Comparing IoT system deployment on Raspberry Pi 4 Model B and Odroid-C2 devices
Ecker, Jonathan (2022)
Ecker, Jonathan
2022
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-202205118686
https://urn.fi/URN:NBN:fi:amk-202205118686
Tiivistelmä
A common problem for microcontroller devices and most technology, in general, is measuring and comparing devices or components to determine which product best suits a situation and the benefits or drawbacks each possesses. The goal is to investigate the ability for alternative microcontrollers to outperform the Raspberry Pi, which is the most common microcontroller device on the market. The common deployment of the Raspberry Pi device results in a large portion of research and education being carried out on one device, with the competing products often overlooked as a result. By attempting to compare the devices, it can be determined whether there are better alternatives for both education and personal use for monitoring and experimentation. Although the scope of this project is limited to two microcontroller devices the idea is that the system should be designed in such a way that it can be implemented on any microcontroller device and have the experiments conducted in the exact same way.
In this project, the Raspberry Pi and Odroid devices-are paired to a Bluetooth weather station which then takes selective measurement values consisting of humidity, pressure, and temperature and performs a series of mathematical operations on the data using principles of edge computing to determine how well each device performs. The use of Linux distributions which support Python is key to the universal nature of the experimentation being successful. If a microcontroller device supports a Linux operating system and can run Python scripts, then the experiments can be repeated exactly as listed and compared to the findings in this study. Regarding the findings it was initially expected for the Raspberry Pi device to be the clear favorite due to the superior hardware components available to the device. Contrary to the initial expectation, while the hardware advantage allowed the Raspberry Pi to perform better on pure mathematical operations it was not quite as dominant in other areas and even lost in performance to the Odroid device at times. For example, one of the aspects that the Odroid device performed better at was accessing and returning information directly from data storage.
The key takeaway of the project and experimentation is that there are many more consumer microcontroller devices available than just the Raspberry Pi and as demonstrated through the experiments, some will perform better in their niches. While the experimentation is not fully comprehensive, it looks at the most common solutions and deployments and crafts a rough estimate of a device’s abilities. While the Raspberry Pi is still a very good option for most microcontroller experimentations, hopefully, the results and findings can compel a look into alternative microcontrollers as well.
In this project, the Raspberry Pi and Odroid devices-are paired to a Bluetooth weather station which then takes selective measurement values consisting of humidity, pressure, and temperature and performs a series of mathematical operations on the data using principles of edge computing to determine how well each device performs. The use of Linux distributions which support Python is key to the universal nature of the experimentation being successful. If a microcontroller device supports a Linux operating system and can run Python scripts, then the experiments can be repeated exactly as listed and compared to the findings in this study. Regarding the findings it was initially expected for the Raspberry Pi device to be the clear favorite due to the superior hardware components available to the device. Contrary to the initial expectation, while the hardware advantage allowed the Raspberry Pi to perform better on pure mathematical operations it was not quite as dominant in other areas and even lost in performance to the Odroid device at times. For example, one of the aspects that the Odroid device performed better at was accessing and returning information directly from data storage.
The key takeaway of the project and experimentation is that there are many more consumer microcontroller devices available than just the Raspberry Pi and as demonstrated through the experiments, some will perform better in their niches. While the experimentation is not fully comprehensive, it looks at the most common solutions and deployments and crafts a rough estimate of a device’s abilities. While the Raspberry Pi is still a very good option for most microcontroller experimentations, hopefully, the results and findings can compel a look into alternative microcontrollers as well.