In recent years, there has been a strong drive to bring together Cloud computing and Internet of Things (IoT) into an integrated and ubiquitous computing platform. IoT systems envision a world with billions of objects having the capabilities for real-time data collection, processing and intelligent decision-making. In order to achieve its goal of providing intelligence based on Artificial Intelligence (AI) algorithms, IoT systems have relied on the computing resources provided by the cloud computing platforms, for processing the huge volume of data, by transporting IoT data to the geographically distributed cloud data centers to support delay tolerant IoT services. The IoT based intelligence has leveraged the scalability, performance and global service accessibility capabilities of the cloud. However, this approach has proved to lead to significant delays and traffic congestion when used for delay sensitive applications and thus cannot satisfy the requirements of ultra-low delay sensitive IoT applications such as real-time computer vision for smart city security, Virtual Reality games, emergency applications, etc. Hierarchical edge computing can address this problem by integrating IoT data processing and AI algorithms as a microservice close to the IoT devices. Thus, the overall goal of the AIMS project is to design, develop and prototype an advanced innovative data-driven platform integrating AI, Cloud and IoT technologies to support efficient, ultra-low delay sensitive, future AI applications as a microservice. It aims to push intelligence to the edge of things by enabling the development of IoT services integrating composable AI based microservices, opening up new applications and ensuring re-usable IoT services. Toward the smart edge of things, edge intelligence can be developed as independently deployable microservices based on message driven communications, allowing easier extensibility, interoperability, and composition of high-level IoT services.
Project leader: Kashif Mahmood
Institution: Telenor Research