1) Description about Fog computing
Fog computing operate at ends of network, not like cloud computing which operates from a central location. It places some resources and processes on edges of cloud. It is also known as edge computing. Due to some limitations related to technical and infrastructure aspects, there are some services and applications which could not fit in paradigm of cloud, to address them, paradigm of cloud computing to edge of network is extended by fog computing.
2) Difference between Fog computing and cloud computing
In Cloud Computing, there are different set of machines distributed and running at various locations while connected to a single hub service or a network.
In Fog computing, one or more than one collaborative multitude of customers carried storage, configuration, measurement, communication, control, and management in a substantial amount.
3) Applications of fog computing
There are some tech giants such as IBM which are driving force behind fog computing. These days in offices, data centre is a common thing or you can see hundreds of devices are connected and communicated to each other and it is expecting that in next few years this number could raised to thousands or lakhs. There are possibilities that direct user-end computing and communication will soon become more relevant.
Here are some practical examples for you to understand where fog computing can be applied:
- It permits prompt interaction between machine and humans, between machine and machine with cooperation of cloud.
- The activities which are going in the cities, on their levels, Sensor data can be obtained by fog computing and then it integrates all network entities which are working independently within those cities.
- As per reports of Markets, it is predicted that by 2017, cloud computing market in field of healthcare will reach $5.4 billion and same thing on a local level will be permit by fog computing.
Fog computing laid focus on a big scale on processes which brings us close to location of data as much as possible. Unvaluable data must be stopped from reaching to cloud and only data which has some worth or has some value should be reach to networks of cloud computing.
Models which helps us to learn machines, through them fog computing is done in best manner which gets training on a small part of data on the cloud. If this model finds suitable, then it is send nearer to the devices. On these devices, there are some algorithms such as decision trees and fuzzy logic can be used on a local level in order to take decision which is less expensive as compared setting up of an infrastructure in cloud which is required to deal with data acquired from tens of thousands of devices.
4) What is next for fog computing?
If you want services on immediate basis, fog computing is a good way. It can also be used in place of internet as speed of net is dependent on carriers.
For future, large companies such as Facebook, Google are looking for alternative methods such as drones and balloons to access web in case any network issue occurs. And small scale organizations are planning to develop a fog of devices which are currently present around them so as to establish close and prompt connections by which resources can be computed.
Aggregated and centralized cloud computing will certainly has a place but as sensors moved to things and data also grows enormously, thus in order to host applications, a new and a fresh approach is required. Right approach for hosting fresh and new applications which could make use of existing devices is Fog computing.
Although relevance of centre is not reduced because of the movement to edge. And on other hand, it reflects that to expand computer architecture, data centre must be a great nucleus.