HPC Parallel Computing for FEA - Customer Examples (1 of 4)
4 likes11,502 views
This document details the impact and benefits of high-performance computing (HPC) at Ansys, emphasizing its role in finite element analysis (FEA) and computational fluid dynamics (CFD). Ansys HPC solutions facilitate faster simulations and increased productivity through parallel processing on various computing platforms, resulting in substantial returns on investment. Key applications showcased include enhanced oil recovery and aerospace component simulations, illustrating the efficiency gains and design improvements achievable with HPC technology.
HPC Parallel Computing for FEA - Customer Examples (1 of 4)
- 1. © 2014 ANSYS, Inc. July 11, 20141 Customer Examples of High-Performance Parallel Computing for FEA Wim Slagter, Ph.D ANSYS, Inc. Series of 4 Presentations on HPC 1: High-Performance Parallel Computing for FEA 2: High-Performance Parallel Computing for CFD 3: High-Performance Parametric Computing for FEA 4: High-Performance Parametric Computing for CFD
- 2. © 2014 ANSYS, Inc. July 11, 20142 High Performance Computing (HPC) at ANSYS: An ongoing effort designed to remove computing limitations from engineers who use computer aided engineering in all phases of design, analysis, and testing. It is a hardware and software initiative! HPC Defined
- 3. © 2014 ANSYS, Inc. July 11, 20143 Why HPC? Modal Nonlinear Multiphysics Dynamics Assemblies CAD-to-mesh Capture fidelity Multiple design ideas Optimize the design Ensure product integrity Impact product design Enable large models Allow parametric studies
- 4. © 2014 ANSYS, Inc. July 11, 20144 Why HPC? Results indicate very substantial returns for investments in HPC: $356 dollars on average in revenue per dollar of HPC invested $38 dollars on average of profits (or cost savings) per dollar of HPC invested Key Findings: Financial ROI Results Source: IDC report “Creating Economic Models Showing the Relationship Between Investments in HPC and the Resulting Financial ROI and Innovation”; October 2013, IDC #243296, Volume: 1.
- 5. © 2014 ANSYS, Inc. July 11, 20145 Take Advantage of the HPC Revolution Recent advances have revolutionized the computational speed available on the desktop • Multi-core processors o Every core is really an independent processor • Large amounts of RAM • SSDs • GPUs
- 6. © 2014 ANSYS, Inc. July 11, 20146 Typical HPC Growth Path Cluster UsersDesktop User Workstation and/or Server Users
- 7. © 2014 ANSYS, Inc. July 11, 20147 Our HPC Solutions for FEA HPC-enabled solutions: • Structural Mechanics (bundled and solver) products: o Mechanical, Structural, Professional, Multiphysics, Autodyn HPC processing solutions for running: • A simulation – in parallel (simultaneously) – on multiple cores: o ANSYS HPC, ANSYS HPC Pack, ANSYS HPC Workgroup • Parametric simulations on multiple cores simultaneously: o ANSYS HPC Parametric Pack
- 8. © 2014 ANSYS, Inc. July 11, 20148 High-Performance Parallel Computing
- 9. © 2014 ANSYS, Inc. July 11, 20149 Summary Design Impact HPC Parallel Usingtoday’smulticorecomputersiskeyfor companiesto remaincompetitive.ANSYSHPCproductsuiteallowsscalability towhatevercomputationallevelrequired,fromsingle-useror smallusergroupoptionsat entry-leveluptovirtuallyunlimited parallelcapacityorlarge usergroupoptionsat enterpriselevel. • Reduce turnaround time • Examine more design variants faster • Simulate larger or more complex models
- 10. © 2014 ANSYS, Inc. July 11, 201410 Application Example Benefit of HPC Parallel - Examine More Design Variants Objective Develop a steerable conductor for enhanced oil recovery. ANSYS Solution • Stress analysis of hydraulic deflection housing using ANSYS Mechanical. • Use ANSYS HPC technology to solve a typical model with about 750K elements and many contacts in an hour or less, compared to about six hours without parallel processing. Design Impact Parallel processing makes it possible to evaluate five to 10 design iterations per day, enabling Cognity engineers to rapidly improve their design. Images courtesy of Cognity Ltd. Click here to read the full story
- 11. © 2014 ANSYS, Inc. July 11, 201411 Application Example Benefit of HPC Parallel - Simulate Larger or More Complex Models Objective Develop components for aircraft, rocket and gas turbine engines. ANSYS Solution • ANSYS structural mechanics simulations calculate temperature and stress time history of models up to 500,000 nodes. • Key to the selection of ANSYS structural mechanics was the program’s scalability – both in terms of parallel performance and the business model to support the order of magnitude increase in the number of design-point analyses to be performed. Design Impact Because of the increase in model size and number of calculations required, Volvo Aero leverages HPC to help track engine part wear, saving customers service and replacement costs. Images courtesy of Volvo Aero. Click here to read the full story
- 12. © 2014 ANSYS, Inc. July 11, 201412 HPC Parallel Scaling - Faster with More Compute Cores at Your Desktop Number of Cores Number of Cores 10.7 Mio Degrees of Freedom Static, linear, structural 1 load step 1 Mio Degrees of Freedom Harmonic, linear, structural 4 frequencies Intel Xeon E5-2690 processors (2.9 GHz, 16 cores total) with 128 GB of RAM.
