Optimization of the job management in a multi-queue environment

Due to the outstanding performance and availability of the Large Hadron Collider (LHC) at CERN, unprecedented volumes of experimental data have been produced in the last years. Since increasing the storage and computing capacities cannot keep up with the experiments' requirements, new solutions are conceived and implemented regarding the optimization of the resource usage, the coupling between the available processing units and data sources, and the improvement of the data transfer management. These changes impact all the centres of the Worldwide LHC Computing Grid collaboration (WLCG), including those of the Romanian Tier-2 Federation (RO-LCG). The main purpose of this paper is to present the solutions recently adopted at DFCTI in order to face the new aforementioned challenges encountered by three of the LHC experiments, namely ALICE, ATLAS and LHCb. Special emphasis is put on the use of appropriate local resource management systems and the software tuning for optimizing the data analysis. Also, an overview of the RO-LCG current status and its contribution to the WLCG collaboration is included.