Dynamic Hardware Customisation for Mobile Users in FPGA-accelerated Edge Infrastructures

• Diogo Gonçalves UNICAMP
• Luiz Bittencourt UNICAMP
• Edmundo Madeira UNICAMP

DOI: https://doi.org/10.5753/eradsp.2024.239937

Resumo

User mobility support is relevant for smart applications in edge network infrastructures. Machine Learning-based solutions have been taking advantage of hardware customisation to improve their performance. In this context, attaching Field Programmable Gate Array (FPGA) technology into edge nodes could offer tailored hardware for ML applications at the edge. This work proposes hardware customisation at the edge network to improve the performance of ML-based solutions. In that scenario, mobile users could configure the hardware and apply it to different edge nodes. Mobile users can use customised hardware to execute applications at the edge in a performance-enhanced environment by carrying hardware settings.

Referências

Fahim, F., Hawks, B., Herwig, C., Hirschauer, J., Jindariani, S., Tran, N., Carloni, L. P., Di Guglielmo, G., Harris, P., Krupa, J., et al. (2021). hls4ml: An open-source codesign workflow to empower scientific low-power machine learning devices. arXiv preprint arXiv:2103.05579.

Gonçalves, D., Puliafito, C., Mingozzi, E., Rana, O., Bittencourt, L., and Madeira, E. (2020). Dynamic network slicing in fog computing for mobile users in mobfogsim. In 2020 IEEE/ACM 13th International Conference on Utility and Cloud Computing (UCC), pages 237–246. IEEE.

Leiserson, C. E., Thompson, N. C., Emer, J. S., Kuszmaul, B. C., Lampson, B. W., Sanchez, D., and Schardl, T. B. (2020). There’s plenty of room at the top: What will drive computer performance after moore’s law? Science, 368(6495):eaam9744.

Puliafito, C., Goncalves, D. M., Lopes, M. M., Martins, L. L., Madeira, E., Mingozzi, E., Rana, O., and Bittencourt, L. (2020). Mobfogsim: Simulation of mobility and migration for fog computing. Simulation Modelling Practice and Theory, 101:102062.

Ramchandani, D., Asgari, B., and Kim, H. (2023). Spica: Exploring fpga optimizations to enable an efficient spmv implementation for computations at edge. In 2023 IEEE International Conference on Edge Computing and Communications (EDGE), pages 36–42. IEEE.

Taleb, T., Ksentini, A., and Frangoudis, P. A. (2016). Follow-me cloud: When cloud services follow mobile users. IEEE Transactions on Cloud Computing, 7(2):369–382.

Yuan, T., Da Rocha Neto, W., Rothenberg, C. E., Obraczka, K., Barakat, C., and Turletti, T. (2022). Machine learning for next-generation intelligent transportation systems: A survey. Transactions on emerging telecommunications technologies, 33(4):e4427.