Challenges for building a cloud native scalable and trustable multi-tenant AIoT platform

ABSTRACTThe arrival of 5G together with advances in artificial intelligence, machine learning, cloud computing, virtualization, and service orchestration have created a ubiquitous computing model at the network edge, enabling a host of new, AI driven edge computing applications. Although edge computing shares many characteristics of cloud computing, there are unique challenges for edge computing to meet the ever growing demands for scalability, security and multi-tenancy, especially in the upcoming 5G era. These challenges are discussed through two typical edge computing use cases: streaming video analytics and industrial IoT. A number of open research problems are discussed to call for help from the design automation community with the focus on new automation methodologies in building a cloud-native end-to-end edge computing platform.