MECCA offloading cloud model over wireless interfaces for optimal power reduction and processing time
| Autor: | Sufian Yousef, Georgios Kapogiannis, Rakan Aldmour, Mohammad Yaghi |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Wireless network
Computer science business.industry Node (networking) Quality of service Real-time computing 020206 networking & telecommunications Cloud computing 02 engineering and technology Energy consumption File size 0202 electrical engineering electronic engineering information engineering Wireless 020201 artificial intelligence & image processing business Energy (signal processing) |
| Zdroj: | SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI |
| DOI: | 10.1109/uic-atc.2017.8397639 |
| Popis: | In this paper, the power consumption and the processing time of smartphones are estimated locally and then compared with the power consumption and processing time when the smartphone executes heavy tasks by offloading on WLAN, 3G, and 4G mobile systems. Different scenarios were tested for different file sizes and wireless network interfaces. The main parameter of the quality of service is the time needed to process the file on the cloud versus the time needed to execute the file locally on the smartphone, as tested by the MECCA (Mobile Energy Cloud Computing algorithm) model. The optimal saving in energy consumption of the smartphone has reached around 90% over the 4G system, while maintaining an approximately similar range of time consumption for similar file sizes. The most important issue is to save time while serving the file. However, it is important, especially for the small nodes, to decrease the power consumption during serving big files, which is normally very high. The cost of the power consumption on smartphone, processing time, and file size for the core cloud and local node, are calculated to extract an immediate input to the processing decision. The Wi-Fi results showed very short processing times comparatively but resulted in very high energy consumption. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|