https://doi.org/10.1051/epjconf/201921403048
Multicore workload scheduling in JUNO[1]
1
Institute of High Energy Physics,
19B Yuquan Road,
Beijing 100049,
P.R. China
2
Soochow University,
1 Shizi Street, Suzhou,
JiangSu Province, 215006,
P.R. China
* E-mal: zhangxm@ihep.ac.cn
Published online: 17 September 2019
The Jiangmen Underground Neutrino Observatory (JUNO) is going to apply parallel computing in its software to accelerate JUNO data processing and fully use capability of multi-core and manycore CPUs. Therefore, it is necessary for the JUNO distributed computing system to explore the way to support single-core and multi-core jobs in a consistent way. To support multi-core jobs, a series of changes to the job descriptions, scheduling, monitoring needs to be considered, in which the pilot-based scheduling for a hybrid of single-core and multi-core jobs is the most complicated part. Two scheduling modes and their efficiency are presented and compared in this paper, and also a way to optimize efficiency is provided.
© The Authors, published by EDP Sciences, 2019
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
