Internet of Things (IoT) addresses the challenge to provide a transparent access to a huge number of IoT resources that can be either physical devices or just data resources. Moreover, because of the large number of resource-constrained devices and the dynamic nature of IoT environments, integrating the resulted data becomes a non trivial task. We believe that the use of mashups, a way to compose new services from existing ones, can be a solution to the above challenge if each resource exposes its functionalities as a Web service. In the IoT environment, this will constitute a Web of Things where mashups development will take advantage of the connected physical world. The huge amounts of IoT-generated data from physical devices and data sources, called IoT Big Data, requires new design solutions to speed up data processing, scale up with the data volume and improve data adaptability. Besides existing techniques for IoT data collection, filtering, and analytics, we present in this article a mashup oriented model, called IoT Big Services, for provisioning data-centric IoT services in the context of IOT mashups. These IoT services are organized in tree structure where each node, called an IoT Big Service, acts as an integrator that collects data from lower level, processes them and delivers the results to higher level in the architecture.