Yamato proposes “a server architecture recommendation and automatic performance verification technology, which recommends and verifies appropriate server architecture on an infrastructure as a service (IaaS) cloud with bare metal servers, container-based virtual servers, and virtual machines.”

From the introductory section, it’s clear the paper is based more on the author’s experience with ongoing research as part of the NTT Software Innovation Center, and some literature review. Though the focus is the IaaS platform OpenStack, the reader is introduced to CloudStack and Amazon Web Services (AWS) for comparisons. Further qualitative comparisons of bare metal, container, and hypervisor are provided. The author also compares bare metal, Docker, and KVM, and the analysis proves that “because bare metal shows better performance than the other two types of servers, it is suitable to use large-scale database processing or real-time processing.”

The author’s proposed system is composed of server architecture recommendations and automatic verification functions (SAFs), which is an eight-step process. The contribution of the paper is that a technology is developed that “creates virtual resources based on a heat concrete template and executes performance verification test cases.”

The research approach used--through performance evaluation of the proposed method, setting up performance measurement conditions and performance measurement environments, and finally reviewing performance measurement results--provides an in-depth framework that another researcher could use to prove the authenticity of this work.

A limitation of this work is the fact that it assumes extensive understanding of fundamental knowledge in cloud technology. The validation of the developed automatic performance verification technology on the IaaS cloud could have been more aggressive if implemented in real production environments to fully validate the proof of concept.