Dassault Systèmes has announced that Abaqus 6.11, the new release of the unified finite element analysis (FEA) product suite from its Simulia brand, utilises Nvidia Quadro and Tesla GPUs, coupled with CPUs, to run computer-aided engineering (CAE) simulations twice as fast as with CPUs alone.
Two of the largest car manufacturers in Europe evaluated Abaqus FEA to analyse the structural behaviour of large engine models. In each case their Nvidia GPU-accelerated simulations were completed in half the time it would have taken using CPUs alone. For the automotive industry, this acceleration enables CAE engineers to perform more analyses in a given period of time, identifying issues earlier and decreasing time to market.
Nvidia GPUs are based on the Cuda parallel processing architecture and can be programmed using industry-standard languages such as C, C++, and Fortran. A broad range of computing applications has been modified to take advantage of GPU acceleration. With power-efficient cores and increasingly fast access to memory, mechanical simulation software from Simulia enables design companies and manufacturers to use computer simulations instead of expensive prototyping.