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
.