- 13. © 2014 ANSYS, Inc. July 11, 201413 1.3x 1.7x 2.7x 2.4x 0 1 2 3 4 5 6 Engine (9 MDOF) Stent (520 KDOF) Clutch (160 KDOF) Bracket (45 KDOF) SpeedupoverR14.5 Improved Scaling at 8 cores by an enhanced domain decomposition method at R15.0 HPC Parallel Scaling - Improved Performance at Higher Core Counts 8-node Linux cluster (with 8 cores and 48 GB of RAM per node, InfiniBand DDR).
- 14. © 2014 ANSYS, Inc. July 11, 201414 1.6x 1.8x 3.8x 4.0x 0 1 2 3 4 5 6 Engine (9 MDOF) Stent (520 KDOF) Clutch (160 KDOF) Bracket (45 KDOF) SpeedupoverR14.5 Improved Scaling at 16 cores by an enhanced domain decomposition method at R15.0 HPC Parallel Scaling - Improved Performance at Higher Core Counts 8-node Linux cluster (with 8 cores and 48 GB of RAM per node, InfiniBand DDR).
- 15. © 2014 ANSYS, Inc. July 11, 201415 1.8x 2.2x 3.9x 5.0x 0 1 2 3 4 5 6 Engine (9 MDOF) Stent (520 KDOF) Clutch (160 KDOF) Bracket (45 KDOF) SpeedupoverR14.5 Improved Scaling at 32 cores by an enhanced domain decomposition method at R15.0 HPC Parallel Scaling - Improved Performance at Higher Core Counts 8-node Linux cluster (with 8 cores and 48 GB of RAM per node, InfiniBand DDR).
- 16. © 2014 ANSYS, Inc. July 11, 201416 2 CPU cores 2 CPU cores + Tesla K20 93 324 3.5X Simulation productivity (with an HPC license) K20 8 CPU cores 7 CPU cores + Tesla K20 275 576 2.1X Simulation productivity (with an HPC Pack) K20 V14sp-5 Model Turbine geometry 2.1 million DOF SOLID187 elements Static, nonlinear analysis One iteration Sparse direct solver Distributed ANSYS Mechanical 15.0 with Intel Xeon E5-2697 v2 2.7 GHz CPU; Tesla K20 GPU with boost clocks. ANSYSMechanicaljobs/day Higher is Better HPC Parallel Scaling - Performance on Newest GPUs at R15.0
- 17. © 2014 ANSYS, Inc. July 11, 201417 Simulation productivity (with a HPC Pack) V14sp-6 Model 4.9 million DOF Static, nonlinear analysis One iteration Sparse direct solver ANSYSMechanicaljobs/day Distributed ANSYS Mechanical 15.0 with Intel Xeon E5-2697 v2 2.7 GHz CPU; Tesla K20 GPU with boost clocks. HPC Parallel Scaling - Performance on Newest GPUs at R15.0 8 CPU cores 7 CPU cores + Tesla K20 180 270 1.5X CPU GPU Benefit 100% 12% 100% 50% Cost CPU-only solution cost is approximated and includes both hardware and software license costs. Benefit is based on the number of completed Mechanical jobs/day. CPU-only solution cost Simulation productivity from CPU-only system Additional productivity from GPU Additional cost of adding a GPU Higher is Better
- 18. © 2014 ANSYS, Inc. July 11, 201418 ANSYS Advantages HPC Parallel - Final Remarks • Good scalability of distributed solvers up to 80 CPU cores • Mainstream solvers can be accelerated by GPUs • License scheme treats each GPU socket in the same manner as a CPU core in all HPC license products at R15.0 – adding lots of value in terms of flexibility and speed • Parallel part-by-part meshing allows simultaneous meshing of multiple parts on multi-core machines – without HPC license! Images courtesy of HP and Aurecon Australia PTY LTD Click here to read about scalability
- 19. © 2014 ANSYS, Inc. July 11, 201419 Watch recorded webinars by clicking below: • Understanding Hardware Selection for ANSYS 15.0 • How to Speed Up ANSYS 15.0 with GPUs • Intel Technologies Enabling Faster, More Effective Simulation • Why HPC for ANSYS Mechanical and CFD Click on webinars related to HPC/IT for more and upcoming ones! Connect with me on LinkedIn by clicking on Wim Slagter Further Information
- 20. © 2014 ANSYS, Inc. July 11, 201420 THANK YOU!