Zeleros, a startup developing an innovative hyperloop system to revolutionize ground travel, optimized the quality and efficiency of its simulations with high performance computing (HPC) on Amazon Web Services (AWS). Migrating HPC workloads to AWS bolstered Zeleros’ capabilities, while the Rescale platform for HPC automation further optimized the company’s performance and productivity. The combination of Rescale and AWS HPC enabled Zeleros to achieve faster computational fluid dynamics (CFD) simulations at a fraction of the cost, streamline workflows for sustainable growth, and integrate innovation across the company.
Team innovation outpaced local computing capabilities
At Zeleros, Dynamics Engineer Federico Lluesma was charged with managing the development of hyperloop model simulations to test and perfect definitive attributes of the company’s prototype and ensure real-world performance. As Lluesma led the team in designing these increasingly innovative and complex aerodynamics simulations, he quickly realized that the company’s local computing capabilities lacked the flexibility to accommodate the changing requirements for simulations and modeling. He also observed significant impediments to productivity, including scarcity of on-premises computing capacity as computers were fully occupied by a single simulation. In search of a solution that could comprehensively enhance and support their process, the Zeleros team turned to AWS for cloud-based HPC.
Rescale and AWS HPC accelerate simulations from one week to one day
With HPC on AWS, the Zeleros team found the scalability, flexibility, and memory required to develop their simulations. Using the Rescale HPC automation platform— which is built for hybrid and cloud HPC, and geared specifically for companies in the R&D phase—Zeleros was able to further optimize AWS services. Engaging Rescale significantly transformed the capabilities of the Zeleros team. In addition to application lifecycle management that streamlined software configuration and licensing updates, Rescale’s automation support unlocked new simulations that had been impossible to execute. By transitioning from their local HPC system to Rescale on AWS HPC, the team’s turnaround time for aerodynamic simulations was instantly reduced from one week to one day. “In some ways, the time we saved was infinite,” Lluesma said. “We are doing simulations that we could not do before. For aerodynamics specifically, our engineers have reached a level of performance that teams in other fields have not yet achieved. And just by unblocking a computer, we’re able to run several cases in parallel.”