Agent-Based Fog Computing: Gossiping, Reasoning, and Bargaining

Fog computing is a layered model that consists of fog nodes residing between cloud data centers and Internet of Things ( IoT ) devices. With a large number of highly distributed fog nodes and IoT devices, decentralized and computationally efficient techniques for resource discovery and selection are necessities of fog computing. Since consumers are charged for using fog and cloud resources, economically efficient resource pricing mechanisms are crucial. This work devises: 1) an agent-based fog computing model , 2) a gossip-based resource discovery algorithm for propagating resource requests, 3) a reasoning technique for determining similarities between user requirements and resource profiles, and 4) a bargaining mechanism for dynamically pricing vertically arranged fog and cloud resources. Mathematical proofs validate that the gossip algorithm and reasoning technique are computationally efficient , and the bargaining mechanism is economically efficient . Since agents can determine resource similarities independently and propagate requests autonomously without centralized control, agent-based resource discovery and selection approaches are decentralized techniques .