Home > Research > Publications & Outputs > Energy-Aware Placement of Device-to-Device Medi...

Links

Text available via DOI:

View graph of relations

Energy-Aware Placement of Device-to-Device Mediation Services in IoT Systems

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published
Publication date25/11/2021
Host publicationService-Oriented Computing: 19th International Conference, ICSOC 2021, Virtual Event, November 22–25, 2021, Proceedings
EditorsHakim Hacid, Odej Kao, Massimo Mecella, Naouel Moha, Hye-young Paik
PublisherSpringer
Pages335-350
Number of pages16
ISBN (Print)9783030914301
<mark>Original language</mark>English
EventInternational Conference on Service-Oriented Computing, ICSOC 2021 - Virtual Event
Duration: 22/11/202125/11/2021
https://link.springer.com/book/10.1007/978-3-030-91431-8

Conference

ConferenceInternational Conference on Service-Oriented Computing, ICSOC 2021
Period22/11/2125/11/21
Internet address

Conference

ConferenceInternational Conference on Service-Oriented Computing, ICSOC 2021
Period22/11/2125/11/21
Internet address

Abstract

Internet-of-Things (IoT) systems are becoming increasingly complex, heterogeneous and pervasive, integrating a variety of physical devices, virtual services, and communication protocols. Such heterogeneity presents an obstacle especially for interactions between devices of different systems that encounter each other at run time. Mediation services have been proposed to facilitate such direct communication by translating between messaging protocols, interfacing different middlewares, etc. However, the decision of where to place a mediation service within an IoT topology has repercussions and is in some cases critical for satisfying system objectives. In this paper, we propose an integer linear programming solution to optimize the placement decision specifically in terms of energy consumption. Our solution takes into account the energy consumed by each interaction at each device along the data transfer paths. Through simulations that use topologies of real-world IoT systems, we show the effect of our approach on energy consumption, messaging delay, and placement decision time. Our algorithm outperforms a state-of-the-art solution in terms of reducing energy consumption by almost a third in large-scale typologies. We also demonstrate the feasibility of our approach in terms of overhead.