Solving The Toughest Engineering Challenges With HPC
Migrating to the cloud can enable engineers and scientists to remove engineering bottlenecks and quicken optimization cycles, providing design simulation results nearly instantly.
Engineering Computing News
Engineering Computing Resources
June 5, 2019
“Five years from now, people will look back and wonder why they did not migrate to the cloud earlier,” thinks Ian Campbell, CEO of OnScale, about the people who continue to rely on legacy high-performance computing (HPC) and manage their own data centers.
He believes that HPC workloads for science and engineering will inevitably migrate to the cloud, enabling engineers and scientists to remove engineering bottlenecks and quicken optimization cycles, providing design simulation results nearly instantly.
“At OnScale, we want to be the enabler of the cloud HPC capability and want to ensure that people understand that cloud HPC is the Future of Engineering, where engineers are not restricted by performance or cost constraints.”
Earlier in his career, Campbell faced difficulties while using legacy computer-aided engineering (CAE) tools running on local HPC servers, as they hindered the designing and optimization of IoT sensors. OnScale fixes that problem. Founded in 2017, OnScale is a spin-off from Thornton Tomasetti, a global leader in science and engineering consulting services. Although the firm is relatively new in the market, it employs an old, proven technology called ‘solver’.
OnScale addresses the rapidly increasing demands for advanced CAE and cloud HPC resources to tackle the engineering problems in IoT sensors, 5G RF, biomedical, and advanced driver-assistance systems (ADAS).
These industries are worth tens of billions of dollars, based on HPC market projections from Hyperion and Grandview Research, combined with CAE market projections. The company caters to the needs of engineers across varied disciplines by giving them ‘solver’ capability in the cloud to address their toughest next-generation engineering challenges.
It brings a SaaS-based cloud platform that can reduce the complex engineering design challenges. However, unlike software-as-a-service, this acronym refers to OnScale’s ‘Solver-as-a-Service’ platform, which offers unparalleled CAE software tools and multi-physics solvers with unlimited capabilities of scalable cloud HPC.
Campbell believes that the only way to get over the engineering hurdle is to have massive amounts of HPC cores running in parallel. Organizations can efficiently break the cost and performance barriers by purchasing only the core hours they need using OnScale’s SaaS platform.
All of OnScale’s clients are renowned companies from highly competitive industries such as IoT, 5G, biomedical, and ADAS that look forward to protecting their core value, which is IP.
“Unlike software-as-a-service, SaaS refers to ‘Solver-as-a-Service’ platform that combines unparalleled CAE software tools and multi-physics solvers with unlimited capabilities of scalable cloud HPC.”
The firm’s SaaS platform is built on AWS, ensuring an unquestionable security mechanism. The framework offers end-to-end security by storing all the user credentials in Cognito, an AWS component. OnScale uses AWS’s data encryption schemas to protect data for its customers that remains secured even when it is transferred.
Encryption is completely handled by AWS Key Management System (KMS) without OnScale retaining any user identity or authentication data. “We believe that it is customers’ cloud which is delivered by AWS and we provide a very convenient means to abstract away the security components and any IP issues so the engineers can just worry about solving their problems and not IP or security,” emphasizes Campbell.
OnScale launched its product a few weeks ago and has been offering it to customers all over the world across Silicon Valley, UK, Japan, Korea, China, Germany, and France.
Currently, the company is working with some of the biggest names in component and engineering design—including Siemens, Qualcomm, TDK InvenSense and more—providing them the support for tackling the toughest engineering challenges. The company is strengthened by a team of capable engineers, scientists, and developers, dedicated to delivering the best of CAE tools on next-generation cloud HPC.
While numerous industries such as automotive, aerospace, and general mechanical engineering demand multi-physics solvers, CAE tools, and cloud HPC, Campbell lays his focus on addressing the needs of underserved markets that big players do not reach, such as the aforementioned 5G, IoT, ADAS, and BioMed. OnScale’s current goal is to support businesses in the underserved markets and later expand to bigger markets after growing its capabilities. Ultimately, OnScale aims to own the market and provide the Future of Engineering.