Machine Learning Operations: A Survey on MLOps Tool Support

Abstract

Machine Learning (ML) has become a fast-growing, trending approach in solution development in practice. Deep Learning (DL) which is a subset of ML, learns using deep neural networks to simulate the human brain. It trains machines to learn techniques and processes individually using computer algorithms, which is also considered to be a role of Artificial Intelligence (AI). In this paper, we study current technical issues related to software development and delivery in organizations that work on ML projects. Therefore, the importance of the Machine Learning Operations (MLOps) concept, which can deliver appropriate solutions for such concerns, is discussed. We investigate commercially available MLOps tool support in software development. The comparison between MLOps tools analyzes the performance of each system and its use cases. Moreover, we examine the features and usability of MLOps tools to identify the most appropriate tool support for given scenarios. Finally, we recognize that there is a shortage in the availability of a fully functional MLOps platform on which processes can be automated by reducing human intervention.

Machine Learning Operations (MLOps): Overview, Definition, and Architecture

“…Operational challenges. In productive settings, it is challenging to operate ML manually due to different stacks of software and hardware components and their interplay as well as the selection of both ( [64], [65]. Therefore, robust automation is required [24], [33].…”

Continual Learning for Predictive Maintenance: Overview and Challenges

“…There are various practical fields where researchers and companies have applied CL. Publications in computer vision [13], Machine Learning Model Operationalization Management (MLOps) [85], robotics and machine vision, industry, and edge computing, have expanded knowledge from the theoretical to more practical situations. An overview of some of the latest and most relevant CL studies for different applications is presented in the following sections:…”

Analysis and Development of the MLOps Pipeline for ML Model Deployment

“…Среди крупных недостатков MLFlow выделяется проблема отсутствия ролей пользователей и функционала безопасности, что приводит к затруднениям совместной работы над моделями [21]. Кроме того, еще одни из проблем MLFlowсложности с развёртыванием моделей на различных платформах и отсутствие мониторинга производительности моделей [21], [22].…”

“…К недостаткам Kubeflow относится отсутствие версионирования данных и конвейера [22]. Кроме этого, Kubeflow сложно настроить, у него высокий порог входа, а для его использования необходимы глубокие знания в Kubernetes [26].…”