Litoiu, Marin2019-03-052019-03-052018-12-052019-03-05http://hdl.handle.net/10315/35891The number of IoT devices has been growing exponentially as new products are being developed and legacy systems are becoming Internet enabled. As a consequence of this trend, the large amounts of traffic generated by IoT devices require new approaches to platform design and workload management. The primary challenge with managing IoT devices is that the traffic can be highly variable due to the type and the time of use. To be able to maintain quality of service standards on an IoT platform, these traffic patterns need to be modeled and understood so we can adapt the architecture dynamically. To address these challenges, we propose an adaptable architecture, a platform to emulate IoT devices, and a smart testing framework to detect bottlenecks that can predict the demand for computing resources. We show that in certain cases we can predict the demand for computing resources with a high degree of accuracy.enAuthor owns copyright, except where explicitly noted. Please contact the author directly with licensing requests.Information technologyInformation technology--Internet of thingsElectronic Thesis or Dissertation2019-03-05Quality of service (Computer networks)Smart Testing