Cloud Computing Design Patterns for MLOps: Applications to Virtual Power Plants

Loading...
Thumbnail Image

Access rights

openAccess

URL

Journal Title

Journal ISSN

Volume Title

A4 Artikkeli konferenssijulkaisussa

Date

2023

Major/Subject

Mcode

Degree programme

Language

en

Pages

7

Series

IECON 2023 - 49th Annual Conference of the IEEE Industrial Electronics Society, IECON Proceedings (Industrial Electronics Conference)

Abstract

Virtual Power Plants (VPPs) are a key factor in smart grids, and they use cloud computing to integrate and manage Distributed Energy Resources (DERs). VPPs use Machine Learning (ML) methods to optimize various tasks. Machine Learning Operations (MLOps) methodology is a set of techniques that targets to develop, deploy and maintain ML applications smoothly on production. Cloud design patterns (CDPs) are general reusable solutions for common cloud problems that can improve the reliability, scalability, and quality of cloud applications. This paper discusses how CDPs can help in building complex ML applications on cloud with MLOps practices which can help VPPs to optimize their workloads. The paper also provides an example implementation on a public cloud provider.

Description

Funding Information: ACKNOWLEDGMENT This research was funded by Business Finland grant number 1516/31/2022. Publisher Copyright: © 2023 IEEE.

Keywords

cloud computing, design patterns, machine learning, MLOps, smart grid, virtual power plant

Other note

Citation

Subramanya, R, Räisänen, P, Sierla, S & Vyatkin, V 2023, Cloud Computing Design Patterns for MLOps : Applications to Virtual Power Plants . in IECON 2023 - 49th Annual Conference of the IEEE Industrial Electronics Society . IECON Proceedings (Industrial Electronics Conference), IEEE, Annual Conference of the IEEE Industrial Electronics Society, Singapore, Singapore, 16/10/2023 . https://doi.org/10.1109/IECON51785.2023.10312212