Many open-source software are included in commercial software. Also, several open-source software are used in the cloud service such as OpenStack and Eucalyptus from standpoint of the unified management, cost reduction and maintainability. In particular, the operation phase of cloud service has a unique feature with uncertainty such as big data and network connectivity, because the operation phase of cloud service changes depending on many external factors. On the other hand, the effective methods of performance assessments for cloud service have only a few presented. Recently, edge computing is the focus of attention because of the problems of connection and processing delay in case of cloud computing. It is known as that cloud computing treats big data. On the other hand, edge computing operates on instant data. We focus on the performance assessments based on the relationship between the cloud and edge services operated by using several open-source software. Then we propose a two-dimensional stochastic differential equation model considering the unique features with uncertainty from big data under the operation of cloud and edge services. Also, we analyse actual data to show numerical examples of performance assessments considering the network connectivity as characteristics of cloud and edge services. Moreover, we compare the noise terms of the proposed model for actual data.

中文翻译：

---

基于随机微分方程和努力数据的边缘计算性能评估

许多开源软件都包含在商业软件中。此外，云服务中使用了多个开源软件，例如OpenStack和Eucalyptus从统一管理、降低成本和可维护性的角度考虑。尤其是云服务的运营阶段具有独特的特点，具有大数据、网络连通性等不确定性，因为云服务的运营阶段会因诸多外部因素而发生变化。另一方面，对云服务进行性能评估的有效方法却寥寥无几。最近，由于云计算的连接和处理延迟问题，边缘计算成为关注的焦点。众所周知，云计算处理大数据。另一方面，边缘计算对即时数据进行操作。我们专注于基于使用多个开源软件运行的云和边缘服务之间的关系的性能评估。然后我们提出了一个二维随机微分方程模型，考虑到云和边缘服务运行下大数据的不确定性的独特特征。此外，我们分析了实际数据，以显示考虑网络连接性作为云和边缘服务特征的性能评估的数值示例。此外，我们比较了所提出模型的实际数据的噪声项。