GPU_Computing_Guide_2019.pdf-资料库

liuliu449017-11990289-16359647583859058666.pdf-第1页.png

第1页 / 共33页

liuliu449017-11990289-16359647583859058666.pdf-第2页.png

第2页 / 共33页

liuliu449017-11990289-16359647583859058666.pdf-第3页.png

第3页 / 共33页

liuliu449017-11990289-16359647583859058666.pdf-第4页.png

第4页 / 共33页

liuliu449017-11990289-16359647583859058666.pdf-第5页.png

第5页 / 共33页

liuliu449017-11990289-16359647583859058666.pdf-第6页.png

第6页 / 共33页

liuliu449017-11990289-16359647583859058666.pdf-第7页.png

第7页 / 共33页

liuliu449017-11990289-16359647583859058666.pdf-第8页.png

第8页 / 共33页

CST STUDIO SUITE R 2019 GPU Computing Guide 9 1 0 2 i e d u G g n i t u p m o C U P G s e m `e t s y S t l u a s s a D c A I L U M I S / M O C . S D 3 Copyright 1998-2019 Dassault Systemes Deutschland GmbH. CST Studio Suite is a Dassault Syst`emes product. All rights reserved.

2 9 1 0 2 i e d u G g n i t u p m o C U P G s e m `e t s y S t l u a s s a D c A I L U M I S / M O C . S D 3 Contents 1 Nomenclature 2 Supported Solvers and Features 2.1 Unsupported Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Operating System Support 4 Supported Hardware 5 Unsupported Hardware 6 NVIDIA Drivers Download and Installation 6.1 GPU Driver Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Verifying Correct Installation of GPU Hardware and Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Uninstalling NVIDIA Drivers 7 Switch On GPU Computing 7.1 Interactive Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Simulations in Batch Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Usage Guidelines 8.1 The Error Correction Code (ECC) Feature . . . . . . . . . . . . . . . . . . 8.2 Tesla Compute Cluster (TCC) Mode . . . . . . . . . . . . . . . . . . . . . 8.3 Disable the Exclusive Mode . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4 Display Link . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.5 Combined MPI Computing and GPU Computing . . . . . . . . . . . . . . 8.6 Service User . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7 GPU Computing using Windows Remote Desktop (RDP) . . . . . . . . . . 8.8 Running Multiple Simulations at the Same Time . . . . . . . . . . . . . . . 8.9 Video Card Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.10 Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.11 Latest CST Service Pack . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.12 GPU Monitoring/Utilization . . . . . . . . . . . . . . . . . . . . . . . . . . 8.13 Select Subset of Available GPU Cards . . . . . . . . . . . . . . . . . . . . 9 NVIDIA GPU Boost 10 Licensing 11 Troubleshooting Tips 12 History of Changes 3 4 4 4 5 14 14 14 18 19 20 20 20 21 21 23 24 24 25 25 25 25 26 26 26 26 27 28 31 31 33

1 Nomenclature The following section explains the nomenclature used in this document. command <...> Commands you have to enter either on a command prompt (cmd on MS Windows or your favorite shell on Linux) are typeset using typewriter fonts. Within commands the sections you should replace according to your environment are enclosed in ”<...>”. For example ”” should be replaced by the directory where you have installed CST STUDIO SUITE (e.g. ”c:\Program Files\CST STUDIO SUITE”). 3 9 1 0 2 i e d u G g n i t u p m o C U P G s e m `e t s y S t l u a s s a D c A I L U M I S / M O C . S D 3

4 9 1 0 2 i e d u G g n i t u p m o C U P G s e m `e t s y S t l u a s s a D c A I L U M I S / M O C . S D 3 2 Supported Solvers and Features • Transient Solver (T-solver/TLM-solver) • Integral Equation Solver (direct solver and MLFMM only) • Multilayer solver (M-solver) • Particle-In-Cell (PIC-solver) • Asymptotic Solver (A-solver; for this solver TCC mode is required on Windows!) • Conjugate Heat Transfer Solver (CHT-solver) Co-simulation with CST CABLE STUDIO is also supported. 2.1 Unsupported Features The following features are currently not supported by GPU Computing. This list is subject to change in future releases or service packs of CST STUDIO SUITE. Solver Unsupported Features on GPU Transient Solver • Subgridding Particle In Cell Solver • Modulation of External Fields • Open Boundaries 3 Operating System Support CST STUDIO SUITE is continuously tested on diﬀerent operating systems. For a list of supported operating systems please refer to https://updates.cst.com/downloads/CST-OS-Support.pdf In general, GPU computing can be used on any of the supported operating systems.

5 9 1 0 2 i e d u G g n i t u p m o C U P G s e m `e t s y S t l u a s s a D c A I L U M I S / M O C . S D 3 4 Supported Hardware CST STUDIO SUITE currently supports up to 16 GPU devices in a single host system, meaning each number of GPU devices between 1 and 16 is supported.1 The following tables contain some basic information about the GPU hardware currently supported by the GPU Computing feature of CST STUDIO SUITE, as well as the require- ments for the host system equipped with the hardware. To ensure compatibility of GPU hardware and host system please check https://www.nvidia.com/object/tesla-qualiﬁed-servers.html Please note that a 64 bit computer architecture is required for GPU Computing. A general hardware recommendation can be found here: https://www.cst.com/products/csts2/hardwarerecommendation 1It is strongly recommended to contact CST before purchasing a system with more than four GPU cards to ensure that the hardware is working properly and is conﬁgured correctly for CST STUDIO SUITE.

