Optimizing Resource Reuse in the Web of Things

The Web of Things enables collecting vast amounts of data about the environment and sharing them over the Internet. Its popularity, however, has brought challenges under constrained environments, such as an increase in the number of connected devices, and a consequent additional consumption of resources, leading to the deterioration of communications and application response times. By discovering and reusing semantically equivalent Things among applications, resources may be spared and responsiveness improved. In this article, an optimization model is developed to address this issue, and both fair and unfair goals are used for performance comparison over multiple scenarios, with different levels of network connectivity, Thing equivalence and placement. Results show that the interplay between network connectivity and equivalence ratio is determinant for responsiveness improvement, and both parameters can be used for an adequate planning of which Thing hosts to reuse. Also, ensuring a fair responsiveness among applications imposes fewer reuses, particularly in low connectivity and equivalence conditions, than using unfair criteria, where some applications can benefit from high responsiveness while others don’t. When a high equivalence ratio is coupled with a low connectivity degree, reuses tend to compound into fewer hosts, independently of fairness.