A Resource Estimation Method in Multi-Cloud Environment with a Model based on a Repairable-Item Inventory System

Multi-cloud services developed by microservices based applications have become increasingly popular using container virtualization. Here, managing multi-cloud services is complicated in resource management, such as how to allocate and age computing resources efficiently. To solve this problem, we propose a model of multi cloud services for resource estimation and show an approximate solution method using this model. This approximate solution method can estimate the amount of resources in a multi-cloud under given conditions. in this paper, we propose a model of services in multi-cloud developed by a orchestration tool, using a model of the existing repairable-item inventory system. For the model, we also show the approximate solution method used in the analysis of this existing repairable-item inventory system. This approximate solution method obtains the amount of resources in the multicloud and enables resource estimation for the target model. The approximation accuracy is verified by comparing the proposed method with Monte Carlo simulation results. Also, we show that the proposed method takes less computation time to obtain the same accuracy results with Monte Carlo simulation results. Through the verification, we show the usefulness of the proposed model and the approximate solution method.