6 9 1 0 2 i e d u G g n i t u p m o C U P G s e m `e t s y S t l u a s s a D c A I L U M I S / M O C . S D 3 CST STUDIO SUITE oﬃcially supports the Nvidia Tesla and Quadro cards listed in the table below. That means that these GPUs are well tested and validated with CST software and you can contact CST support in case you run into any problems. List of supported GPU hardware for CST STUDIO SUITE 2019 2 3 Platform Series Card Name Quadro RTX 8000 4 Turing Workstations Quadro RTX 6000 4 Turing Workstations Quadro RTX 5000 4 Turing Workstations Quadro RTX 4000 4 Turing Workstations Volta Workstations Quadro GV100 Servers Tesla V100-SXM2-32GB (Chip) Volta Servers Tesla V100-PCIE-32GB Volta Tesla V100-SXM2-16GB (Chip) Volta Servers Servers Volta Tesla V100-PCIE-16GB Servers Pascal Tesla P100-SXM2 (Chip) Servers Pascal Tesla P100-PCIE-16GB Tesla P100 16GB Pascal Servers Pascal Tesla P100-PCIE-12GB Servers Quadro P6000 4 Pascal Workstations Quadro GP100 Pascal Workstations Tesla P40 4 Pascal Tesla P4 4 Pascal Tesla M60 4 Maxwell Servers/Workst. Tesla M40 4 Maxwell Servers Quadro M6000 24GB 4 Maxwell Workstations Quadro M6000 4 Maxwell Workstations Tesla K80 Kepler Kepler Tesla K40 m/c/s/st/d/t Kepler Workstations Quadro K6000 Tesla K20X Kepler Kepler Tesla K20m/K20c/K20s Tesla K10 4 Kepler Servers Servers/Workst. Servers Servers Servers/Workst. Min. CST Version 2019 SP6 2019 SP6 2019 SP6 2019 SP6 2018 SP6 2018 SP6 2018 SP6 2018 SP1 2018 SP1 2017 release 2017 release 2017 release 2017 SP2 2017 SP 2 2017 SP2 2018 release 2018 release 2016 SP4 2016 SP4 2016 SP4 2015 SP4 2014 SP6 2013 SP5 2013 SP4 2013 release 2013 release 2013 release Servers Servers 2Please note that cards of diﬀerent series (e.g. ”Maxwell” and ”Pascal”) can’t be combined in a single host system for GPU Computing. 3Platform = Servers: These GPUs are only available with a passive cooling system which only provides suﬃcient cooling if it’s used in combination with additional fans. These fans are usually available for server chassis only! Platform = Workstations: These GPUs provide active cooling, so they are suitable for workstation com- puter chassis as well. 4 Important: The double precision performance of this GPU device is poor, thus, it is recommended for T-solver simulations only.

7 9 1 0 2 i e d u G g n i t u p m o C U P G s e m `e t s y S t l u a s s a D c A I L U M I S / M O C . S D 3 Hardware Type NVIDIA Tesla P100 Chip (for Servers) NVIDIA Tesla P100 PCIe1 (for Servers) Min. CST version required 2017 release 2017 release Number of GPUs Max. Problem Size (Transient Solver) Form Factor Memory Bandwidth 1 1 approx. 160 million mesh cells approx. 160 / 120 million mesh cells Chip Passive Cooling Dual-Slot PCI-Express Passive Cooling 16 GB CoWoS HBM2 16 / 12 GB CoWoS HBM2 732 GB/s 732 GB/s / 549 GB/s Single Precision Performance 2 10.6 TFlops Double Precision Performance 2 5.3 TFlops 9.3 TFlops 4.7 TFlops Power Consumption 300 W (max.) 250 W (max.) System interface NVIDIA NVLink 1x PCIe Gen 3 (x16 electrically) Power Supply of Host System min. 750 W Min. RAM of Host System 3 64 GB min. 750 W 64 GB 1 The 12 GB version has about 25 percent less performance compared to the 16 GB version. 2 Measured with BOOST enabled 3The host system requires approximately 4 times as much memory as is available on the GPU cards. Al- though it is technically possible to use less memory than this recommendation, the simulation performance of larger models will suﬀer. CST assumes no liability for any problems caused by this information.

8 9 1 0 2 i e d u G g n i t u p m o C U P G s e m `e t s y S t l u a s s a D c A I L U M I S / M O C . S D 3 Hardware Type NVIDIA Quadro GP 100 (for Workstations) NVIDIA Quadro P60001 (for Workstations) Min. CST version required 2017 SP 2 2017 SP 2 Number of GPUs Max. Problem Size (Transient Solver) Form Factor Memory Bandwidth 1 1 approx. 160 million mesh cells approx. 240 million mesh cells Dual-Slot PCI-Express Dual-Slot PCI-Express 16 GB HBM2 24 GB GDDR5X 720 GB/s Single Precision Performance 2 10.3 TFlops Double Precision Performance 2 5.2 TFlops 432 GB/s 12.0 TFlops 0.2 TFlops Power Consumption 300 W (max.) 300 W (max.) System interface 1x PCIe Gen 3 (x16 electrically) 1x PCIe Gen 3 (x16 electrically) Power Supply of Host System min. 750 W min. 750 W Min. RAM of Host System 3 64 GB 96 GB 1 The double precision performance of this GPU device is poor, thus, it is recommended for T-solver simulations only. 2 Measured with BOOST enabled 3The host system requires approximately 4 times as much memory as is available on the GPU cards. Al- though it is technically possible to use less memory than this recommendation, the simulation performance of larger models will suﬀer. CST assumes no liability for any problems caused by this information.

资料库

GPU_Computing_Guide_2019.pdf

相关推荐

开发技术

热门标签

最新资料