Resource Management for Internet of Things by Flávia C. Delicato Paulo F. Pires & Thais Batista

Author:Flávia C. Delicato, Paulo F. Pires & Thais Batista
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

5.3.2 Discussion

Unlike approaches that consider the IoT devices as passive data sources and delegate all work to the cloud, the proposals presented in this section leverage the collaborative processing between IoT devices and cloud. Such collaboration takes advantage of the devices computing capacity , considering such nodes as active players in the resource provision process. As showed in [18], among the benefits of such approaches, by exploiting the resources available at millions of smart mobile devices, user’s requests can be met more quickly, and the cost of cloud management can be significantly reduced. Moreover, besides the benefits related to cost and performance, issues related to user privacy, one of the controversial limitations of cloud usage in the IoT context, can be resolved. Personal data can be processed within a device or the obfuscated version of such data can be sent to the cloud [18]. One key drawback of cloud only approaches that can be overcome when considering the Things layer as active part of the RM is achieving real-time data delivery. Clouds are mainly suitable for exchanging historical data due to the delay introduced in the transmission to/from the cloud, which can yield unacceptable response times in the critical process such as eHealth applications. In this context, the IoT cloud approach provides means to build flexible sharing models such as the one described in [18] where sharing occurs at the device level and not only on the data produced by them.

Table presents the comparison of the IoT cloud approaches based on the IoT requirements tackled by them. Like the works in the cloud-only Section, all the works presented in this section propose RM solutions that take into account the high scale of IoT data as well as the heterogeneity of devices and applications. The dynamicity of the IoT environment is addressed by [15–18, 20] through mechanisms that monitor the context of the environment and adapt the system state accordingly. Unlike one would expect in works that considers the IoT devices on the RM scheme, key requirements for managing IoT resources , such as mobility, real-time and data stream processing are addressed by only few works. Data stream processing is addressed in [15, 28, 32]; mobility in [17, 22, 25, 30, 31]; and real-time processing is addressed by two works: [17, 25]. The work in [31] can be considered addressing price modelling since the cooperation among the components of the system is stimulated by a pricing mechanism and users demand. Fault tolerance is another requirement neglected by most of the work, only [20] and [32] explicitly take into consideration some level of fault tolerance in their solutions. Opportunistic interactions are another requirement investigated by few works [15–17] (Table 5.2). Table 5.2Comparison of approaches for IoT cloud

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.