA system and their utilization. Data popularity is represented as the set of parameters that are used to predict the future data state in terms of popularity levels.