Optimal Design of Hierarchical Cloud-Fog&Edge Computing Networks with Caching

Area of Science:

• Computer Science
• Electrical Engineering
• Network Engineering

Background:

• Hierarchical cloud-fog&edge computing (FEC) networks integrate cloud, fog, and edge tiers for distributed computation.
• Devices in FEC networks can utilize multiple computing modes: cache-assisted, cloud-assisted, or joint device-fog&edge.
• Optimizing resource allocation in these complex networks is crucial for energy efficiency and performance.

Purpose of the Study:

• To investigate the optimal design of a three-tier hierarchical cloud-fog&edge computing network.
• To formulate and solve an energy minimization problem by jointly optimizing computing mode selection, local computing ratio, computation frequency, and transmit power.
• To guarantee task completion deadlines, computation capability, and transmit power thresholds.

Main Methods:

• The energy minimization problem, formulated as a mixed integer nonlinear programming problem, was decomposed into three subproblems.
• Optimal solutions for each subproblem were derived.
• An efficient optimal caching, cloud, and joint computing (CCJ) algorithm was proposed to solve the primary problem.

Main Results:

• The proposed CCJ algorithm achieved superior system performance compared to benchmark schemes.
• A smaller achievable transmit power threshold for devices resulted in greater energy savings.
• Increased task data size led to a reduced local computing ratio.

Conclusions:

• The developed CCJ algorithm provides an effective solution for energy minimization in hierarchical FEC networks.
• Network design parameters, such as transmit power thresholds and task data size, significantly impact energy efficiency.
• The study demonstrates the effectiveness of joint optimization in complex, multi-tier computing environments.