MLOPS BASED PROTOTYPE OF AI SYSTEM FOR EDGE COMPUTING

Oleh Sinkevych, Yaroslav Boyko, Lyubomyr Monastyrskyy

Abstract


The paper represents an approach for developing the edge AI system which is based on the modern MLOps concept. For the edge hardware part we utilize Nvidia Jetson Nano microcomputer which provides server side for network requests processing, data storage and machine learning models of self-deployment. We propose the working MLOps pipeline fully designed by the industrial software solutions like TensorFlow 2, Mlflow, Apache Airflow, which is integrated into the developed application. The considered pipeline scheme consists of three operational stages: a) data storage and processing, which stands for fetching the data from database, cleansing and transformation; b) machine learning modeling with synchronous hyper-parameters optimization and model registration; c) model deployment and serving. The whole pipeline is wrapped by the REST API created via FastAPI micro-framework and orchestrated using Apache Airflow service. To implement the described pipeline we chose the time dependant temperature data to be learned and short-term predicted by the GRU-based recurrent neural network. The latter one is tuned in terms of hyper-parameters configuration by genetic algorithm which is embedded into the second stage of the pipeline. Also, a design which combines Nvidia Jetson Nano server with the inference edge device like STM32 H745 microcontroller via sockets is discussed.

Key words: edge computing, MLOps, machine learning, Mlflow, genetic algorithm.


Full Text:

PDF

References


[1] Edge computing use case examples [Електронний ресурс]. – 2020. – Mode of access: https://stlpartners.com/edge_computing/10-edge-computing-use-case-examples/.

[2] Sriram G. Edge Computing vs. Cloud Computing: An overview of Big Data challenges and opportunities for large enterprises / G. Sriram. // International Research Journal of Modernization in Engineering Technology and Science. – 2022. – №4. – P. 1186 – 1190.

[3] Bringing computation closer towards user network: Is edge computing the solution? / [A. Ejaz, A. Ahmed, I. Yaqoob та ін.]. // IEEE Communications Magazine. – 2017. – №55. – С. 138–144. DOI: 10.1109/MCOM.2017.1700120.

[4] Verma A. Comparative Study of Cloud Computing and Edge Computing: Three Level Architecture Models and Security Challenges / A. Verma, V. Verma. // International Journal of Distributed and Cloud Computing. – 2021. – №9. – С. 13–17.

[5] Recent Advances in Evolving Computing Paradigms: Cloud, Edge, and Fog Technologies / [N. Angel, D. Ravindran, P. Vincent та ін.]. // Sensors. – 2022. – №22. – С. 196–234. DOI: 10.3390/s22010196.

[6] Edge MLOps: An Automation Framework for AIoT Applications / E.Raj, D. Buffoni, M. Westerlund, K. Ahola. // IEEE International Conference on Cloud Engineering (IC2E). – 2021. – С. 191–200. DOI: 10.1109/IC2E52221.2021.00034.

[7] A Digital Twin Architecture Model Applied with MLOps Techniques to Improve Short-Term Energy Consumption Prediction / [T. Yukio, V. Hayashi, R. Arakaki та ін.]. // Machines. – 2022. – №10. – С. 23–49. DOI: 10.3390/machines10010023.

[8] Fowler M. Python Concurrency with asyncio / Matthew Fowler. – Shelter Island : Manning Publications, 2022. – 378 p.

[9] FastAPI [Електронний ресурс] // FastAPI. – Режим доступу: https://fastapi.tiangolo.com(дата звернення: 22.05.2022). – Title from screen.

[10] Benchmarks - FastAPI [Електронний ресурс] // FastAPI. – Режим доступу: https://fastapi.tiangolo.com/benchmarks/ (дата звернення: 22.05.2022).

[11] DEAP documentation – DEAP 1.3.1 documentation [Електронний ресурс] // DEAP documentation – DEAP 1.3.1 documentation. – Режим доступу: https://deap.readthedocs.io/en/master/ (дата звернення: 22.05.2022).

[12] Harenslak B. Data Pipelines with Apache Airflow / Bas Harenslak, Julian de Ruiter. – [Б. м.]: Manning Publications, 2021. – 480 с.




DOI: http://dx.doi.org/10.30970/eli.17.7

Refbacks

  • There are currently no refbacks.