Article Preview
TopIntroduction
The goal of this study is to develop a practical edge computing architecture and design, which can be used to support smart construction environments with high QoS. Approach taken by this study is a practical one. From one side, it investigates the potential for smart IoT applications in the construction domain and their detailed technical requirements. The initial review of existing smart applications shows a gap that can be addressed in the near future by relying on more dependable edge computing infrastructures and applications. This section explains the methodology and the detailed design of smart construction applications, which is based on advanced cloud, edge and fog computing approaches and software engineering technologies.
The identified challenge is to define a set of Edge Management. Services as that can be used to manage Smart IoT Construction Environments. Within the smart environments various smart applications are deployed to support automation. The primary goal of the Edge Management Services is to contribute to higher QoS in comparison to existing cloud systems and therefore also to be able to meet stringent dependability requirements of every potential smart application (Shangguang et al., n.d.). The present work takes a top-down approach. Based on edge and fog computing definitions exiting technology reviews, the identified technical requirements of the potential applications, and experience gained in the course of advanced cloud computing and software engineering projects, such as OSAIC, SWITCH and ENTICE, an adequate conceptual approach for edge computing in construction is prepared. It defines the main necessary services that can be used to obtain high QoS. Following the perspective, the dependability can be also defined as system's ability to avoid failures that are more frequent or severe, and outage durations that are longer, than it is acceptable to the user, for example, to perform safety and business critical operations. To measure the system dependability, it is necessary to consider various dependability attributes, such as: high QoS, availability, reliability, safety, security, maintainability, integrity and many others (Roman et al., n.d.).
Addressing all of them within a single architecture is out of scope of this work. A dependable system can assume a subset of these attributes, which particularly relate to its safety and/or business critical operation, for example, the achieved QoS. Moreover, some of the dependability attributes are related, for instance, higher availability contributes to higher reliability and requires higher security, but that significantly increases the operational cost. Thus, building a dependable system in practice is a trade-off between attributes and their importance varies according to the business needs. When the key dependability requirements are not satisfied, severe consequences are inevitable. For instance, this can cause additional financial costs, compromised safety at the construction site, construction process delays, loss of reputation and many other repercussions (Choo et al., n.d.).
For various reasons, the cloud computing paradigm is unable to meet certain requirements (e.g. low latency and jitter, context awareness, mobility support) that are crucial for several applications (e.g. vehicular networks, augmented reality). To fulfill these requirements, various paradigms, such as fog computing, mobile edge computing, and mobile cloud computing, have emerged in recent years. These large-scale, commodity-computer data centers have enough computing resources to serve a very large number of users (Guo et al., 2017). However, this centralization of resources implies a large average separation between end user devices and their clouds, which in turn increases the average network latency and jitter. Because of this physical distance, cloud services are not able to directly access local contextual information, such as precise user location, local network conditions, or even information about users’ mobility behavior.