Open AccessAdaptive threshold modeling algorithm for monitoring indicators of power network server based on Chebyshev inequality
Author(s) -
Benran Hu,
Yanjun Li,
Jiyuan Ren,
Yiqun Li
Publication year - 2020
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 - 10.1088/1742-6596/1693/1/012089
Subject(s) - chebyshev's inequality , computer science , chebyshev filter , threshold model , set (abstract data type) , kernel density estimation , power (physics) , kernel (algebra) , algorithm , real time computing , data mining , inequality , mathematics , statistics , machine learning , linear inequality , mathematical analysis , physics , kantorovich inequality , quantum mechanics , combinatorics , estimator , computer vision , programming language
The business system server under the IT automation operation and maintenance platform generates massive data samples, based on which the threshold can be set to realize the allocation and management of hardware resources. The traditional threshold selection method is to determine an appropriate threshold based on human experience. If the threshold is too high, it will not play its due role. But if the threshold is too low, it will frequently produce false positives. To solve this problem, an adaptive threshold method based on Chebyshev inequality theory combined with kernel density estimation is proposed to determine monitoring indexes, and a new dynamic implicit threshold model is established to analyse the data generated by the business system server for real-time monitoring and alarm processing. Through the experimental study on the CPU utilization data of the power grid server, lower missing and false positive rate are obtained, which verifies the feasibility and effectiveness of the proposed method.