An Auction-Based Mechanism for Task Offloading in a Secure Fog-Cloud Environment

With the rapid growth of terminal devices, fog-cloud systems become promising for solving delays and congestions in task provisioning. When the fog computing resources are insufficient, task offloading to the remote cloud can be performed to improve performances. Auctions are considered as a convenient tool to provide incentives for task offloading, and efficiently allocate resources in the fog-cloud environment. However, due to virtualization, the cloud segment in the fog-cloud system is prone to malicious attacks. Security requirements are often in contrast with performance requirements since the operation of security mechanisms consumes a part of computation capacities. Therefore, a comprehensive study is needed to address resource allocation with a security assessment. In this paper, we propose a novel simulation model for resource allocation in the hierarchical fog-cloud system with task offloading. To improve the experience and prevent performance deterioration, we introduce task differentiation into delay-sensitive and delay-tolerant tasks. Resource allocation of the fog and the cloud layer is based on the truthful double auction. A Vicrey-Clarke-Grove (VCG) driven resource allocation is established for winner determination. The proposed simulation model is used to analyze utility functions in the observed fog-cloud environment depending on the offered security level.