Fog Computing: An Overview of Big IoT Data Analytics

Table 1

Key featured difference between fog and cloud.

Features	FOG computing	Edge computing	Cloud computing
Availability of server nodes	Availability high range of servers	Less scalable than fog computing	Availability of few servers
Type of services	Distributed and localized limited and special for specific domain	Mostly uses in cellular mobile networks	Worldwide and global services
Location identification	Yes	Yes	No
Mobility features	Provided and fully supported	Provided and partially supported	Limited
Real-time interaction	Supported	Supported	Supported
Real-time response	Highest	Higher	Lower
Big data storage & duration	Short duration and targeted to specific area	Depends on the scenario of services and applications	Life time duration as its managing for big data
Big data analytic capacity and computation quality	Short time capacity with high level computation functionality	Short time capacity for prioritized computing facilities	Long time capacity only with categorization computing facilities
Working environment & positions	Streets, roadside, home, malls, field tracks (e.g., every Internet existing areas)	Deployed by the specific services provider in specific indoor areas	Indoors with massive components at cloud service provider owned place
Architectural design	Distributed	Distributed	Centralized
Number of users facilitated	Locally related fields (e.g., IIoT, STL devices)	Specific related fields (e.g., mobile users)	General Internet connected users
Major service provided	Cisco IOx,; Intel	Cellular network companies	Google, Amazon, IBM, and Microsoft Azure