The Internet of Things (IoT) has grown from a mere concept, into an unstoppable technology platform that encompasses Cloud Computing, connected sensors, devices, and machines. An integral part of emerging technologies like autonomous vehicles, automated production lines, and even smart in-home services, IoT currently consists of around 15 billion devices and sensors. By the year 2020, that number will grow to at least 50 billion, and if you agree with companies like Intel, the number could even be as high as 200 billion.
In IoT, the ability to collect data, process it, and make it relevant to devices, is critical. With traditional closed networks and even off-site cloud systems, there are some significant challenges. Emerging IoT devices require speedy processing of data to control automations and provide insights for operators. In some critical industries, the current solutions do not work fast enough. Fog Computing is an emerging type of architecture that enables efficiency in IoT, while overcoming some of the inherent weaknesses of current solutions.
Downsides of Cloud Computing
Storage, communication, control, management, and configuration, are four essential tasks that any network must be able to accommodate for effective IoT implementation. Up until this time, Cloud Computing has been seen as the leading enabler of IoT, and while the cloud does provide a number of benefits, there are also some key areas where Cloud Computing is underperforming with new IoT deployments.
- Cloud Computing means higher latency for devices and sensors.
- Relatively high bandwidth requirements.
- Data and logic processing is performed by remote machines.
- Potentially reduced privacy and security for operators.
- Inefficient for certain applications.
Data and computing requirements are growing exponentially. Centralized servers and datacenters are no longer ideal when you consider that some organizations deploy thousands of individual IoT devices. Fog Computing looks at the current state of IoT and the growing concept of the Internet of Everything (IoE), and presents a new type of network architecture that still leverages off of the cloud, while partially decentralizing storage and logic functions.
Fog Computing makes More Sense for Closed and Critical Systems
Fog Computing has sometimes been described as networks where devices are smarter. With IoT up to this point, many of the embedded sensors and connected machines have been based around simple designs that leverage off of the processing and storage capabilities of a central server. This increases latency, creates a problem of reliance on the cloud, and as mentioned, introduces security and privacy concerns. Fog Computing takes things back to some of the philosophies that existed before Cloud Computing, without sacrificing interconnectivity between devices. This means that a fog network would rely more on localized storage and processing, with some processing capabilities embedded into the devices themselves.
Cisco, the current leading innovator in Fog Computing, uses a large jetliner as an excellent example of why Fog Computing is necessary. There are thousands of individual sensors installed in a large airliner, monitoring everything from engine and fuel systems, to cabin pressure and avionics. Sending data from these sensors to a cloud system would be inefficient, and introduce latency into the transfer of critical data. It is more important that pilots are able to receive interpreted data that can be displayed in real time. This means that much of the storage and computing power needs to exist within the closed network of the plane, with other relevant data being sent to external systems (an external cloud server, flight control operators etc.). Other forms of transportation, such as autonomous vehicles, also need access to speedy data and logic processing. It has been proposed that roadside machines and sensors on local nodes would be more efficient, and offer less latency than cloud systems. These two examples show how decentralized fog network architecture would provide robust functionality that exceeds what a cloud network could offer.
In essence, Fog Computing allows for big data to be processed locally, or at least in closer proximity to the systems that rely on it. Newer machines could incorporate more powerful microprocessors, and interact more fluidly with other machines on the edge of the network. While fog isn’t a replacement for cloud architecture, it is a necessary step forward that will facilitate the advancement of IoT, as more industries and businesses adopt emerging technologies.
Do you have any experience of FOG applications? Perhaps you can see an area where FOG would be a major advantage – let us know in the comments section below.