GPU Computing Solutions

SIMULIA Abaqus Acceleration on NVIDIA GPUs

NVIDIA and SIMULIA, the Dassault Systèmes brand for Realistic Simulation, have collaborated to deliver the power of GPU computing for Abaqus customers. Available in the Abaqus 6.11 release, NVIDIA GPU acceleration enables faster results for more efficient computation and job turnaround times, delivering more license utilization for the same investment.

Innovate Faster with GPU-Accelerated Abaqus FEA:

> Abaqus 6.11 for structural response simulations

> Abaqus presentations at the GTC 2010 and NAFEMS 2011 World Congress:

Performance studies conducted together with SIMULIA engineers and customers resulted in the following GPU speed-up performances with Abaqus 6.12:

Abaqus 6.12 is used on a SuperMicro node with Sandy Bridge CPUs 2.6GHz [8 core], Tesla K20X GPU

For a test drive of Abaqus 6.12 on GPUs, please visit www.AccelerateAbaqusonGPU.com. This test drive site also supports a ROI Calculator that provides a performance and cost analysis of Abaqus/Standard on Generation N-1 (Westmere CPU + Fermi GPU) and Generation N (Sandy Bridge CPU + Kepler GPU) server configurations.

Recommended Professional Products
The powerful GPU computing capabilities in Abaqus were developed on NVIDIA Tesla and Quadro GPU computing products and require the use of recent CUDA-capable NVIDIA GPUs, such as the Tesla 20-series products, and supporting CUDA v3.1 or above (learn more).

Tesla and Quadro GPU computing products are designed to deliver the highest computational performance with the most reliable numerical accuracy, and are available and supported by the world's leading professional system manufacturers.

Tesla Benefits
Highest Computational Performance
> High-speed double precision operations
> Large dedicated memory
> CUDA-Accelerated math libraries
> High-speed bi-directional PCIe communication
> NVIDIA GPUDirect™ with InfiniBand
Most Reliable
> ECC memory
> Rigorous stress testing
> Abaqus QA testing
Best Supported
> Professional support network
> Large and growing developer community
> OEM system integration
> Long-term product lifecycle
> 3 year warranty
> Cluster & system management tools
   (server products)
> Windows remote desktop support
Recommended Configurations
> New System Purchase

Each node with
  • 2 CPU sockets (6 to 8 cores each); recommend 1 GPU per CPU socket
  • Total 96 to 192 GBs of CPU memory
  • Disk with minimum 2000 GB (scratch)
GPU OPTIONS: 2 x Tesla K20X/K20 or Tesla M2090

> Existing System
  • Tesla C2075 or K20c (6GB)
> New System Purchase
  • Total 2 CPU sockets; recommend 1 GPU per CPU socket ;
  • Total 48 GBs of CPU memory or more;
  • Disk with minimum 500 GB
  • Tesla K20c
  • Quadro 4000/K5000 for pre/post

OR a single GPU
  • Quadro 6000 for compute and pre/post

NVIDIA Tesla and Quadro products are available from all major professional workstation OEMs. Only Tesla GPU computing products are designed and qualified for compute cluster deployment.

Buy Optimized Tesla Systems

NVIDIA partners with system vendors to provide optimal solutions that accelerate your HPC workloads. Buy now and enjoy all the benefits of GPU-acceleration on SIMULIA Abaqus.

For a complete list of Tesla Preferred Providers, click here.