Abstract
This article presents and discuss the industrialization process of ML (Machine Learning) projects with a focus on the principles of MLOps (Machine Learning Operations) and the challenges encountered when putting an ML project into production. The paper also proposes a set of tools used in an MLOps context to facilitate the deployment of ML projects and their production release. This paper is a guide to discover the MLOps domain in its theoretical (MLOps concepts, pipeline and life cycle) and practical (technical and tools) aspects. MLOps must provide answers to the use of ML applications hosted on servers with high performance, also for applications embedded in equipment with minimal sizing, since we are talking about the fourth industrial revolution with the increase of number of sensors in the world generating a mass of data that must be efficiently processed and analyzed.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Sculley, D., et al.: Hidden technical debt in machine learning systems. In: NIPS, pp. 2494–2502 (2015)
Symeonidis, G., Nerantzis, E., Kazakis, A., Papakostas, G.: MLOps – definitions, tools and challenges, pp. 0453–0460. https://doi.org/10.1109/CCWC54503.2022.9720902 (2022)
Ruf, P., Madan, M., Reich, C., Abdeslam, O.D.: Demystifying MLOps and presenting a recipe for the selection of open-source tools. Appl. Sci. 11, 39 (2021). https://doi.org/10.3390/app11198861
Cajamarca, A.: Andres Felipe Cajamarca Mantilla Responsabilidad Social. https://doi.org/10.13140/RG.2.2.35401.06241(2021)
Fursin, G., Guillou, H., Essayan, N.: CodeReef: an open platform for portable MLOps, reusable automation actions and reproducible benchmarking (2020)
Liu, Y., Ling, Z., Huo, B., Wang, B., Chen, T., Mouine, E.: Building a platform for machine learning operations from open source frameworks. IFAC-PapersOnLine 53, 704–709 (2020). https://doi.org/10.1016/j.ifacol.2021.04.161
Mäkinen, S., Skogström, H., Laaksonen, E., Mikkonen, T.: Who needs MLOps: what data scientists seek to accomplish and how can MLOps help? (2021)
Chen, A., et al.: Developments in MLflow: a system to accelerate the machine learning lifecycle, pp. 1–4. https://doi.org/10.1145/3399579.3399867(2020)
Bisong, E.: Building machine learning and deep learning models on Google Cloud Platform: a comprehensive guide for beginners. https://doi.org/10.1007/978-1-4842-4470-8(2019)
Site officiel dvc [online]. Available on: https://dvc.org/
Ray, P.P.: A review on TinyML: state-of-the-art and prospects. J. Sens. 2022, 11 (2022)
ZenML [online]. Available on: https://zenml.io/
Neptune [online]. Available on: https://neptune.ai/blog/best-mlops-tools
Censius [online]. https://censius.ai/blogs/dvc-vs-mlflow
Databricks [online]. https://databricks.com/product/managed-mlflow
Chakraborty, A., Mandal, N.: Introduction of industry 4.0 introduction of industry 4.0. Turk. Online J. Qual. Inq. 12, 8342–8350 (2022)
Dharmadhikari, S.: Industry 4.0 Challenges. 40 (2022)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Bodor, A., Hnida, M., Najima, D. (2023). MLOps: Overview of Current State and Future Directions. In: Ben Ahmed, M., Boudhir, A.A., Santos, D., Dionisio, R., Benaya, N. (eds) Innovations in Smart Cities Applications Volume 6. SCA 2022. Lecture Notes in Networks and Systems, vol 629. Springer, Cham. https://doi.org/10.1007/978-3-031-26852-6_14
Download citation
DOI: https://doi.org/10.1007/978-3-031-26852-6_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-26851-9
Online ISBN: 978-3-031-26852-6
eBook Packages: EngineeringEngineering (R0)