Affiliations: Chongqing College of Electronic Engineering, Chongqing, China
Correspondence:
[*]
Corresponding author. Zhong Yi, Chongqing College of Electronic Engineering, Chongqing, 401331, China. E-mail: ukv9313@163.com.
Abstract: With the rapid development of the IoT, the traditional traffic scheduling optimization model is difficult to adapt to the development needs of emerging services, bringing new challenges and problems to data center management. In order to solve the problem of data traffic management in data center network, a clustering algorithm is constructed to analyze its key technologies. The algorithm divides the data to be observed into a certain number of “class clusters” by some predetermined features, so that the similarity of the data in the cluster is measured by a certain “distance function” within each “class cluster". By analyzing the RFID automatic radio frequency identification technology, a data classification model based on RFID automatic radio frequency identification technology is constructed. The original data of the unbalanced state is processed based on the hierarchical partitioning method, and the sampling data analysis result is obtained. The results of data training experiments on the model show that for the prediction of a few samples, the prediction of the unbalanced data set has been further improved, and the AUC value has reached 98.72%. Research has provided new ideas for the operation and management of data centers.
Keywords: RFID automatic radio frequency identification technology, data center, algorithm optimization
