Implementasi Openshift Cluster Menggunakan KVM

Abstract

Now, technological developments are increasing rapidly, one of which is in cloud computing stored on servers. The more users accessing the service, the heavier the server load will be. Therefore, cluster technology is needed. With cluster technology, every incoming request will be divided equally or equally to each server. However, to run a cluster, a cluster orchestration software is needed. In this study, we compare two orchestrations, namely Openshift and Kubernetes. The virtualization technology used is Kernel-Based Virtual Machine (KVM). In this study, we compare latency, CPU usage, and memory usage when accessing the web between Openshift cluster and Kubernetes cluster technologies. The highest latency test results were 18201 ms by Kubernetes cluster and the lowest latency was 1374 ms by Openshift cluster technology. The latency of the Openshift cluster is lower because the nodes are divided into compute, infra, and controller. Good latency is low latency because it can open web pages faster. In standby, the lowest CPU usage and the lowest memory usage are Openshift clusters, namely 0.13% and 60%, this is because in Openshift cluster technology each node is divided into computing, infrared , and controller. In a busy state, the highest CPU usage and the highest memory are Kubernetes clusters, namely 75.14% and 98%, this is because in Kubernetes there is no node sharing.