The Implementasi Kubernetes Cluster Menggunakan KVM

Abstract

Rapid technological development requires services that can be accessed at any time. The increasing number of client requests will result in the server being down due to more workloads. Therefore, clusters are present to handle increased server workloads and evenly divide each server. However, to run a cluster requires cluster orchestration software that is Kubernetes. Kubernetes is used for container management. The virtualization technology used in Kubernetes is Kernel-Based Virtual Machine (KVM) virtualization. In this research, we compare latency, CPU usage, and memory usage when opening web pages, between conventional technology and Kubernetes cluster. The results of the highest latency test are 16614 ms by Kubernetes cluster and the lowest latency is 1522 ms by conventional technology. In conventional systems, the response from the user is directly received by the web server, processed and then sent. In Kubernetes clusters, there are nodes running on top of virtual machines, so it takes longer to respond. Good latency is low latency because it can open web pages faster. In the standby state, the lowest CPU usage and memory usage occurred in conventional were 0.07% and 28% because there were no nodes running on the virtual machine. In a busy state, the lowest CPU usage occurs on the Kubernetes cluster server is 1.06% due to the division of workload on virtual machines. The lowest memory usage occurs on a conventional server is 57% because there are no nodes running on a virtual machine.