These slides were presented at the NAFEMS World Congress 2025, held in Salzburg, Austria from May 19–22, 2025.

Abstract

The transition from on-premise high-performance computing (HPC) to cloud-based HPC is a rapidly growing trend in the field of engineering simulation. With advancements in cloud technology, organizations are increasingly moving their simulation workloads to the cloud to benefit from greater scalability, flexibility, and accessibility. Cloud-based solutions allow engineers to leverage cutting-edge hardware without the significant upfront investment required for on-premise infrastructure. However, this shift also introduces a set of unique challenges, requiring careful consideration and strategic planning to fully harness the advantages of the cloud. This presentation explores the transition to cloud-based simulation from the perspective of a seasoned simulation engineer with over 25 years of experience, including a decade of specialization in cloud environments. The focus is on providing practical advice and technical strategies for engineers looking to migrate their simulation workloads to the cloud, addressing the unique challenges and opportunities inherent to this paradigm shift. The discussion will begin by examining hardware configuration challenges, particularly how to identify and configure the optimal hardware setups to meet diverse simulation requirements. Understanding the fundamental differences between cloud infrastructure and traditional on-premise systems is critical for optimizing simulation performance. The nuances of data management in cloud environments will also be addressed, including best practices for handling large simulation datasets with regard to storage, transfer protocols, and robust security measures to ensure data integrity and confidentiality. Attention will be given to the entire simulation lifecycle, encompassing preprocessing, simulation execution, and post-processing workflows. Special emphasis will be placed on optimizing these workflows for a distributed cloud architecture to enhance efficiency and scalability. Cost optimization is another key focus, with strategies to balance computational performance against budget constraints by leveraging cloud-native tools, instance selection, and workload scheduling techniques. Furthermore, the complexities of license management in cloud-based settings will be explored, with practical approaches to efficiently allocate and utilize licenses across distributed resources. The presentation also delves into collaborative and data-sharing dynamics enabled by cloud platforms. It highlights methods for fostering seamless teamwork among geographically dispersed engineering teams and ensuring effective sharing of simulation data within and across organizations. Adapting cloud-based workflows to project-specific requirements and deadlines will be discussed, showcasing how flexibility in resource allocation and scaling can align with varied engineering objectives. Real-world examples and case studies will illustrate these concepts, offering actionable insights into overcoming common obstacles while maximizing the benefits of cloud-based simulation. By leveraging cloud technologies, engineers can achieve substantial improvements in efficiency, reduce operational costs, and accelerate innovation cycles. This presentation aims to equip engineers with the knowledge and tools necessary to navigate the complexities of cloud-based simulation and unlock its full potential for modern engineering challenges.