Determining Resource Utilization in Cloud Systems: An Analytical Algorithm for IaaS Architecture

Abstract

Cloud systems have become a key element in today's ICT systems. One solution in this area is to build systems based on IaaS architecture, in which user-dedicated virtual machines are created on available physical devices. When building such systems, it is essential to ensure optimal utilization of individual resources such as RAM, disk, processor, and available bandwidth while limiting their overuse, which can negatively impact the machine's lifespan. Moreover, it is crucial to avoid situations in which some parameters are underutilized while others are highly loaded. Therefore, the authors have developed an analytical algorithm to determine the average utilization of individual component resources of the system during its operation. The proposed analytical algorithm's results were compared with the results obtained through digital simulation of a cloud system based on the IaaS architecture. The proposed algorithm can successfully optimize the system design to ensure appropriate utilization of all its components in the future.