ATLAS computing on CSCS HPC

21st International Conference on Computing in High Energy and Nuclear Physics

物理学;计算机科学

Filipcic, A.^1 ; Haug, S.^2 ; Hostettler, M.^2 ; Walker, R.^3 ; Weber, M.^2

Jozef Stefan Institute, Ljubljana, Slovenia^1

Albert Einstein Center, University of Bern, Bern, Switzerland^2

Ludwig Maximilian University of Munich, Munich, Germany^3

关键词: ATLAS experiment;    Detector simulations;    GPU accelerations;    GPU accelerators;    Job submission;    Possible futures;    Technical solutions;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/664/9/092011/pdf
DOI  :  10.1088/1742-6596/664/9/092011

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

The Piz Daint Cray XC30 HPC system at CSCS, the Swiss National Supercomputing centre, was the highest ranked European system on TOP500 in 2014, also featuring GPU accelerators. Event generation and detector simulation for the ATLAS experiment have been enabled for this machine. We report on the technical solutions, performance, HPC policy challenges and possible future opportunities for HEP on extreme HPC systems. In particular a custom made integration to the ATLAS job submission system has been developed via the Advanced Resource Connector (ARC) middleware. Furthermore, a partial GPU acceleration of the Geant4 detector simulations has been implemented.