Update: New experimental data has been collected as part of recently completed work Thesis work. A comprehensive, reproducible dataset and framework repository will be made public soon.
Ruslan Kain, Sara A. Elsayed, Yuanzhu Chen, Hossam Hassanein
Data may also be accessed on Borealis
Citations:
- Kain, Ruslan; Elsayed, Sara A.; Chen, Yuanzhu; Hassanein, Hossam S., 2022, "Resource Usage of Applications Running on Raspberry Pi Devices", https://doi.org/10.5683/SP3/GOZAJE, Borealis, V1, UNF:6:FjVtgSYUu2Iy08LQ2ra6fQ== [fileUNF]
- R. Kain, S. A. Elsayed, Y. Chen and H. S. Hassanein, "DRUDGE: Dynamic Resource Usage Data Generation for Extreme Edge Devices," GLOBECOM 2023 - 2023 IEEE Global Communications Conference, Kuala Lumpur, Malaysia, 2023, pp. 5342-5347, doi: 10.1109/GLOBECOM54140.2023.10437760.
The collection and construction of this dataset were organized by the Queen's Telecommunications Research Lab (TRL) and led by Ruslan Kain, a Ph.D. student at Queen's School of Computing. The dataset includes dynamic resource usage information associated with running edge-native applications on a set of four heterogeneous Single Board Computers.
Four Raspberry Pi 4 devices have 2, 2, 4, and 8 GB RAM sizes, and CPU frequencies of 1200, 1500, 1500, and 1800 MHz. This is to establish heterogeneity of the devices used and collected data and to enable data-based applications for Edge Computing Research. The resource usage measurements have a five-second granularity. We managed to collect more than 550 thousand unique data points representing the 768 hours of running applications on Raspberry Pi Devices.
| Worker specifications and labels | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
| Dataset | |||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
| Augmented Reality | Dunio Coin Mining | Youtube Streaming | Gaming |
|---|---|---|---|
 |
 |
 |
 |