Serverless on Machine Learning: A Systematic Mapping Study

Abstract

Machine Learning Operations (MLOps) is an approach to managing the entire lifecycle of a machine learning model. It has evolved over the last years and has started attracting many people in research and businesses in the industry. It supports the development of machine learning (ML) pipelines typical in the phases of data collection, data pre-processing, building datasets, model training, hyperparameters refinement, testing, and deployment to production. This complex pipeline workflow is a tedious process of iterative experimentation. Moreover, cloud computing services provide advanced features for managing ML stages and deploying them efficiently to production. Specifically, serverless computing has been applied in different stages of the machine learning pipeline. However, to the best of our knowledge, it is missing to know the serverless suitability and benefits it can provide to the ML pipeline. In this paper, we provide a systematic mapping study of machine learning systems applied on serverless architecture that include 53 relevant studies. During this study, we focused on (1) exploring the evolution trend and the main venues; (2) determining the researchers' focus and interest in using serverless on machine learning;(3) discussing solutions that serverless computing provides to machine learning. Our results show that serverless usage is growing, and several venues are interested in the topic. In addition, we found that the most widely used serverless provider is AWS Lambda, where the primary application was used in the deployment of the ML model. Additionally, several challenges were explored, such as reducing cost, resource scalability, and reducing latency. We moreover discuss the potential challenges of adopting ML on serverless, such as respecting service level agreement, the cold start problem, security, and privacy. Finally, our contribution provides foundations for future research and applications implying machine learning in serverless computing.

3D Placement Optimization in UAV-Enabled Communications: A Systematic Mapping Study

“…The research adheres to the established guidelines for conducting an SMS, which follows a systematic approach to provide an overview of the research area [58], which has also been used in similar studies [59]. The tasks involved in this research were organized into three main stages: planning, conducting, and reporting.…”

On the challenges of optical disaggregated data center networking for ML/AI applications

“…Generally speaking, ML applications require of CPU, memory and storage resources for both the training and inference phases while also requiring GPU resources during the training phase in order to accelerate the learning process. Additionally, from an application structure perspective, ML applications can be split broadly into four different sub-applications [9]. First, a storage sub-application that stores the historical data and data inputs for the training of the ML as well as the real-time data for the inference of the model, which require large storage spaces.…”

Managing Cold-start in the Serverless Cloud

“…It outlines the architecture or process that combines these models, providing a structured overview for understanding their collaborative role in enhancing cloud infrastructure. Table 6.Integration framework of tcns and ensemble policies in cloud management [14] Communication Step…”

“…The proposed Ensemble Policy Architecture is designed to provide a robust and adaptive framework for start challenges and optimizing resource allocation in serverless cloud environments. It comprises a hierarchical structure that incorporates multiple policy layers, each responsible for that combines these models, providing a structured overview for understanding their collaborative role n framework of tcns and ensemble policies in cloud management [14] Development of the Experimental Framework and Validation developed to evaluate the performance and efficacy of the integrated world cloud environments. A series of controlled experiments and simulations were conducted to assess the predictive accuracy, resource utilization efficiency, and overall system performance under varying workload conditions.…”