Offloading computing tasks beyond the edge: A data-driven analysis

Sadia Khizar (1, 2) , Marcelo Dias de Amorim (2) , Vania Conan

1 Thales SIX GTS France
2 NPA - Networks and Performance Analysis

Mobile devices have always been left out of the network infrastructure due to their limited capacities. However, it turns out that they are becoming more and more sophisticated and, above all, increasingly numerous. Given their ubiquity, they offer untapped resources to extend the computing capacity of the MEC (multi-access edge computing). Nevertheless, such resources vary over time due to the dynamics of the network. In this paper, we investigate how the mobility of nodes impacts the task offloading process. To this end, we use traces of actual user equipment (UE) mobility from a cellular operator. We also quantify the impact of task duration and completion delay. Our results show that mobile nodes' offloading potential beyond the edge is significant, even for short delays. Index Terms-Beyond-the-edge computing, mobility, data analysis, task offloading.

Domaines: Informatique [cs]; Sciences de l'ingénieur [physics]

Sadia Khizar, Marcelo Dias de Amorim, Vania Conan. Offloading computing tasks beyond the edge: A data-driven analysis. 13th IFIP Wireless and Mobile Networking Conference, Oct 2021, Montreal (online), Canada. ⟨hal-03375422⟩