What’s New and What’s Coming for RADIOSS® on Intel®

It’s been an exciting two years for Altair solver software as our product quality and feature set has improved dramatically. I want to take a moment to update you on what’s going on with RADIOSS, our leading structural analysis solver that’s become a standard for crash test (and other compute-intensive) simulations.

We’re proud that RADIOSS is the most scalable solver in its class, and one of the top solvers in auto, aero, consumer goods, electronics, sports and other industries. It’s really hard to believe we’ve come so far…back in 1987 it took 20 hours to perform the first full car crash computation — with only 20,000 elements and very limited accuracy. Today we can perform a 15 million element car crash simulation in less than 5 hours!

What makes this possible is RADIOSS’s massively parallel architecture — we use Intel MPI coupled with OpenMP to achieve a Hybrid Massively Parallel Program (HMPP), for the best scalability in the marketplace.

Speaking of Intel, a big part of Altair’s quest to achieve higher levels of performance for RADIOSS has been our partnership with Intel. In addition to the Intel MPI element of the HMPP architecture, we use a wide range of other Intel tools like Intel MPI, Intel Fortran & C++ compilers, Intel MKL Library, Intel VTune™ Amplifier, Intel Trace Analyzer and Collector. RADIOSS is also Intel Cluster Ready certified, so users know it is ready to run on any Intel certified cluster.

And of course, Intel has offered their close collaboration and access to hardware resources plus technical expertise to help us optimize our software on Intel systems. You can read about Altair and Intel’s collaboration here.

What we’re excited about for Q1 2014 is the outstanding performance we’ve seen on our most recent RADIOSS testing with Intel Xeon® processor E5-2600 v2 product family processors. From testing a series of processors, from Woodcrest all the way up to Intel Xeon processor E5-2600 v2, we saw performance levels on a typical cluster node achieve an order of magnitude (10x) improvement!

See below for the performance results from this testing:IMG_24042014_155103








This breakthrough comes from significant hardware and software optimizations — the most important factor is the increase in number of cores and the scalability achieved by RADIOSS. If you are interested in getting a copy of RADIOSS 12.0, which runs out of the box on all Intel Xeon processor E5-2600 v2 product family processors, click here to begin the process.

Most recently, we announced some excellent results for RADIOSS on future Intel Xeon processor E7-8800/4800/2800 v2 product families. Intel kindly provided Altair with access to their SDP platform for testing of an upcoming product. We saw very impressive absolute performance and have shared those on the Intel blog. RADIOSS runs 2.75x faster on the 4-way Intel Xeon processor E7-4890 v2 @2.80GHz – just another milestone on the road to better engineering with Altair and Intel!

Eric Lequiniou
Eric Lequiniou

About Eric Lequiniou

An expert in software optimization and parallelization on clusters, multi-core architectures, accelerators and coprocessors, Eric developed the Hybrid MPP parallel version of RADIOSS finite element software. After receiving an MSc degree in computer science, Eric started his career in 1994 at CNRS, Laboratoire Informatique du Parallélisme. He joined Mecalog in 1994 and worked for this company until its acquisition by Altair in 2006. Eric holds an Executive MBA from HEC French business school obtained in 2007.