Simulation-based Queueing Models for Performance Analysis of IoT Applications - Archive ouverte HAL Access content directly
Conference Papers Year :

Simulation-based Queueing Models for Performance Analysis of IoT Applications

(1, 2) , (3) , (1) , (1)


To facilitate the development of Internet of Things (IoT) applications, numerous middleware protocols and APIs have been introduced. Such applications built atop reliable or unreliable protocols and they expose different characteristics. Additionally, with regard to the application context (e.g., emergency response operations), several Quality of Service (QoS) requirements must be satisfied. To study QoS in IoT applications, the provision of a generic performance analysis methodology is required. Queueing network models offer a simple modeling environment, which can be used to represent IoT interactions by combining multiple queueing model types for building queueing networks. The resulting networks can be used for performance analysis through analytical or simulation models. In this paper, we present several types of queueing models that represent different QoS settings of IoT interactions, such as intermittent mobile connectivity, message drop probabilities, message availability/validity and resource constrained devices. Using MobileJINQS, we simulate our models demonstrating the significant effect on response times and message success rates when varying QoS settings.
Fichier principal
Vignette du fichier
csndsp_bouloukakis_et_al_2018.pdf (234.35 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01797930 , version 1 (23-05-2018)


  • HAL Id : hal-01797930 , version 1


Georgios Bouloukakis, Ioannis Moscholios, Nikolaos Georgantas, Valérie Issarny. Simulation-based Queueing Models for Performance Analysis of IoT Applications. 11th International Symposium on Communication Systems, Networks, and Digital Signal Processing (CSNDSP), Jul 2018, Budapest, Hungary. ⟨hal-01797930⟩
126 View
1028 Download


Gmail Facebook Twitter LinkedIn More