Leveraging HPC and Optimizing the IT Environment for ......ANSYS Mechanical ANSYS Fluent 13.0 Dec...
Transcript of Leveraging HPC and Optimizing the IT Environment for ......ANSYS Mechanical ANSYS Fluent 13.0 Dec...
-
© 2012 ANSYS, Inc. February 7, 2013 1
Leveraging HPC and Optimizing the IT Environment for Simulation
Shane Moeykens ANSYS, Inc.
-
© 2012 ANSYS, Inc. February 7, 2013 2
Focused Engineering simulation is all we do. Leading product technologies in all physics areas Largest development team focused on simulation
Capable 2,500 employees 65 locations, 40 countries
Trusted 96 of top 100 FORTUNE 500 industrials ISO 9001 and NQA-1 certified
Proven Recognized as one of the world’s most innovative and fastest-growing companies*
Independent Long-term financial stability CAD agnostic
*BusinessWeek, FORTUNE
ANSYS, Inc.
-
© 2012 ANSYS, Inc. February 7, 2013 3
Trends in Engineering Driving Demand for HPC
Increase product performance and integrity in less time • Consider more design variants
• Find the ‘optimal’ design
• Ensure performance across a range of conditions
Increased product complexity! • Assess larger, more detailed models
• Consider more complex physics
• From single component to system innovations
HPC is a key enabler to gain higher-fidelity insight
HPC is a key enabler to assure greater product integrity through robust design
-
© 2012 ANSYS, Inc. February 7, 2013 4
HPC is a key enabler to amplify engineering productivity
Do more with existing engineering and design teams • Scale-up of HPC to support more
simulation workloads
• Centralized HPC infrastructure (or cloud) with effective remote access
• Managing collaboration and explosive growth of engineering data
Trends in Engineering Driving Demand for HPC
-
© 2012 ANSYS, Inc. February 7, 2013 5
Case Study
Shape Optimization: “50:50:50 Study”
Challenge • Simulate 50 shape variants of the Volvo XC60 with high-fidelity CFD simulations
using a computational mesh of 50 million cells in a total elapsed time of 50 hours (after initial case setup)
Solution • ANSYS Workbench platform
– To drive shape parameters (RBF Morph) – To create DOE and perform Goal Driven Optimization
• ANSYS Fluent and ANSYS HPC • Intel Xeon based cluster
Benefits • Automatic with least human effort • Ability to explore a large design space • Accurate because of high-fidelity CFD • Shorten vehicle development process
48% improvement
-
© 2012 ANSYS, Inc. February 7, 2013 6
Hardware Trends
A Software Development Imperative
Today’s multi-core / many-core hardware evolution makes HPC a software development imperative.
Source: AnandTech
-
© 2012 ANSYS, Inc. February 7, 2013 7
150M cell model
Non-reactive species
LES Turbulence
Running on Cray XE6
Scalable at ~18K cells per core
50
100
150
200
250
300
350
400
2048 3072 4096 5120 6144 7168 8192
Sim
ula
tio
ns
pe
r d
ay
Number of Cores (10 time steps)
Rating
Ideal
Number of Cores
Cells/Core Efficiency
2048 73K 100%
4096 36K 98%
5632 26K 88%
6656 22K 81%
7680 19K 74%
8192 18K 70%
Scaling Performance Milestones
ANSYS Fluent 14.5 on ~8000 Cores
-
© 2012 ANSYS, Inc. February 7, 2013 8
0
500
1000
1500
2000
2500
3000
3500
4000
0 2048 4096 6144 8192 10240 12288
Number of Cores
Fluent 14.5
Fluent 15.0 (devt. version)
Truck (111M cells) benchmark R
atin
g
Scaling Performance Milestones ANSYS Fluent 15.0 on ~12,000 Cores
- Pre-release results
- Scalability at ~10K cells per core
-
© 2012 ANSYS, Inc. February 7, 2013 9
Release
ANSYS Mechanical ANSYS Fluent
13.0 Dec 2010
SMP, Single GPU, Sparse
and PCG/JCG Solvers
14.0 Dec 2011
+ Distributed ANSYS;
+ Multi-node Support
Radiation Heat
Transfer (beta)
14.5 Oct 2012
+ Multi-GPU Support;
+ Hybrid PCG;
+ Kepler GPU Support
+ Radiation HT;
+ GPU AMG Solver
(beta), Single GPU
