Altair RADIOSS now supported on Microsoft Azure

In July 2015, we launched Linux RDMA on Microsoft Azure. With this release we embarked on a cloud journey to make high performance computing (HPC) mainstream so every engineer or scientist on the planet can take advantage of super computing capabilities to solve complex engineering problems.

In this quest to make HPC more mainstream we understand the importance of the ISV ecosystem. We want to create and nurture a vibrant, accessible and open ISV ecosystem for the end users and for our partners. This approach puts the end users in the driver’s seat to select and choose key components for their HPC stack on Azure.

One of the key elements of any HPC stack is the application layer or as in the CAE (Computer Aided Engineering) world they call it a “solver.”  Solvers/applications are at the heart of any HPC system. Engineers and scientists are using these applications across massively parallel computer systems to solve complex problems.

Today we are happy to announce support of one of the mainstream CAE solvers, RADIOSS, part of Altair’s HyperWorks suite on Azure. Altair RADIOSS® is an industry leading structural analysis solver for highly non-linear problems under dynamic loadings. It is highly differentiated for scalability, quality and robustness, and consists of features for multi-physics simulation and advanced materials such as composites. RADIOSS is used across all industries worldwide to improve the crashworthiness, safety, and manufacturability of structural designs.

We have been working closely with Altair over the past several months and see excellent performance and scalability of RADIOSS on Microsoft Azure Linux RDMA platform (A9 virtual machines in Azure). Azure can deliver high level of scalability and performance with RADIOSS because of its dedicated high speed low latency network fabric that uses RDMA and Infiniband technology, available only on A8 and A9 instances. The A8 and A9 VM sizes in Azure provide RDMA virtualized through Hyper-V with near “bare metal” performance providing close to three microseconds of latency and greater than 3.5 Gbps bandwidth.  You can clearly see in the chart below how Azure has delivered close to bare metal performance for RADIOSS across hundreds of cores.

We ran “Neon 1 million,” a refined version of the Dodge Neon public domain benchmark, full vehicle crash simulation model using RADIOSS® on Azure A9 instances (Intel E-2670 @ 2.6 GHz, Memory 128GB, 1600MHz DDR3, QDR InfiniBand) using SLES 12 (Suse Linux 12) and Intel MPI.  In comparison we ran RADIOSS on bare metal hardware (Intel E5-2670 @ 2.6 GHz, QDR InfiniBand). In the chart below, the X-Axis represents number of cores and the Y-Axis represents the RADIOSS® solver run time in seconds.

image1

Another important pillar of the Azure HPC cloud ecosystem is the ISV application licensing model. ISV application licensing flexibility and affordability are the keys to make HPC more mainstream. Altair, with its innovative HyperWorks licensing model, has come up with a cloud-friendly, affordable, game-changing licensing model that includes solutions for structural analysis, CFD (computation fluid dynamics), electromagnetic analysis and optimization. More details on RADIOSS.

We encourage you to try HyperWorks RADIOSS on Microsoft Azure A8 and A9 instances by opening a free trial account in Microsoft Azure.

This blog post was originally published on Microsoft Azure Blog.

Tejas Karmarkar

Tejas Karmarkar

Senior Program Manager-Azure High Performance Computing at Microsoft HQ
Tejas is a Senior Program Manager for Azure High Performance Computing in Microsoft HQ in Redmond, WA. He has worked in Microsoft for 9 years starting from the data center business, and then moving on to Private Cloud and then to High Performance Computing. Currently, Tejas is responsible for engineering and business strategy for Big Compute and High Performance Computing solutions running on Microsoft Azure Cloud for the CAE industry.
Prior to joining Microsoft, Tejas worked with Altair for more than 8 years and has extensive experience in the automotive industry. Altogether, he has more than 17 years of experience in the manufacturing industry. Tejas has a Bachelor and Master’s Degree in Mechanical Engineering, as well as an MBA from Michigan State University.
Tejas Karmarkar

Latest posts by Tejas Karmarkar (see all)

Tejas Karmarkar

About Tejas Karmarkar

Tejas is a Senior Program Manager for Azure High Performance Computing in Microsoft HQ in Redmond, WA. He has worked in Microsoft for 9 years starting from the data center business, and then moving on to Private Cloud and then to High Performance Computing. Currently, Tejas is responsible for engineering and business strategy for Big Compute and High Performance Computing solutions running on Microsoft Azure Cloud for the CAE industry.
Prior to joining Microsoft, Tejas worked with Altair for more than 8 years and has extensive experience in the automotive industry. Altogether, he has more than 17 years of experience in the manufacturing industry. Tejas has a Bachelor and Master’s Degree in Mechanical Engineering, as well as an MBA from Michigan State University.