An Accurate Model for Computation Offloading in 6G Networks and a HAPS-Based Case Study
Open Access
- 28 October 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 3, 1963-1977
Abstract
The undeniable potential of computation offloading has been attracting attention from researchers for more than a decade. With advances in multi-access edge computing (MEC), computation offloading has become a more critical issue because of the heterogeneity in the computational power of edge devices and the elevated importance of extending their lifespan. Due to the apparent advantages, the use of MEC in 6G networks, where a vertical heterogeneous network composed of space, air, and ground networks is only natural. The non-terrestrial networking elements constitute effective computational resources. However, recent research investigating the potential of computational offloading in 6G networks has involved models that do not adequately reflect the complexity of the underlying processes. In this study, we propose a realistic computation model for 6G networks that considers crucial properties of the offloaded job, including the inter-dependency of the job tasks and the decomposability of the job. Our model is based on the mature application domain of MEC, where proven solutions are already studied. We also investigate the potential of a high altitude platform station (HAPS)-aided MEC platform using this model. The proposed model allows us to design offloading strategies to enable adaptive computational offloading. Through numerical analyses, we show that the proposed model provides sufficient insight to reduce the total processing time significantly.Keywords
This publication has 55 references indexed in Scilit:
- Efficient Multi-User Computation Offloading for Mobile-Edge Cloud ComputingIEEE/ACM Transactions on Networking, 2015
- Evaluation of connected-component labeling algorithms for distributed-memory systemsParallel Computing, 2015
- Characterizing and modeling cloud applications/jobs on a Google data centerThe Journal of Supercomputing, 2014
- Analysis, Modeling and Simulation of Workload Patterns in a Large-Scale Utility CloudIEEE Transactions on Cloud Computing, 2014
- Beyond Poisson: Modeling Inter-Arrival Time of Requests in a Datacenter2014
- Workload Analysis, Implications, and Optimization on a Production Hadoop Cluster: A Case Study on TaobaoIEEE Transactions on Services Computing, 2013
- Inside dropbox2012
- Interactive analytical processing in big data systemsProceedings of the VLDB Endowment, 2012
- Random graph generation for scheduling simulations2010
- Adaptive Offloading for Pervasive ComputingIEEE Pervasive Computing, 2004