Research Article
SOSE: Smart Offloading Scheme Using Computing Resources of Nearby Wireless Devices for Edge Computing Services
@INPROCEEDINGS{10.1007/978-3-030-23943-5_5, author={Ali Al-ameri and Ihsan Lami}, title={SOSE: Smart Offloading Scheme Using Computing Resources of Nearby Wireless Devices for Edge Computing Services}, proceedings={Emerging Technologies in Computing. Second International Conference, iCETiC 2019, London, UK, August 19--20, 2019, Proceedings}, proceedings_a={ICETIC}, year={2019}, month={7}, keywords={Offloading Edge computing Cooperative Mobile cloud computing}, doi={10.1007/978-3-030-23943-5_5} }- Ali Al-ameri
Ihsan Lami
Year: 2019
SOSE: Smart Offloading Scheme Using Computing Resources of Nearby Wireless Devices for Edge Computing Services
ICETIC
Springer
DOI: 10.1007/978-3-030-23943-5_5
Abstract
Offloading of all or part of any cloud service computation, when running processing-intensive Mobile Cloud Computing Services (MCCS), to servers in the cloud introduces time delay and communication overhead. Edge computing has emerged to resolve these issues, by shifting part of the service computation from the cloud to edge servers near the end-devices. An innovative Smart Cooperative Computation Offloading Framework (SCCOF), to leverage computation offloading to the cloud has been previously published by us [1]. This paper proposes SOSE; a solution to offload sub-tasks to nearby devices, on-the-go, that will form an “edge computing resource, we call SOSE_EDGE” so to enable the execution of the MCCS on any end-device. This is achieved by using short-range wireless connectivity to network between available cooperative end-devices. SOSE can partition the MCCS workload to execute among a pool of Offloadees (nearby end-devises; such as Smartphones, tablets, and PC’s), so to achieve minimum latency and improve performance while reducing battery power consumption of the Offloader (end-device that is running the MCCS). SOSE established the edge computing resource by: (1) profiling and partitioning the service workload to sub-tasks, based on a complexity relationship we developed. (2) Establishing peer2peer remote connection, with the available cooperative nearby Offloadees, based on SOSE assessment criteria. (3) Migrating the sub-tasks to the target edge devices in parallel and retrieve results. Scenarios and experiments to evaluate SOSE show that a significant improvement, in terms of processing time (>40%) and battery power consumption (>28%), has been achieved when compared with cloud offloading solutions.
