Seagate Accelerates Data Storage Innovation With GPU-Powered Engineering Simulations

In the race to meet the ever-growing demand for data storage, Seagate Technology is leveraging cutting-edge GPU-accelerated computing to revolutionize its design process. By harnessing the power of NVIDIA GPUs, Seagate has dramatically accelerated its computational fluid dynamics (CFD) simulations, enabling faster innovation and more efficient product development.

Seagate’s engineering team faced a significant challenge: Traditional CPU-based simulations were becoming a bottleneck in the design process. Enter GPU acceleration, which has transformed Seagate’s workflow, enabling:

• Simulation speed increased by up to 50 times
• Ability to run larger, more complex models
• Reduced time to market for new products

“Through more upfront modeling and design optimization, Seagate engineers reduce prototyping and physical testing,” says Kent Forbord, Senior Director of Advanced Mechanical Development at Seagate. “This helps them meet the ultimate goal of shortening our design cycle.”

With the ability to distribute tasks across thousands of cores simultaneously, GPUs excel in handling complex computations swiftly. Ansys collaborates closely with NVIDIA to leverage this parallel processing, continually refining and enhancing GPU-accelerated simulation solvers and algorithms. The collaboration focuses on optimizing various aspects of simulation workflows, including model building, solution algorithms, post-processing, and visualization tasks—helping drive energy efficiency, performance, and scalability to new heights. With the release of Ansys Fluent’s multi-GPU solver in 2022, users can achieve speedups of 30X or more. Such performance gains are just one benefit, as studies have shown that engineers who run their fluids simulations on GPUs can reduce hardware costs by up to 7X and energy consumption by up to 4X. These energy savings can help customers move forward toward their long-term sustainability goals.

Previously, Seagate’s hardware included a workstation comprising 40-core CPUs, resulting in a time-consuming, cumbersome workflow. Seagate’s CPU-based workstation runtimes were approximately one month for chassis-level aeroacoustics models with a mesh size of about 50 million cells, while finer, higher-quality mesh containing up to 100 million cells took even longer. Today, Seagate harnesses the full power of an NVIDIA DGX™ system to run the Fluent GPU solver.

With the ability to run more simulations in less time, Seagate is pushing the boundaries of what’s possible in data storage. By exploring novel designs that were previously too time-consuming to simulate, engineers can optimize airflow within the HAMR-enabled Mozaic 3+ hard drives for better performance and reliability. 

The increased efficiency of GPU-accelerated simulations aligns with Seagate’s commitment to efficiency, resulting in:

• Reduced energy consumption in the design process
• Less physical prototyping required, reducing material waste
• Developing more energy-efficient storage solutions