Task Allocation for Energy Optimization in Fog Computing Networks With Latency Constraints

Fog networks offer computing resources of varying capacities at different distances from end users. A Fog Node (FN) closer to the network edge may have less powerful computing resources compared to the cloud, but the processing of computational tasks in FN limits long-distance transmission. How should the tasks be distributed between fog and cloud nodes? We formulate a universal non-convex Mixed-Integer Nonlinear Programming (MINLP) problem minimizing task transmission- and processing-related energy with delay constraints to answer this question. It is transformed with Successive Convex Approximation (SCA) and decomposed using the primal and dual decomposition techniques. Two practical algorithms called Energy-EFFicient Resource Allocation (EEFFRA) and Low-Complexity (LC)-EEFFRA are proposed and their effectiveness is tested for various network and traffic scenarios. Using EEFFRA/LC-EEFFRA can significantly decrease the number of computational requests with unmet delay requirements when compared with baseline solutions (from 48% to 24% for 10 MB requests). Utilizing Dynamic Voltage and Frequency Scaling (DVFS) minimizes energy consumption (by one-third) while satisfying delay requirements.

Automated summary

Fog networks provide varying computing resources at different distances from end users. This study focuses on the task distribution between fog and cloud nodes and proposes algorithms to minimize task transmission and processing energy while satisfying delay constraints. The results show a significant decrease in the number of computational requests with unmet delay requirements and reduced energy consumption using the proposed algorithms.