Perbandingan Algoritma K-Means Dan K-Medoids Pada Pengelompokan Humidity, Temperature, Dan Voltage Di Data Center Perawang

Abstract

A data center is a facility managed by a company for data storage (database) and telecommunications from all computer system components and needs. Monitoring the level of humidity, temperature, and voltage is needed to support the performance of the data center and can be done with AKCP. Because of that, it is necessary to group Humidity, Temperature and Voltage of Perawang DC to optimize the monitoring process. Various methods are used to make it easier for companies to determine the best grouping of performance data found in the data center. In this research, data was obtained from Systemlog AKCP PT. Arara Abadi, Perawang from January 21, 2022 – March 19, 2022. This research is expected to make it easier for companies to determine which algorithm is the right one for grouping air humidity, temperature and voltage levels in the Perawang data center by comparing two algorithms namely K-Means and K-Medoids. Based on the research results, K-Means is better in grouping the Humidity, Temperature and Voltage data of Perawang DC in PT. Arara Abadi, Perawang because it has accurate cluster accuracy compared to K-Medoids with a DBI value of 0.306 in the K=2 experiment and the process time is only 1 minute 22 seconds.