Edge Computing: Challenges, Research Focus, and Related Paradigms

Edge computing is presented as the next generation computing model after distributed, grid, and cloud computing, integrating cloud, network, terminal, and intelligence to optimize resource allocation, improve service intelligence, and enable deep collaborative scheduling across computing, storage, transmission, and applications.

The article identifies six major motivations for edge computing: (1) the need for high reliability and low latency in data‑intensive applications such as industrial IoT and VR/AR; (2) excessive energy consumption of centralized data centers with low utilization; (3) the growing pressure of big‑data processing due to massive unstructured data and IoT device proliferation; (4) low utilization of cloud resources caused by time‑, application‑, and region‑specific demand fluctuations; (5) the emergence of intelligent front‑ends that can perform local AI inference; and (6) security and privacy challenges associated with transmitting sensitive data to centralized clouds.

By distributing computation and storage toward the network edge, edge computing can alleviate these issues, reduce latency, lower energy use, and enhance data privacy.

The article then surveys related paradigms:

Fog Computing – introduced by Cisco in 2011, fog computing extends cloud services to the edge with a horizontal, system‑level architecture that provides nearby compute, storage, control, and networking functions, often using low‑performance, dispersed devices.

Mobile Edge Computing (MEC) – defined by ETSI, MEC leverages 5G architectures to bring cloud capabilities to the mobile access network, delivering high‑performance, low‑latency services for mobile users and supporting the convergence of mobile and IoT traffic.

Sea Computing – a concept proposed by the Chinese Academy of Sciences in 2010, sea computing envisions computation among physical world objects, forming a layered architecture that spans perception, self‑organization, and distributed intelligent processing.

Intelligent Edge Computing (AI‑EC) – a 5G‑edge platform that offers AI inference resources (vision, audio, NLP) at the base‑station level, enabling on‑site AI processing and reducing the need to send raw data to central clouds.

Overall, the article emphasizes that edge computing complements cloud computing by providing decentralized, distributed resources that address the identified challenges and support emerging technologies such as IoT, AI, and 5G.