widen, considerations are needed for reliable gateways which can help to convert the IoT vision
into a functional and workable reality while ensuring complete uptime guarantee.
IoT can make lights, appliances, doors, windows, and your home objects connect to the
Internet but this also means that the strain on the current data center and internet infrastruc-
ture will also maximize. According to Gartner, by 2020, IoT applications will run through
26billion devices.
One of the common approaches to deal with this is the centralization of cloud data
processing at one site. This helps to improve application security and reduce costs. But this
structure will receive large amounts of information from multiple sources, each positioned at
various sites. Therefore, a backup is necessary.
The majority of the enterprise data is sent to the cloud where it is analyzed and stored;
followed by planning to take the next decision. However, eciency is not one of the hallmarks
of this approach. Hence, in order to add eciency in the mix, a reliable solution is necessary that
can handle voluminous amounts of data “smartly”.
IDC has emphasized on the necessity of smart processing and illustrated the fact that
almost 40 percent of the data is run through analysis on those devices that are “physically” near
to the IoT ecosystems.
Solution
Fog computing oers decision-making, action-taking, and computing for IoT devices and facilitates
to add only the “relevant” data in the cloud. The term Fog Computing was first coined by Cisco.
According to Cisco, the fog helps to bring the cloud nearer to those “things” that generate
and perform actions for the internet of things data. “Fog nodes” was the term that was used to
describe these devices. Fog nodes are deployed at any location that has network connectivity—
it can be a power pole, vehicle, oil rig, factory floor, or a railway track. This means that any
device that has network connectivity, computing, and storage can serve as a fog node. Switch-
es, embedded servers, routers, controllers, and video surveillance cameras are some of the
examples of fog nodes.
To learn the Fog computing architecture, consider the following.
• Runs analysis in the network edges for sensitive data. Hence, the processing is executed
near the edge where the data is produced rather than pushing huge IoT datasets in the
cloud.
• Follows a policy to perform actions on IoT data in milliseconds.
• Forwards the data that is elected at the cloud in order to help with long-term storage and
historical analysis.
Advantages
A fog computing architecture helps IoT ecosystems with the following advantages.
• Saves network bandwidth.
• Helps with latency.
• Improves security for all network levels.
• Gathers and accumulates large amounts of data.
• Helps in the provision of reliable operation with quick actions.
• Routes data to the most appropriate location for processing.
86 Internet of Things
Internet_of_Things_CH04_pp081-104.indd 86 9/3/2019 10:13:30 AM
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.