July 13, 2023
The ever-increasing demand for faster and more accurate engineering simulations has driven the need for cutting-edge technologies that can handle complex computational workloads efficiently. In a groundbreaking collaboration, Ansys has partnered with AMD and Amazon Web Services (AWS) to deliver an exceptional performance boost for engineering simulations.
A recent study titled, “Ansys Fluent on Amazon EC2 Instances Featuring AMD EPYC™ 7003 Series Processors” showcases the remarkable potential of this partnership, transforming the way engineers approach computational fluid dynamics (CFD) simulations using Ansys Fluent. Let's dive into the key findings of this study and uncover the benefits of this collaboration.
CFD simulations heavily rely on the performance and efficiency of the underlying hardware infrastructure. These processors feature high core counts, massive memory bandwidth, and exceptional floating-point performance, making them ideal for computationally intensive tasks like fluid dynamics simulations. The benchmarking results showcase substantial performance improvements when using AMD EPYC™ processors on Ansys Gateway powered by AWS, achieving faster simulation turnaround times and increased productivity for engineers to make quicker design decisions. By leveraging the inherent capabilities of AMD EPYC™ processors, engineers can experience significant acceleration of their Fluent simulations.
Scalability is a crucial aspect of engineering simulations, particularly when dealing with large and complex models. The study highlights the scalability advantages offered by the combination of AMD EPYC™ processors and Amazon EC2 instances. With the ability to scale up to thousands of cores, engineers can tackle simulations that were previously impractical or time-consuming. Furthermore, the collaboration between AMD, Ansys and AWS provides a cost-effective solution, allowing engineers to pay only for the computational resources they need, thereby optimizing their budget while maintaining high-performance simulations.
The study emphasizes the seamless migration process from on-premises or other cloud environments to AWS. Engineers using Fluent can easily transition their simulations to AWS HPC instances powered by AMD EPYC™ processors without the need for extensive code modifications or reconfigurations. This simplicity ensures a smooth and efficient transfer of simulation workloads, enabling engineers to benefit from the enhanced performance and scalability of this powerful collaboration promptly.
The collaboration between AMD, Ansys, and AWS not only delivers enhanced performance but also streamlines the overall simulation workflow, leading to improved productivity. With AWS providing a robust and scalable cloud infrastructure, engineers can access the required computational resources on-demand, eliminating the need for costly on-premises hardware investments and maintenance. The simplicity of provisioning Amazon EC2 instances on Ansys Gateway powered by AWS and the ease of managing simulation jobs allowing engineers to focus more on their simulations and less on infrastructure management.
The performance brief displays Ansys Fluent 2023 R1 running the benchmarks on Ansys Gateway powered by AWS shown in Figures 1 and 2 on Amazon EC2 Hpc6a instances featuring AMD EPYC 7003 Series Processors. Each instance includes 96 physical cores with AMD Simultaneous Multithreading (SMT) disabled and a 100 Gbps Elastic Fabric Adapter networking for fast internode communication that supports scaling from one to 16 instances.
For the larger models, Ansys Fluent generally exhibits super-linear scaling through 16 instances by delivering an average speedup up of ~17.73x at 16 instances (1536 cores). For the Open Racecar 280M model, the average speedup is ~7.87x at 16 instances (1536 cores) versus two instances which is almost an ideal (linear) scaling.1
This study demonstrates the exceptional benefits of the collaboration between Ansys, AMD, and AWS for engineering simulations using Fluent on Ansys Gateway powered by AWS. By leveraging the high-performance computing capabilities of AMD EPYC™ processors and the scalability of Amazon EC2 instances, engineers can achieve faster simulation turnaround times, improved scalability, and optimized cost efficiency. This powerful collaboration paves the way for engineers to tackle complex simulations, make quicker design decisions, and drive innovation across industries.