case study
The Power of Software Optimization
ANSYS Scales
Simulation Performance
by optimizing ANSYS® Mechanical™ for the increasing parallelism
of Intel® architecture
The Need for Speed in Simulation-Based Design
Engineering simulation software has changed how companies design products, enabling them to explore and test more design
options faster, while reducing the need for ...physical prototyping. ANSYS software has played a central role in this transition
1
and is now used by 96 of the top 100 industrial companies on the FORTUNE 500* list.
ANSYS customers have an insatiable need for higher computing performance, so they can model and test bigger assemblies
with greater fidelity. Many of them also want to generate deeper insights with each simulation, by integrating additional
physics variables, exploring nonlinear and composite materials, and evaluating more complex and dynamic environmental
conditions. Yet, fast time to results remains a critical factor for most customers to meet aggressive time-to-market
requirements. As the size and complexity of the simulations grows, the speed and capacity of the computing platform must
also grow.
Up to 3X the performance for ANSYS ™Mechanical
Gold Upgrade: Move to a New System and Add Intel® Xeon® Phi Coprocessors
Intel® Many Integrated Core Ansys® Mechanical 16.0 V16sp–5
R16.0 Benchmark SP5
Architecture Offers a Path 600
To meet the continually growing Intel® Xeon® processor E5-2690
550 Intel® Xeon® processor E5-2697 v3
compute demand, ANSYS has worked 500 Intel® Xeon® processor E5-2697 v3 with
closely with Intel for several years to Intel® Xeon Phi® coprocessor 3120
450
Rating Intel® Xeon® processor E5-2697 v3 with
optimize ANSYS Mechanical software Intel® Xeon Phi® coprocessor 7120
400 +2.2x
for increasing parallelism in each new Solver 350
generation of multicore Intel® Xeon® 300
Core
processors. More recently, ANSYS +2.8x
250
and Intel have extended these efforts 200
to the many-core Intel® Xeon Phi™ +3.2x
150
coprocessor. These highly parallel 100
coprocessors provide up to 61 cores, 50
244 threads, and 1.2 teraflops of
double-precision peak performance 0 3120 7120 7120 7120 7120 7120
3120 3120 3120 3120 3120 7120 7120 7120 7120 7120 7120 7120 7120
Phi Phi Phi
Phi Phi Phi Phi Phi
2
per coprocessor. They can run the 0 0 0 Phi Phi Phi Phi Phi Phi
0 0 0 0 0 Phi Phi Phi Phi Phi Phi Phi Phi Phi Phi Phi Phi Phi
core core core
core core core core core 4 2 2 3 3 3
1 1 1 2 2 1 1 1 1 1 1 1 2
same code as Intel Xeon processors, so core core core core core core
core core core core core core core core core core core core core
16 20 24
1 2 3 4 8
independent software vendors (ISVs), 12 14 16 17 20 21
1 2 3 4 6 1 2 3 4 5 7 8 6
such as ANSYS, are not required to By optimizing ANSYS® Mechanical™ 16.0 for the latest multicore Intel® Xeon® processors
write and manage multiple code bases. and many-core Intel® Xeon Phi™ coprocessors, ANSYS
Read the full case study.