A few months back everything was in the "cloud." However, the buzzword these days is "fog computing." Fog computing is related to how efficiently data is being stored and accessed. Basically, cloud computing is the ability to store and retrieve data from an off-site location. Cloud computing is a major reason for how traditional phones got "smart." Phones don’t have ample, built-in space to store the information that is needed to access applications and services. All the data is being transmitted to and from the cloud in order to provide the services we need. But, there is a problem with cloud computing technology — it’s the limitation of bandwidth.

A report by the World Economic Forum (WEF) predicted that the U.S. ranks 35th in the world for bandwidth per user. This is a serious issue if you’re trying to transmit data wirelessly.

The idea behind fog computing is an attempt to knock down some of these physical limitations. With fog computing technology, all the processing happens on devices physically closer to where the data is collected, instead of sending data to the cloud. With the evolution of the Internet of Things, more and more physical objects (devices) are being added to the network and they all are connected wirelessly to transmit and receive data.

Fog computing has also been referred to as "edge computing."  Edge computing is supposed to resolve problems by storing data close to the "ground." In other words, it stores data in local computers and storage devices, rather than routing all the information through a centralized DC in the cloud. Fog or edge computing is a prototype campaigned by a few of the leading IoT technology players, such as Cisco, IBM, and Dell. They are pioneers of representing the change in architecture wherein intelligence is pushed from the cloud to the "fog" or "edge." Basically, fog computing is responsible for enabling quick response time, reducing network latency and traffic, and supporting the backbone bandwidth savings to achieve a better Quality of Service (QoS). It is also supposed to selectively relay applicable data to the cloud.

IDC predicts that by the end of 2025 about 45 percent of the world’s data would be moved closer to the network edge. It is believed that fog computing is the only technology that can withstand AI, 5G, and IoT in the years to come.

Another study by IDC estimates that by 2020, 10 percent of the world’s data will be produced by edge devices. This will drive the need for more efficient fog computing solutions that could provide reduced latency.

So, what’s the main difference between edge computing and fog computing? Cisco coined the term "fog computing" and IBM calls it "edge computing." Basically, edge computing is a subset of fog computing. It simply refers to data being processed close to where it originated. Fog computing allows data to be accessed and processed more efficiently, which reduces the risk of data latency.

A brilliant use case for fog computing is the smart traffic light system that has the capability of preventing accidents and can reduce traffic congestion by changing its signals based on its surveillance of incoming traffic. Also, this data is sent for further analysis to the clouds.

The growth of fog computing frameworks provides much more choices to organizations for processing information and data wherever it is appropriate. There are certain applications where data may need to be processed as quickly as possible. For example, in a manufacturing industry where all machines are connected to a network, they need to be able to react to an incident as soon as possible. Fog computing helps to create low-latency network connections between devices and their analytics endpoints. This architecture, in turn, reduces the amount of bandwidth needed when compared to the cloud. It can also be used in scenarios where there is no bandwidth connection needed to transfer data. Hence, data is processed close to where it originated. An added benefit is advanced security features that can be applied by users in a fog network. This will be from segmentation of network traffic to virtually extending firewalls to protect the network.

It would be wise for any enterprise relying on someone else’s data center for storing their data to strongly consider fog computing. They should also figure out how their businesses might get affected, if they continue using traditional ways to store data in the days to come, due to lack of bandwidth to access the data.