Treffer: A Self-learning Approach for Service Offloading in Heterogeneous Fog-Integrated Cloud System.

The service requirements of the Internet of Things (IoT) are fulfilled by the Fog-Cloud system. Service placement in fog-integrated cloud environment is NP-class problem due to dynamism in resource availability and workload. In recent, Deep Reinforcement Learning (DRL) algorithms have gained popularity for addressing service scheduling problems due to their adaptability to the dynamic environment. This work proposes a DRL-based Twin Delayed Deep Deterministic Policy Gradient (TD3) model for service offloading in a fog-integrated cloud. While the standard TD3 model exhibits certain limitations, this work addresses those by incorporating few key enhancements. Additionally, Transfer Learning (TL) is applied to minimize the computational complexity of multi-agent based TD3 model. The proposed model has been simulated in Python platform using Google trace-2019 data. The efficacy is measured on various Quality of Service (QoS) metrics such as latency, energy, computing cost, and deadline violation, where it produces on average 14% 12%, 16%, and 1.5% better results respectively as compared to state-of-art models. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Network & Systems Management is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)