Explore projects

Assistance Systems Project is a web application that provides personalized health recommendations and data analysis. The frontend is built using Streamlit, and it integrates a chatbot developed with Rasa. Machine learning models created with scikit-learn are used to generate recommendations based on user input. The application is containerized with

Objectives:

The primary objective of this thesis is to evaluate and compare the performance of WebAssembly and Docker as containerization solutions for edge computing and IoT ap- plications. The study aims to provide insights into the strengths and limitations of each technology in terms of resource utilization, execution speed, and overall efficiency.

Methodology:

To achieve this objective, a comprehensive methodology was adopted. Hard- ware representative of edge devices, such as the Raspberry Pi 3 B+ and HP Pavillion x360 running linux 20.0 Mint Edition, was selected for benchmarking. A diverse set of CPU- intensive algorithms, including compression, encryption, and graph-based operations, was implemented in C and executed natively, within WebAssembly, and in Docker containers. Benchmark results, including CPU performance, memory usage, and file system I/O, were collected and analyzed. Statistical methods, including t-tests, were employed to quantify performance differences.

Results: The findings reveal nuanced performance variations between WebAssembly and Docker in different aspects. Native execution consistently outperformed both WebAssem- bly and Docker, emphasizing the impact of containerization overhead. Docker exhibited efficient execution on certain algorithms, particularly those well-suited for its runtime envi- ronment. WebAssembly, while versatile, incurred overhead in terms of resource consump- tion.

Keywords: WebAssembly, Docker, Containerization, Virtualization, IoT, WebAssembly Sys- tem Interface, Cloud computing

Modern Internet Technologies WS21-22 Course Project

Computer Vision Project - Forest fire detection

Einstieg in Auto ML Präsentation Seminar Aktuelle Themen der KI -- KI-B-22