Back to search

Harokopio University

Fog node self-control middleware: towards autonomous decision making in fog colonies

Abstract

dc:description

The Fog Computing paradigm brings computational resources closer to the IoT, able to prefilter IoT data before offloading it to the Cloud and able to provide cloud-like functionalities closer to the IoT. Within the Fog Computing paradigm, computational resources, called Fog Nodes, are organized in groups, called Fog Colonies. Traditionally, Fog Colony Control has been heavily centralized, with specialized nodes, named Orchestrators and Fog Controllers, controlling crucial aspects of the Fog Colony's operation, such as is formation, load balancing, and interfacing with the external world to provide services. Although decentralized approaches are often employed in distributed systems, such efforts are not yet applied in Fog Colony Control. However, based on current literature such an aspect seems to be promising. The aim of this research is to explore the benefits of decentralized Fog Colony Control, especially as a colony size is increasing. As centralized control is the prominent paradigm, we focused on enhancing the autonomous operation of Fog Nodes and developed a Self-Control Fog Node middleware in order to efficiently support decentralized Fog Colony Control. To advance their autonomous decision making, Context Awareness has been embedded in the middleware. The concepts of policies and policy rulebooks have been introduced to enable the standardization of operations performed by all nodes, and facilitate the diffusion of node status between colony members. Finally, both centralized and decentralized context diffusion patterns were explored using the ECTORAS Publish/Subscribe protocol developed as part of our research. Two different real-world case studies were utilized as a testbed for the Self-Control Fog Node middleware: A smart management building application and an evacuation application based on IoT devices and sensors. In both cases, the performance of the nodes themselves and of the services provided by the Fog Colony was explored as the colony size increases. This allowed to explore the difference between context-aware and non context-aware versions of the self-control node middleware, as well as decentralized versus centralized context diffusion schemes. Results indicate the potential of a decentralised control in Fog Colonies especially as their size is increasing.

Degree

thesis:*
Grantor dc:publisher
Harokopio University
Year dc:date
2022

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Nikolopoulos, Vasileios
  • Νικολόπουλος, Βασίλειος

Subjects

dc:subject × 14

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/52322

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Nikolopoulos, Vasileios; Νικολόπουλος, Βασίλειος. Fog node self-control middleware: towards autonomous decision making in fog colonies. Harokopio University, 2022. http://hdl.handle.net/10442/hedi/52322