- 10-20% improvement in job runtime
- Satisfies NASA software requirements
- Achieves high-performance scalable storage
- Drives improved economics
Overview
Wisk Aero has developed the first-ever autonomous electrical vertical take-off and landing (eVTOL) aircraft and is using Amazon Web Services (AWS) to build high performance compute (HPC) clusters to run simulations. The company relies on HPC to run computationally intensive and complex simulations, each of which uses thousands of CPU cores. Purchasing on-premises computers for its HPC workload presented challenges, such as cost and managing enough CPU cores for peak runs. Wisk Aero migrated its HPC clusters to AWS to improve job runtime, achieve scalable storage, and drive improved economics.
Opportunity | Using Amazon EC2 to Improve Job Runtime for Wisk Aero
Wisk Aero is an aviation company focused on developing eVTOL aircraft and revolutionizing mobility through quiet, fast, and clean air travel. The company has over 10 years of experience, has locations around the world, and is backed by the Boeing Company and Kitty Hawk Corporation. To study the in-flight airflow, Wisk Aero engineers perform computational fluid dynamics (CFD) simulations using in-house and NASA CFD applications, such as OVERFLOW and FUN3D. Wisk Aero focuses more on using CFD than traditional aircraft builders because CFD supports rapid design iteration as the team explores different aircraft designs and architectures, especially in the early phase of the design process.
The use of CFD simulations gives engineers a clear understanding of the aircraft’s expected performance under various loading and boundary conditions. Because of the novel design of Wisk Aero’s sixth-generation four-seat self-flying eVTOL, it is not possible to use previous simulations or design models. Wisk Aero engineers rely on HPC to run these computationally intensive and complex CFD simulations, each using thousands of CPU cores. To purchase on-premises computers for these HPC workloads, Wisk Aero would need to spend more on hardware that might go entirely unused when not running at peak jobs. Wisk Aero also had to address the increased operational overhead of managing physical hardware as the size of the on-premises cluster increased. To solve these challenges, Wisk Aero turned to the AWS HPC team and Converge Technology Solutions (Converge), an AWS Advanced Consulting Partner, to assist in migrating the company’s HPC simulations to Amazon Elastic Compute Cloud (Amazon EC2), which offers secure and resizable compute capacity for virtually any workload.
“Using AWS, we quickly scaled and added the needed on-demand compute power for the CFD team, compared with the months required and significant capital to build and scale an on-premises HPC cluster.” –Colin Haubrich, Head of IT, Wisk Aero
Solution | Choosing AWS for Agility, Elasticity, Storage, and Security
The Converge client-executive supporting Wisk Aero for its on-premises infrastructure introduced Converge’s Cloud Platforms team and its AWS offerings to the engineering manager of core infrastructure at Wisk Aero. Converge shared a similar use case when Converge—using its AWS Competency Program, which highlights AWS technical expertise and specialization—helped the client successfully migrate its HPC workload to AWS.
Converge, alongside the AWS HPC team, created a pilot environment on AWS for the Wisk Aero team. The fully funded environment helped Wisk Aero to benchmark performance of the Amazon EC2 Hpc6a Instances—HPC instances powered by 3rd generation AMD EPYC processors—and run the necessary software to simulate a smooth transition to AWS. In addition to meeting technical and performance requirements, Wisk Aero worked with Converge to make sure the financial model for using AWS was also part of the pilot deliverables.
After the successful pilot, Wisk Aero chose to use AWS for another round of CFD simulations for its eVTOL aircraft…
Read the full case study to learn more. Reminder: You can learn a lot from AWS HPC engineers by subscribing to the HPC Tech Short YouTube channel, and following the AWS HPC Blog channel.