Engineering companies must constantly reinvent themselves in the global and dynamic market environment—most recently reinforced by the trend toward sustainability and positioning in the field of e-mobility. This puts high pressure on the research and development (R&D) departments, which have to realize ever faster innovation cycles and develop and test new products for a wide range of challenging operating conditions.
Computer-aided engineering (CAE) and numerical simulations accelerate this product development process: with these tools, engineers can, for example, predict the optimal configuration for cooling an electric motor or optimize gearboxes in terms of losses and longevity. Dive uses a cloud-native simulation environment to take the complexity out of this process and offers a modern Software as a Service (SaaS) package—all based on Microsoft Azure.
The challenge: Cumbersome, limited on-premises environments are an outdated norm
“Experiments are out. Simulations are the way forward,” says Johannes Gutekunst, founder and Chief Technology Officer (CTO) at dive solutions GmbH. For a long time, test stands and prototypes were the standard approach for engineers. Computer-aided simulation began making inroads in a variety of industries in the early 1990s. But this requires large data centers, which are expensive for companies to set up and require constant on-premises hardware support. Realistic representation of use cases ties up computers for weeks solving physical equations, so if the hardware is busy with one project, the next one has to wait. If, on the other hand, there is no simulation to run, the data center incurs running costs but doesn’t deliver any results.
In light of the enormous demands on hardware and the unprecedented triumph of SaaS in other industries, right from the start it was clear to Gutekunst that Dive needed to develop a cloud-native SaaS solution. However, this ran counter to the concerns of its customers, who mainly come from the R&D departments of large automotive manufacturers and production companies. “This is a fairly conservative market that to date has been fairly cloud-averse.”
Dive had to overcome customers’ security concerns and a lack of references. “In the beginning, we couldn’t say ‘Others are doing it this way too,’ or ‘You can see the result here,’” Gutekunst says. The development of today’s machines involves the processing of highly sensitive data during simulations. “Data security is a company’s holy grail, with everything subject to the strictest secrecy. So our biggest hurdle wasn’t technological; it was the mindset of our target group,” Gutekunst says.
The solution: Cloud-native CAE based on a secure and scalable Microsoft Azure hardware environment
Dive developed its first version on a different cloud platform, but “when we saw that our product was well-received, we wanted to build it again—and better,” Gutekunst says. To this end, he and his team switched to Microsoft Azure. “Especially in large companies, Microsoft enjoys excellent standing and a high level of trust. Many of our customers have very stringent compliance requirements, so the fact that Azure offers high security standards and Europe as a data storage location is quite persuasive,” Gutekunst explains.
Dive itself also benefits from the move to Azure. “When it comes to high-performance computing (HPC), Microsoft is ahead of the game,” says Gutekunst. “We’re using that to our advantage. We don’t have to focus on server operations, but can invest our energy in developing the optimal code.” He praises the support provided by the Microsoft for Startups program and the collaboration with industry experts. “Microsoft understands enterprise customers and has helped us build good solutions for them,” Gutekunst says. “One main thing we realized is that our customers crave performance. The cloud can win them over in this respect as well.”
To accelerate development cycles, companies often work on multiple projects simultaneously and take full advantage of the scalability of the cloud. This means that “it’s not uncommon for allocations of central processing unit (CPU) cores to peak at several thousand, and for resource allocation to fluctuate by a factor of ten within a few hours. Azure HPC systems fit very well with the heterogeneous requirements of our customers. And they pay only for the capacity they actually use,” Gutekunst says.
Meanwhile, Dive is working on its next advance: soon the startup will offer additional use cases and integrate physical models for thermodynamic analyses, for example. “In the future, we will offer an integrated cloud-native experience for CAE,” Gutekunst says. Here, Dive is working with Microsoft and its hardware partners to optimize the hardware-software interface to achieve even faster simulations and analysis.
“Many of our customers have very stringent compliance requirements, so the fact that Azure offers high security standards and Europe as a data storage location is quite persuasive.”
Johannes Gutekunst, founder and CTO, dive solutions GmbH
Follow Microsoft