Dockemu is developed because of the need for a well designed tool for emulating networks. The tool utilizes technologies that are tailored to a current researcher's needs, delivering a robust and dynamic framework. In the past, most emulation tools have tried to provide solutions without taking into consideration that the installation and configuration of the software can be very time consuming, not to mention, the complexity of setting up and running an actual experiment can become also a very complex task. Our approach provides the researcher the flexibility to rapidly create networks (wired or wireless), up to date operating systems and a user friendly method of installation and configuration, that translates into a streamlined workflow for the emulation of experiments. Also, the development of applications or prototypes can be made directly into a real world OS. This saves time, because the prototypes will only be developed once (not for a simulator first and then for a final OS) but the results will also be more accurate. Dockemu utilizes virtualization with Linux Containers through Docker and Linux Bridging along with NS-3 for the emulation of layers 1 and 2 of the OSI model.

No abstract

One of the requirements for existing personal computers, PCs, is that the hardware inside must maintain an operating temperature as low as possible. One way to achieve that is to place the hardware components at locations with enough airflow around it. However, the relationship between the airflow and temperature of the components is unknown before they are placed at specific locations inside a PC. In this work a Computational Fluid Dynamics (CFD) analysis is coupled to a Design of Experiment (DOE) methodology to answer typical minitower key questions: a) how do the possible positions of hardware components affect their temperature?, and b) is it possible to get an optimal placement for these hardware components using the data collected by the CFD simulation results? The DOE methodology is used to optimize the analysis for a very large number of possible configurations. The results help in identifying where the efforts need to be placed in order to optimize the positioning of the hardware components for similar configurations at the designing stage. Somehow the results show that general conclusions could be drawn, but that there are not specific rules that could be applied to every configuration.