GPUs and Accelerators
ANSYS and NVIDIA Collaborations
-
© 2012 ANSYS, Inc. February 7, 2013 10
8 8
12
19
0
5
10
15
20
25
AN
SY
S M
echanic
al N
um
ber
of
Jobs
Per
Day
Results from HP Z820; 2 x Xeons
(16 Cores, use of only 8) 128GB
memory, Win7; 2 x Tesla C2075
V14sp-5 Model
Turbine geometry
2,100 K DOF
SOLID187 FEs
Static, nonlinear
One iteration
ANSYS Mechanical14.5
Direct sparse solver
Results for Distributed ANSYS 14.5 Preview and Xeon 8-Core CPUs
Higher is
Better
Xeon E5-2687W 8 Cores + Tesla C2075
ANSYS Mechanical – GPU Speedup
Xeon E5-2687W 8 Cores + 2 x Tesla C2075
2.4x Speedup
with 2 GPUs
http://www.hp.com/go/wsansys
-
© 2012 ANSYS, Inc. February 7, 2013 11
2832
933
517 517
0
1000
2000
3000
Dual Socket CPU
Dual Socket CPU + Tesla C2075
AN
SY
S F
luent
AM
G S
olv
er
Tim
e (
Sec)
2 x Xeon X5650, Only 1 Core Used
1.8x
5.5x
Lower is
Better
2 x Xeon X5650, All 12 Cores Used
Helix geometry
1.2M Hex cells
Unsteady, laminar
Coupled PBNS, DP
AMG F-cycle on CPU
AMG V-cycle on GPU
Helix Model
NOTE: All jobs
solver time only,
~65% of total time
ANSYS Fluent - AMG Solver on GPU(s) Work-in-Progress
Preview of ANSYS Fluent 14.5 Performance
-
© 2012 ANSYS, Inc. February 7, 2013 12
• Rapid scale-up of high performance computing • New/emerging architectures
• More and bigger workloads
• Efficient centralized infrastructure (or cloud) • With effective remote access
• Collaboration hubs with secure and scalable data access • Support for mobile platforms
Deployment Trends
IT Environment for Engineering Simulation
-
© 2012 ANSYS, Inc. February 7, 2013 13
HPC
Local computing infrastructure for simulation is transitioning to centralized HPC resources, shared by a globally distributed workforce
Global Collaboration on Consolidated Infrastructure
Optimizing Remote Simulation Workflow
Users
ANSYS Customer Initiatives • Datacenter Consolidation • Globally Connected R&D • IP Protection & Leverage • Virtual Desktop
Enabling “Simulation Practices” • Work-in-Progress Data Management
• Data remains in the datacenter • Eliminate file transfer bottlenecks
• Mobility and Remote Access • For job management • For interactive simulation process
-
© 2012 ANSYS, Inc. February 7, 2013 14
Current Trends / ANSYS Strategy Spectrum of Private, Hosted and Public Cloud
Hardware: Private cloud or hosted HPC (w/ data security )
Software & User Environment: Seamless usability and remote access
Business: Evolution / Add flexibility of “usage” to traditional license model
Graphic courtesy of IBM
later
-
© 2012 ANSYS, Inc. February 7, 2013 15
Combined usage (on premise + cloud)
Browser access
Cloud Portal Secure account access; data management; job management
Thin client access WIP Data
Graphics servers
Compute cluster Batch jobs
Interactive jobs
ANSYS Cloud – Partner-Enabled Solutions
Enable customers to “Outsource HPC” • Burst or steady-state extension of in-house capacity
Leverage partnerships with HPC Cloud Partners • Consistent vendor-neutral strategy
Cloud License
Partner built, partner owned and operated.
Tested, sold, and supported by ANSYS
On-Premise license
On-Premise Data
Customer Facility
-
© 2012 ANSYS, Inc. February 7, 2013 16
Wrap Up
HPC Focus at ANSYS
• Architecting for extreme scalability – Performance at 10,000’s of cores
– Increased core density and GP-GPUs
– Hybrid distributed/shared memory and vector processing paradigms
– Ongoing load balancing and performance tuning
• Scalability across all components and full simulation process – Meshing, setup, solver, I/O, visualization, optimization…
• Usability and “fit” to the IT environment – Mobile job management
– Integrated job and data management
– Remote display and mobile interactive access
All to achieve next-generation performance and effectiveness!