Open AccessResource Usage Evaluation Study in HPC Cluster
Author(s) -
Jiaxu Guo,
Gaochao Xu,
Xiaodong Fu,
Yang Li,
Peng Liu
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1302/2/022075
Subject(s) - computer science , principal component analysis , entropy (arrow of time) , resource (disambiguation) , cluster (spacecraft) , data mining , computer cluster , field (mathematics) , measure (data warehouse) , node (physics) , supercomputer , distributed computing , artificial intelligence , parallel computing , mathematics , engineering , operating system , computer network , physics , structural engineering , quantum mechanics , pure mathematics
In order to evaluate the running state of computing clusters, this paper proposes a measure index for evaluating the resource utilization of nodes in high-performance computing clusters -- Load Score of Computing Node(LSCN), and compares two methods for determining parameter weights, principal component analysis (PCA) and entropy weight determination method (EWD), in combination with relevant algorithms in the field of machine learning, so as to provide useful guidance for the running and maintenance of computing clusters.