原廠網址:http://www.nvidia.com/object/tesla-msc-nastran-accelerations.html
鴻鵠國際為台灣地區官方合作廠商,並且可提供技術支援與收費的教育訓練。
歡迎來電或來信諮詢。
MSC Nastran 2012 Acceleration on NVIDIA GPUs
NVIDIA and MSC Software have collaborated to deliver the power of GPU computing for Nastran customers. Available in the latest release of MSC Nastran 2012, NVIDIA GPU acceleration enables faster results for more efficient computation and job turnaround times, delivering more license utilization for the same investment.
Improve Productivity with GPU-Accelerated MSC Nastran 2012:
> MSC Nastran 2012 for structural response simulations
MSC Software's Press Release, 3 November 2011: MSC Nastran 2012 Delivers Significantly Expanded Simulation of Real World Behaviors
NVIDIA Press Release: GPU-Accelerated Engineering Simulation Application to Enable Highest Quality Products for Aerospace and Other Mission Critical Fields
Recent performance studies conducted together with MSC engineers demonstrated that the Tesla C2050 GPU acceleration of Intel Xeon (Westmere and Nehalem) CPUs resulted in speed-ups in the range of 4-6X with a single GPU over a serial run and in the range of 1.4-1.8X with 2 GPUs over a 8 core DMP run.
The GPUs supported with this implementation are the NVIDIA Tesla 20-series and Quadro GPUs based on the Fermi architecture (compute capability 2.0). Linux and Windows 64-bit platforms are supported.
Sparse direct solver intensive SOL101 (linear statics), SOL400 (nonlinear) and SOL108 (direct frequency) analysis are potential candidates for GPU computing. SOL108 would need a complex sparse direct solver that is not supported in the present implementation (MSC Nastran 2012). Likewise, conventional SOL111 (modal frequency) also should benefit from GPU computing in a later release.
The powerful GPU computing capabilities in MSC Nastran 2012 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 v4.0 (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.
Workstations
> Existing System
Tesla C2050 (3GB)
Tesla C2075 (6GB)
> New System Purchase
Total 8-12 CPU cores
Total 32-64 GBs of CPU memory
Disk with minimum 2x500 GB
> GPU OPTIONS
Tesla C2050 or Tesla C2075
Quadro 400 for pre/post
> OR a single GPU
Quadro 6000 for compute and pre/post
Servers
> Existing System
Tesla S2050 (3GB)
Tesla S2070 (6GB)
> New System Purchase
Total 4 CPUs with 6-8 CPU cores each
Total 4 x16 PCIe (one for each GPU)
Total 128 to 192 GBs of CPU memory
Disk with minimum 2000 GB (scratch)
4 x Tesla M2050 or Tesla M2070
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.