The rising popularity of the Internet of Things (IoT) has led to a plethora of highly heterogeneous, geographically dispersed devices. In recent years, IoT platforms have been used to provide a variety of services to applications such as device discovery, context management, and data analysis. However, the lack of standardization currently means that each IoT platform comes with its own abstractions, APIs, and interactions. As a consequence, programming the interactions between an application and an IoT platform is often time consuming, error prone, and depends on the developers' level of knowledge about the IoT platform. To address these issues, we propose offering to application developers on the client side the possibility to declare variables that are automatically mapped to sensors and whose values are transparently updated with sensor observations. For this purpose, we introduce IoTVar, a middleware between IoT applications and platforms. In IoTVar, all the necessary interactions with IoT platforms are managed by proxies. This paper presents IoTVar integrated with the FIWARE platform, which is used for developing IoT Future Internet applications. We also report results of some experiments performed to evaluate IoTVar, showing IoTVar reduces the effort required to declare and manage IoT variables and its impact in terms of CPU, memory, and energy.

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Nowadays, it is estimated that half the connected devices are related to the Internet of Things (IoT). The IoT paradigm contributes to the increase of the Information Technology energy demand. The energy demand is due on one side to the huge number of IoT devices, and on the other side to the plethora of IoT end user applications consuming data produced by those devices. However, taking into account energy consumption in the development of such applications, consuming data produced by IoT devices is still challenging. There is a lack of knowledge on what are the best practises to develop green IoT applications. The work presented in this paper aims to raise the awareness of application designers concerning the impact of the choice of IoT protocols and interaction patterns on the energy consumption of the applications. For this purpose, we have experimentally analysed the energy consumption of HTTP and MQTT, which are two of the most popular, mature and stable protocols for IoT consumer applications. For the HTTP protocol, we have studied both the publish-subscribe and the request-reply interaction patterns. For MQTT, we have studied the publish-subscribe interaction pattern with the three available Quality of Services. We also examine the impact of message payload on energy consumption. The results show that the publish/subscribe interaction pattern has lower energy consumption (around 92% less) than the synchronous interaction pattern and HTTP consumes 20% more energy than the MQTT protocol for the publish/subscribe interaction pattern. Finally, we show that the payload has a low impact on energy consumption, having a 9% overhead on payloads ranging from 24 to 3120 bytes