Multi-VO support in IHEP's distributed computing environment

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

物理学;计算机科学

Yan, T.^1 ; Suo, B.^1 ; Zhao, X.H.^1 ; Zhang, X.M.^1 ; Ma, Z.T.^1 ; Yan, X.F.^1 ; Lin, T.^1 ; Deng, Z.Y.^1 ; Li, W.D.^1 ; Belov, S.^2 ; Pelevanyuk, I.^2 ; Zhemchugov, A.^2 ; Cai, H.^3

Institute of High Energy Physics, 19B Yuquan Road, Beijing

100049, China^1

Joint Institute for Nuclear Research, Joliot-Curie 6, Moscow region Dubna

141980, Russia^2

Department of Physics, Wuhan University, 299 Bayi Road, Wuhan

430072, China^3

关键词: Accounting system;    Air showers;    Distributed computing environment;    Distributed computing resources;    Hardware cost;    Storage resources;    System administration;    System architectures;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/664/6/062068/pdf
DOI  :  10.1088/1742-6596/664/6/062068

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Inspired by the success of BESDIRAC, the distributed computing environment based on DIRAC for BESIII experiment, several other experiments operated by Institute of High Energy Physics (IHEP), such as Circular Electron Positron Collider (CEPC), Jiangmen Underground Neutrino Observatory (JUNO), Large High Altitude Air Shower Observatory (LHAASO) and Hard X-ray Modulation Telescope (HXMT) etc, are willing to use DIRAC to integrate the geographically distributed computing resources available by their collaborations. In order to minimize manpower and hardware cost, we extended the BESDIRAC platform to support multi-VO scenario, instead of setting up a self-contained distributed computing environment for each VO. This makes DIRAC as a service for the community of those experiments. To support multi-VO, the system architecture of BESDIRAC is adjusted for scalability. The VOMS and DIRAC servers are reconfigured to manage users and groups belong to several VOs. A lightweight storage resource manager StoRM is employed as the central SE to integrate local and grid data. A frontend system is designed for user's massive job splitting, submission and management, with plugins to support new VOs. A monitoring and accounting system is also considered to easy the system administration and VO related resources usage accounting.