[Paper Review] Genetic-based optimization in Fog Computing: current trends and research opportunities

Carlos Guerrero, Isaac Lera+1 more|arXiv (Cornell University)|Dec 3, 2021

IoT and Edge/Fog Computing183 references32 citations

TL;DR

This paper presents a systematic review of genetic algorithm (GA)-based optimization in fog computing, analyzing 70 recent studies to identify trends, design patterns, and research gaps. It proposes a taxonomy for optimization scope and GA design, revealing underexplored areas like data management and service migration, and calls for advanced parallel, hybrid, and adaptive GAs tailored to fog's heterogeneity and distribution.

ABSTRACT

Fog computing is a new computational paradigm that emerged from the need to reduce network usage and latency in the Internet of Things (IoT). Fog can be considered as a continuum between the cloud layer and IoT users that allows the execution of applications or storage/processing of data in network infrastructure devices. The heterogeneity and wider distribution of fog devices are the key differences between cloud and fog infrastructure. Genetic-based optimization is commonly used in distributed systems; however, the differentiating features of fog computing require new designs, studies, and experimentation. The growing research in the field of genetic-based fog resource optimization and the lack of previous analysis in this field have encouraged us to present a comprehensive, exhaustive, and systematic review of the most recent research works. Resource optimization techniques in fog were examined and analyzed, with special emphasis on genetic-based solutions and their characteristics and design alternatives. We defined a classification of the optimization scope in fog infrastructures and used this optimization taxonomy to classify the 70 papers in this survey. Subsequently, the papers were assessed in terms of genetic optimization design. Finally, the benefits and limitations of each surveyed work are outlined in this paper. Based on these previous analyses of the relevant literature, future research directions were identified. We concluded that more research efforts are needed to address the current challenges in data management, workflow scheduling, and service placement. Additionally, there is still room for improved designs and deployments of parallel and hybrid genetic algorithms that leverage, and adapt to, the heterogeneity and distributed features of fog domains.

Motivation & Objective

• To address the growing need for systematic analysis of genetic algorithm (GA)-based optimization in fog computing, a field with increasing research but no prior comprehensive survey.
• To identify key challenges and research gaps in fog resource optimization, particularly in data management, workflow scheduling, and service placement.
• To provide a structured taxonomy for optimization scope and GA design to guide future research and development in fog computing.
• To evaluate the strengths and limitations of existing GA-based approaches in fog environments, focusing on scalability, heterogeneity, and real-time performance.
• To propose future research directions, including parallel, hybrid, and adaptive GAs, and integration with emerging paradigms like osmotic computing.

Proposed method

• Conducted a systematic literature review using Google Scholar and Web of Science to identify 70 relevant papers on GA-based optimization in fog computing.
• Developed a two-tiered taxonomy: one for optimization scope (e.g., service placement, workflow scheduling, data management) and another for GA design (e.g., single-objective, multi-objective, parallel, hybrid GAs).
• Classified and analyzed each paper based on the defined taxonomies, focusing on optimization objectives, fitness functions, and genetic operators (crossover, mutation, selection).
• Assessed the benefits and limitations of each approach in terms of performance, scalability, and adaptability to fog-specific constraints like device heterogeneity and network asymmetry.
• Identified recurring patterns in GA implementation, such as the dominance of NSGA-II and standard-weighted GAs, and highlighted underutilized approaches like multi-population and distributed GAs.
• Synthesized findings into future research directions, emphasizing the need for advanced GA designs that better exploit fog infrastructure characteristics.

Experimental results

Research questions

• RQ1 What are the dominant optimization scopes in GA-based fog computing research, and how have they evolved over time?
• RQ2 How are genetic algorithms designed and implemented in fog environments, and what are the most common variants (e.g., NSGA-II, single-objective GA) used in practice?
• RQ3 What are the key limitations and challenges in current GA-based approaches for fog resource optimization, particularly regarding scalability, heterogeneity, and real-time constraints?
• RQ4 Which optimization areas in fog computing remain underexplored despite growing demand, and what opportunities exist for future research?
• RQ5 How can future GA designs be enhanced to better address the distributed, heterogeneous, and dynamic nature of fog infrastructures?

Key findings

• The most frequently studied optimization scopes are service placement (20.0%), workflow scheduling (18.6%), and service orchestration (15.0%), indicating strong research focus on service management.
• NSGA-II and standard-weighted GAs are the most widely used GA variants, with 22.9% and 20.0% of papers respectively, suggesting limited exploration of alternative GA designs.
• Only 6.7% of the surveyed papers address data management, highlighting a critical research gap despite the increasing volume of data in fog environments.
• Service migration is under-researched, with only 4.3% of papers addressing it, despite its importance in mobile and user-centric applications.
• Parallel and distributed GA approaches are underdeveloped, with only three studies proposing parallel GAs—none focusing on multi-population or fine-grained distribution strategies.
• Hybrid and adaptive GA designs are largely unexplored, indicating a significant opportunity to improve performance and convergence in complex fog environments.