Dataflow Optimization in Edge-Cloud Continuum: AI-Enabled Computational Models for Efficient Data Processing
DOI:
https://doi.org/10.63282/3117-5481/AIJCST-V2I5P102Keywords:
Edge Computing, Cloud Computing, Dataflow Optimization, Artificial Intelligence, Machine Learning, Deep Learning, IoT, Resource ManagementAbstract
The widespread emergence of Internet of Things (IoT) devices, along with the rapid expansion of data creation, have put a tremendous burden on the traditional data processing in the cloud-based systems. Integrating the edge computing and cloud infrastructure has led to the emergence of the edge-cloud continuum framework, which is proposed as a viable solution to overcome the latency issue, bandwidth issues, and scalability problems. The present paper provides an extensive work on AI-based dataflow optimization computer models in the edge-cloud continuum. The suggested solution takes the advantage of using machine learning (ML) and deep learning (DL) to distributed intelligent allocation of work and ensure calculation demands, and distribute resources more intelligent. The substantial simulations and case studies have shown that AI-based dataflow optimization can cut down the latency by up to 35 percent, enhance the throughput by 28 percent and decrease the energy usage by 22 percent in comparison with traditional approaches. Our results create a powerful emphasis on the importance of AI to boost the performance, efficiency, and sustainability of distributed computing models as a groundbreaking source of reference to researchers and practitioners who are interested in designing scalable and adaptable edge-cloud solutions
References
[1] Yousefpour, A., et al. (2019). All one needs to know about fog computing and related technologies. Computers & Electrical Engineering, 72, 176–197.
[2] Bandi, A., & Hurtado, J. (2018). Edge Computing as an Architectural Solution: An Umbrella Review. Northwest Missouri State University
[3] Sammu, J. (2018). Predictive Maintenance Using Edge and Cloud Collaboration. ResearchGate
[4] Bittencourt, L. F., Immich, R., Sakellariou, R., da Fonseca, N. L. S., Madeira, E. R. M., Curado, M., Villas, L., da Silva, L., Lee, C., Rana, O. (2018). The Internet of Things, Fog and Cloud Continuum: Integration and Challenges. arXiv preprint arXiv:1809.09972.
[5] Enabling Mission-Critical Communication via VoLTE for Public Safety Networks - Varinder Kumar Sharma - IJAIDR Volume 10, Issue 1, January-June 2019. DOI 10.71097/IJAIDR.v10.i1.1539
[6] Yao, C., Wang, X., Zheng, Z., Sun, G., Song, L. (2018). EdgeFlow: Open Source Multi layer Data Flow Processing in Edge Computing for 5G and Beyond. arXiv preprint arXiv:1801.02206.
[7] Akidau, T., Bradshaw, R., Chambers, C., Chernyak, S., Fernández Moctezuma, R., Lax, R., McVeety, S., Mills, D., Perry, F., Schmidt, E., Whittle, S. (2015). The Dataflow Model: A Practical Approach to Balancing Correctness, Latency, and Cost in Massive Scale, Unbounded, Out of Order Data Processing. Proceedings of the VLDB Endowment, 8(12), 1792 1803.
[8] Zhang, P., Han, Y., & Babar, M. (2015). A dataflow optimisation mechanism for service oriented cloud workflow. International Journal of Computational Science and Engineering, 10(1), 15 31.
[9] Sharma, S. D., Acharya, A., Saltz, J. (1998). Deferred Data Flow Analysis: Algorithms, Proofs and Applications. UMIACS TR 98 46.
[10] Kung, S. Y., Lewis, P. S., Lo, S. C. (1987). Performance Analysis and Optimization of VLSI Dataflow Arrays. Journal of Parallel and Distributed Computing, 4(6), 592 618.
