The rapid growth of new computing paradigms such as Cloud Computing and Big Data has unleashed great opportunities for companies to shift their business model towards a fully digital strategy. A major obstacle in this matter is the requirement of highly specialized ICT infrastructures that are expensive and difficult to manage. It is at this point that the IaaS (infrastructure as a service) model offers an efficient and cost‐affordable solution to supply companies with their required computing resources. In the Big Data context, it is often a hard task to design an optimal IaaS solution that meets user requirements. In this context, we propose a methodology to optimize the definition of IaaS cloud models for hosting Big Data platforms, following a threefold criterion: cost, reliability, and computing capacity. Specifically, the proposed methodology hinges on evolutionary heuristics in order to find IaaS configurations in the cloud that optimally balance such objectives. We also define measures to quantify the aforementioned metrics over a Big Data platform hosted within an IaaS cloud model. The proposed method is validated by using real information from three IaaS providers and three Big Data platforms. The obtained results provide an insightful input for system managers when initially designing cloud infrastructures for Big Data applications.

Leaf wetness often emerges as the result of the exchange of atmospheric water-soluble gases between the Earth surface and the atmosphere. The importance of this feature resides in the relationship that exists between leaf wetness and various plant diseases. In order to measure this variable, there is a need for deploying physical sensors to capture wetness readings of a crop area. However, the installation and maintenance of these sensors is a hard task that involves qualified people, time and high economical costs. Moreover, the acquisition, storage and analysis of data must be taken into consideration to infer this information and issue countermeasures preemptively. This work presents a leaf wetness soft-sensing approach that relies on predictive machine learning models to estimate the wetness of the leaves of a specific crop. Specifically, among the learning algorithms that are evaluated for this purpose, we include Random Vector Functional Link (RVFL) networks, a family of neural networks that embrace randomization at their core to yield a highly efficient training process. By virtue of machine learning, physical sensors can be replaced by soft-sensors capable of providing the information related to the wetness of the leaves of the crop. In this way, human effort and costs are largely reduced, while ensuring a high precision of the wetness estimation as proven by experiments with real-world data.

No abstract

No abstract