In today's IoT infrastructures, increasingly newly added computational resources at the edge of a network are added to acquire faster response and increased privacy. Such edge networks bring an opportunity for deploying edge application services in proximity to IoT domains and the end-users. In this paper, we consider the problem of utilizing various computational resources established by multiple heterogeneous edge devices in dynamic edge networks. A new lightweight decentralized mechanism (i.e., configurator) is required to monitor an edge infrastructure to enable deploying, orchestrating, and monitoring edge applications at the edge. In this setting, one critical task is to determine the node where the configurator should be placed (deployed) and run (executed) at the edge. In this paper, we propose an efficient approach that solves the configurator's placement problem on the most suited edge device in a given dynamic edge network. Our approach supports the system coping with the dynamicity and uncertainty of the environment and adapts based on the configurator's service quality. We discuss the architecture, processes of the approach, and the simulations we conducted to validate its feasibility.
|Title of host publication||Proceedings of IEEE 2nd International Conference on Cognitive Machine Intelligence|
|Publication date||Oct 2020|
|Publication status||Published - Oct 2020|
|Event||2nd IEEE International Conference on Cognitive Machine Intelligence, CogMI 2020 - Virtual, Atlanta, United States|
Duration: 1 Dec 2020 → 3 Dec 2020
|Conference||2nd IEEE International Conference on Cognitive Machine Intelligence, CogMI 2020|
|Period||01/12/2020 → 03/12/2020|
|Series||Proceedings - 2020 IEEE 2nd International Conference on Cognitive Machine Intelligence, CogMI 2020|
Bibliographical noteFunding Information:
Research partially supported by the Smart Communities and Technologies (Smart CT) at TU Vienna and the EU H2020 Marie Skłodowska-Curie grant No. 764785 FORA–Fog Computing for Robotics and Industrial Automation.
© 2020 IEEE.
- Edge Computing
- Internet of Things
- Resource Management