Fog computing an architecture that uses one or more collaborative end-user clients or near-user edge devices to carry out a substantial amount of storage.
Characteristics of fog computing include close proximity to end devices, geographical distribution and mobility support. This approach decreases loads on data center and network, decreases delays in service and increases quality.
The concept of fog computing is especially important for Industrial Internet of Things, where speed and reliability of data transmission become critical (production automation, transport, sensor and devices networks). Fog computing provides distributed information processing depending on required reaction speed for events and relevancy of events for the functioning of the system as a whole.
Typical examples of fog computing architecture application:
Provision of connectivity for legacy equipment not supporting IP
Connection of a number of systems, which require analytics, related to correlation of events from such systems, with minimum time delays in transmission and processing of information
Connection of equipment, which require analytics, related to filtration of incoming signals
Provision of full-featured security for systems and separate equipment
In case a parallel data collection and processing system is required, in addition to existing ones, when existing system cannot provide necessary functions (ex.: to increase discreteness of information), or there is a risk to disrupt current information collection system during its modification