A clustering algorithm for multidimensional time series devices based on feature extraction

With the increasing number and types of data center equipment, the use of centralized monitoring has some disadvantages, such as network bandwidth limit and excessive data processing pressure of the monitoring server. The equipment grouping according to the running status and trend of the equipment...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Journal of physics. Conference series Ročník 2849; číslo 1; s. 12112 - 12118
Hlavní autori: Li, Lei, Zhang, Yong, Zheng, Yiming, An, Gaoxiang, Guo, Feng, You, Ziyi
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Bristol IOP Publishing 01.09.2024
Predmet:
Algorithms
Clustering
Data processing
Fault detection
Feature extraction
Monitoring
Time measurement
Time series
ISSN:1742-6588, 1742-6596
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:With the increasing number and types of data center equipment, the use of centralized monitoring has some disadvantages, such as network bandwidth limit and excessive data processing pressure of the monitoring server. The equipment grouping according to the running status and trend of the equipment can meet the dynamic changes of the equipment, and the equipment with similar running status is divided into groups for monitoring, which greatly improves the sensitivity of fault detection. Therefore, this paper designs a multidimensional time series device clustering algorithm based on feature extraction. First, a feature extraction algorithm of time series fluctuations based on correlation measure is designed to extract the subject running trend of monitoring data, reduce the dimension of time series data, and improve the accuracy of clustering. Second, a multidimensional time-series clustering algorithm based on Fuzzy C mean is designed to obtain the optimal clustering results. The experimental simulation results show that the proposed algorithm improves the clustering effect and can more accurately cluster the devices with similar running characteristics into one group for monitoring.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/2849/1/012112