Data Awareness Circles: Data lives where it is generated
Starting this article to express the importance of any kind of data generated by us or another thing can be quite boring, and probably you would not have interest in reading the rest of the article. Even this case is known to me, the principal aim of writing this article is to show the importance of the data generation location and its valuable meaning for the new technologies.
Assuming that we focus together on a subject in order to find a solution. During that time, we isolate ourselves from the surrounding things and nearly put a virtual distance to every object or idea. While doing that, we actually build data awareness circles, which allow us to deep dive in a topic (indeed a data cluster or space) and do not see the details of all other things. This use case can be seen an extreme example, and if you think so, let see together the other two examples: 1) The world in which we live is full of the events, however, these events can be observed as long as we are close to them. If we increase our distance to the world, the observation quality decreases dramatically, and we will not be able to see the details. Another use case comes from our typical life: The governments or companies are mostly hierarchically constructed in order to simplify the management and coordination. The top manager, CEO, has in general no idea what is the subject of each individual developer or worker within the organization, since she has different responsibilities and receives constrained information from the bottom level. These use cases are depicted below.
Figure 1: Use Cases for Data Awareness Circles Concept
Even though all use cases are different from each other, there is a critical point where all aforementioned use cases coincide: The more we are far from a topic or a thing, the more we lose our observation capabilities on it.
The number of the examples is limitless, and this article can be extended with other use cases, however, the intention is here to shed light on a specific point, which is a common point for all examples. The data is generally more valuable at the location where it is generated, and its importance weakens when it is evaluated in a far point. This statement is probably true in all universe, and we can say “if you want to see the details, then you have to look more closely”. Probably we would agree with the following visualization, since it depicts the same thing.
Figure 2: Data Awareness Circles Layered and Top-Down Views
Relation between IoT and Data Awareness Circles
Devices or services produce and consume data since decades on many platforms. Especially, Internet of Things becomes known in the last decades and noticed that its applicability is valid nearly everywhere, and has a huge potential to reshape our entire life, since we can collect massive data from the environment, thus allowing us to better comprehend and interpret the processes, and optimizing them. In other words, for the first time, we obtained the opportunity to look at our surroundings in detail.
The massive data generation leads to redesign the legacy internet architecture, which is constructed on the cloud architecture, because the connection of millions of devices cannot be easily processed in the required amount of time. For example, The generated massive data by devices in smart factories or autonomous vehicles on street cannot be processed in real time. If the critical operations cannot be completed on time, the risk level in those operations can increase, which may end up with undesired situations. The problem is the locally generated data does not need indeed to be stored on the cloud, since as explained above, the locally generated data has mostly less effect on the global data. Even though this is the case, we still require the data to be processed, however sending it to the cloud is not only costly but also less efficient.
Many technologies have been offered over time, two of them, namely, fog- and edge computing, have great potential to address the aforementioned issues. Even though these two terms are in general mixed, edge can be seen as the computation power close to IoT devices, whereas fog computing is placed between edge and cloud computing. The solution is basically to find adequate computation locations being close to IoT devices to process the data. By doing that, many advantages are gained, for instance, the network pipeline towards the cloud will not be overloaded, the time-critical applications will function in real-time, the local data will be kept at local and the security surface for the device will diminish, and so on.
Considering all aforementioned aspects, it is clearly seen that the data processing and data itself are pushed or kept to/at the location close to the devices, which demonstrates indeed the usage of data awareness circles. Even we assume that we have an immense cloud computation, it does not make sense to transmit the data to the cloud, since the raw data does not enough important at the cloud level compared to the edge level. For this reason, it can be claimed that wide usage of IoT devices will actually contribute to the Data Awareness Circles, and the edge-fog-cloud-space-etc. computing can be evaluated the layers of this approach.
Figure 3: Relation between X-Computing & Data Awareness Circles
Use Cases for Data Awareness Circles within IoT World
The number of use cases for demonstrating Data Awareness Circles can be found nearly in all IoT use cases, i.e. smart factory, smart cities, smart logistics, smart buildings, smart homes, etc. Smart factories generate huge amount of data during the production, and the IoT-Edge-Fog-Cloud connection for processing the data is a more than a necessity. Likewise, smart logistics requires real-time information and if autonomous vehicles are part of it, the IoT-Edge-Fog-Cloud is again a fundamental requirement for aggregating, processing and interpreting the data. The computation intensive tasks and time-critical operations can also be operated at smart buildings and -homes, where IoT gateways can behave an edge point, whereas the upper layers within the network can be consumed for long term operations.
Apart from all these use cases, smart cities can be also fulfilled with various use cases where the computation is distributed across different hierarchical nodes within the network. For instance, the interaction between IoT devices and autonomous vehicles so called V2X can be established using the smart city technology. Citizens or tourists can also actively consume the smart city services, ranging from more data consuming services to time-intensive applications.
In all examples, we clearly see how Data Awareness Circles is the main part of IoT and X-Computing. We can already see the change in the cloud architecture that moves from the core to the edge networks, and this change will accelerate in the next years through the introduction of new technologies.