Novel infrastructure with common API using docker for scaling the degree of platforms for smart community services

The development of smart communities has diversified not only service execution platforms but also the resolvers of multiple service requirements, each of which has different requirements in terms of processing delay, anonymity, computational cost, the amount of data at a given level of granularity, etc. To meet these requirements, an infrastructure that easily performs service migration and provides services with the correct processing nodes using IP-independent distributed processing methods such as Authorized Stream Contents Analysis (ASCA) is becoming a pressing need of smart communities. ASCA is an advanced method of analyzing packet streams and filtering necessary packet streams according to the marker tags in the contents of the streams under the Opt-In manner. Moreover, smart communities require that every service be able to perform ASCA and gather necessary data because of the diversified nature of the services. Consequently, in this paper, we have implemented a service infrastructure using a Docker container that facilitates service migration and provides services with a common Application Programming Interface (API) using ASCA. The API provides a process throughput of over 60 Gbps on a Docker container using Zero-Copy mechanism.

I. Introduction

The proliferation and deployment of Information and Communication Technology (ICT) in the infrastructure of cities has increased interest in smart communities [1] [2]. The goal of smart communities is to improve the Quality of Life (QoL) of the residents of a community. To achieve this goal, the development of smart community services is indispensable. Consequently, in recent years, the design of new smart community services has been actively discussed [3]: electric power system stabilization with a large amount of information represented by the Internet of Things (IoT), traffic management systems, and monitoring services for children and the elderly.