Page 38

Industrial Ethernet Book 103

Technology Industrial web-based computing: is data intelligence finally here? Fog computing and cloud computing are no longer strictly processing partitioned, which leaves more power in the control centre at quite minimal costs. Another key advantage is that no longer are very powerful computers needed, nor is there a need for high levels of human intervention or even monitoring. DATA INCLUDING TEXT, PICTURES, RAW BITS and bytes that tell humans things has been around for a long time, from the early grunts and cave drawings, to the printed word. Making sense of it all and to put data to use, on the other hand, takes understanding and processing. Processing needs a brain or a computer to give us information. Why? The amount of data can be huge and to bring it down to a manageable size may lose points of interest, so encapsulating that data in a wider sense with understanding of where the data is from and what it is about helps enormously in forging solutions. The UK Meteorological Office takes 10 million observations on the weather every day but if you watch a weather forecast the information presented rarely takes more than 5 minutes. How is this done; by targeting the audience and encapsulating the raw data in information packets, packets which have been processed, analysed and organised in order to provide succinct details. An interpretation of this is much like one word being able to insinuate many ideas; the term “congested road” gives the picture of a traffic jam or could simply be a road containing many cars difficult to pass, would this also lead to frustration and anger? If we add just one further idea of “congested urban road” then just that one extra piece of data puts the visualisation into context and makes clear the thought on the subject. This leads to the preposition that digital data can provide intelligence. In software the idea of concatenating bits into binary arrays where a 16 bit word contains digital flags indicating many things, can be carried further by defining an object structure of say “Door Operation”, a door being opened or closed and a time for how long the door has been opened or closed. Add to the structure who opened the door and then an extrapolation may be made as to why the door was opened. Such situations are seen as data providing intelligence rather than being intelligent in its own right. What if we pass to a centralised core not real time data but extrapolated data on the subject? First, let us provide some definitions on which we can more easily describe the segmented parts of the subject under discussion. The cloud An ethereal place of residence on networks circling the globe, and today, of near orbit, where servers and storage can be found and used for data derived in a more physical world. Provided and patrolled by an invisible though powerful entity, the community at large, it is very much indefinable though effectively easily usable. Cloud Computing By having the processing carried out in the cloud by multi-user and powerful computers it alleviates the cost of having a powerful computer in the local environment; optimising machine cost and, to some extent, development costs due to the application frameworks or Applications as a Service, more commonly today referred to as Software as a Service (SaaS), provided by cloud computers. Such points as regular backups and validated recovery mechanisms are benefits to this method of computing but there are ongoing regular costs to take into account also for such things as bandwidth usage and amount of data storage. Fog In the Operational Technology (OT) world the fog is so termed as it is, like the climatic condition, is close to the ground or, more precisely, close the real world interface layer at the edge of the industrial topology used to create effective plant networks. Fog computing As the name implies, this is processing done on the data but the computing engine is close to the edge or at the point of data collection. Unlike cloud computing this is very much under the control of the application developer and it is usually left to the developer to not only implement it in the system but also to ensure ancillary services such as connectivity and backups are maintained as well. Grammar Akin to language we now look at how data can be visualised more succinctly. The very definition of grammar is placing meaningful words in a correct contextual sense whereby accepted words of the language are used within an accepted boundary, the context. This also provides the implicit facility to have the same word having several different meanings and raises the question as to how we can utilise similar techniques with the processed data both to aid understanding but also ease the necessary processing to achieve optimum usage and efficient deployment in the network environment. Shrink boundary & extend the area It would seem incongruous to attempt to take a circle and shrink the circumference but increase the area covered by the circle. Here we now move between the physical world and the metaphysical, or possibly more understandably, the modelled world. 38 industrial ethernet book 11.2017 SOURCE: MOXA Using a ring topology can automatically correct standard errors and faults in the transport layers.


Industrial Ethernet Book 103
To see the actual publication please follow the link above