Edge-Computing Architectures for Internet of Things Applications: A Survey

Area of Science:

• Computer Science
• Networking
• Distributed Systems

Background:

• The proliferation of Internet of Things (IoT) devices generates vast amounts of data, overwhelming limited device resources.
• Traditional cloud computing introduces latency issues for time-critical IoT applications.
• Edge-cloud computing emerges as a solution, enabling distributed data processing and storage.

Purpose of the Study:

• To conduct an in-depth study and classification of edge-computing architectures for IoT (ECAs-IoT).
• To compare ECAs-IoT based on data placement, orchestration, security, and big data handling.
• To identify capabilities, features, and gaps by mapping ECAs-IoT to existing IoT layered models.

Main Methods:

• In-depth analysis and classification of existing ECAs-IoT.
• Comparative study of architectures based on key features.
• Mapping ECAs-IoT to established IoT layered models.
• Identification of limitations and proposal of solutions.

Main Results:

• A comprehensive classification of ECAs-IoT based on critical factors.
• Detailed comparison highlighting the strengths and weaknesses of various architectures.
• Identification of current limitations and research gaps in ECAs-IoT.
• Mapping reveals specific capabilities and areas for improvement.

Conclusions:

• Edge-cloud computing is crucial for addressing IoT data processing and latency challenges.
• The study provides a framework for understanding and selecting appropriate ECAs-IoT.
• Recommendations are offered for overcoming existing limitations and future research directions.
• Four distinct scenarios for ECAs-IoT application are proposed.