In the illusory world of digital transformation, the question of where the applications will best be hosted often triggers a lively discussion. In particular, the cloud vs. edge debate has left many wondering which direction to choose. Turns out, the answer is not as clear-cut as picking one over the other in the hope of finding a winner. It is a case of ‘horses for courses’ and an efficient enterprise-wide implementation may well require elements of both.
Wikipedia defines edge computing as a distributed paradigm that brings data storage and processing closer to the location where it is needed – the shop floor, for instance. Cloud computing, on the other hand, is generally considered as on-demand access to computer resources available to many different users over the Internet – the data centre.
As an Industry 4.0 platform, cloud applications offer almost unlimited scalability in terms of data storage and computing power. This makes it easy to run data mining and analytics algorithms on plant data with a view to optimising overall process and energy efficiency. It also offers older plants running legacy equipment an easy way to get started with the new technologies of the IIoT. One of the downsides of cloud-based systems is the inherent latency, which is fine for any application in which ‘near’ real-time response is acceptable, but not so good for machine applications that require a true real-time reaction.
A DCS is an example of an edge system (before the term became popularised), but in an IIoT context, edge computing comes into its own where the value of the system is linked to its reaction time. Artificial intelligence and machine learning are good applications.
An example of how artificial intelligence can be incorporated into machines through the use of sensors and data processing at the edge is Forpheus, Omron’s ping-pong playing robot. Forpheus uses its cameras and sensors to observe the mood and movements of the opposing player, as well as the trajectory of the ball. It then rapidly analyses this data to anticipate the opponent’s next shot so that it can hit the return. Through this constant assessment of a person’s play, it determines their skill level and modifies its own game to present an appropriate challenge. Forpheus’ objective is not to beat the other player, but to gauge their skill level and then help them to train and improve. This is an example of how smart machines could be used to assist people to make the most of their potential.
Although edge computing has a distinct advantage over the cloud in these types of application, this does not make it a substitute. For instance, predictive maintenance is a natural target for cloud deployment because there is simply no need for a ‘zero latency’ response. See the article ‘Does edge computing have the edge?’ for more on the cloud vs edge conundrum, and how the two will most likely coexist in the future.
Editor: SA Instrumentation & Control
|Tel:||+27 11 543 5800|
|Fax:||+27 11 787 8052|
|Articles:||More information and articles about Technews Publishing (SA Instrumentation & Control)|
© Technews Publishing (Pty) Ltd | All Rights Reserved