ATLAS I/O performance optimization in as-deployed environments

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

物理学;计算机科学

Maier, T.^1 ; Benjamin, D.^2 ; Bhimji, W.^3 ; Elmsheuser, J.^1 ; Van Gemmeren, P.^4 ; Malon, D.^4 ; Krumnack, N.^5

Ludwig-Maximilians-Universität München, Germany^1

Duke University, United States^2

University of Edinburgh, United Kingdom^3

Argonne National Laboratory, United States^4

Iowa State University, United States^5

关键词: ATLAS experiment;    Integrated programs;    Large scale physics;    Performance optimizations;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/664/4/042033/pdf
DOI  :  10.1088/1742-6596/664/4/042033

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

This paper provides an overview of an integrated program of work underway within the ATLAS experiment to optimise I/O performance for large-scale physics data analysis in a range of deployment environments. It proceeds to examine in greater detail one component of that work, the tuning of job-level I/O parameters in response to changes to the ATLAS event data model, and considers the implications of such tuning for a number of measures of I/O performance.