We discuss engineering and scientific computing in the Cloud. Users today have three major choices of computing: workstations, servers, and cloud. We compare benefits and challenges of each, and present a solution: the online UberCloud community, experiment, and marketplace for engineers and scientists to discover, try, and buy compute power on demand, in the cloud. Our approach of application containerization and tight software/hardware integration removes many of the known cloud roadblocks.

1. SMART MANUFACTURING:
CAE AS A SERVICE
IN THE CLOUD
Wolfgang Gentzsch
President, The UberCloud
NAFEMS Conference,
Manchester, October 20-22, 2014
Courtesy Time Magazine
What is Smart Manufacturing?

2. Summary
 Workstations / servers / clouds: benefits and challenges
 Exploring cloud challenges: CAE experiments
 Case studies
 Lessons learned
 CAE Cloud Marketplace
 Finally, the ‘secrete’ sauce: application containers
2

3. Wolfgang Gentzsch
President,The UberCloud
Experiment and Marketplace
=>
Product innovation requires computing
Want to build the best bike in the world?
Build it in the computer first!

4. Engineers & scientists computing tools:
workstations, servers, and clouds

5. Stay competitive with computing
 Desktop: 94% of engineers
 Server: 5% of engineers
 Cloud : less than 1%

6. However:
Workstations have limited capacity
 Computing: too slow
 Memory: too small
57 % of users are dissatisfied with their desktop
computing capacity*
* Source: US Council of Competitiveness: http://www.compete.org/

7. Benefits of servers
 More compute power and memory
 Higher quality design and products
 Reducing product failure
 Shorten time to market

8. However:
Servers: expensive and complex
$70,000 server is $1 millionTCO over 3 years

9. Benefits of clouds
 More computing, on demand, pay per use
 Scaling resources up and down
 Low risk with multiple providers
 Result: better, faster, cheaper.

10. However:
Clouds: still many challenges
 Security, licensing, control,
data transfer, expertise, …
 … a crowded cloud market,
difficult to find your ideal
cloud service

11. UberCloud Experiments
 SMBs & research organizations
 to explore the end-to-end process
 of using remote computing resources,
 as a service, on demand, at your finger tip
Since July 2012: 2000+ participants, 155 experiments
Learning how to resolve the roadblocks !

12. Experiment teams
 End-User
 Team Expert
 SoftwareVendor
 Cloud Resource Provider
 Finally, writing the Case Study
 42 case studies in Compendium I & II

13. The UberCloud HPC Experiments
Example: AmazonAWS in the UberCloud:
 Team 2:
 Team 20:
 Team 30:
 Team 40:
 Team 65:
 Team 70:
 Team 116:
 Team 142:
 Team 147:
13
Simulation of a Multi-resonant Antenna System
Turbo-machinery Application Benchmarks
HeatTransfer Use Case
Simulation of Spatial Hearing
Weather Research with WRF
Next Generation Sequencing Data Analysis
Quantitative Finance Historical Data Modeling
VirtualTesting of Severe Service ControlValve
Compressor Map Generation Using Cloud-Based CFD

14. The UberCloud HPC Experiments
Started July 2012, 1500 participants, 72 countries
Example: Bull extreme factory in the UberCloud:
 Team 5:
 Team 8:
 Team 32:
 Team 52:
 Team 85:
 Team 89:
 Team 120:
14
2-phase Flow Simulation of a Separation Column
Flash Dryer with Hot Gas to EvaporateWater from a Solid
2-phase flow simulation of a separation columns
Simulations of Blow-off in Combustion Systems
Combustion simulations of power plant equipment
Simulations of Enzyme-Substrate reactions
Simulation of water flow around self-propelled ship
Courtesy: Marc Levrier, Bull

15. © 2013 ANSYS, Inc. October 23, 201415
Some Lessons Learned
- UberCloud HPC Experiment
Team 8: Flash Dryer Simulation (ANSYS Fluent)
Simulation throughput criterion was met
‼ Remote visualization solution required
‼ Time for downloading results
‼ IP concern
Team 9: Irrigation Simulation (ANSYS CFX)
Timely, high fidelity results were obtained
‼ Windows above Linux preferred
‼ HPC workshop services for SMEs requested
Ability to conduct parametric simulations
‼ Sufficient number of licenses needed
‼ Remote visualization solution required
‼ Disappointing hardware performance results
Team 34: Wind Turbine Simulation (ANSYS Fluent)
Courtesy: Wim Slagter, ANSYS

16. © 2013 ANSYS, Inc. October 23, 201416
Some Lessons Learned
- UberCloud HPC Experiment
Team 36: IC-Engine Simulation (ANSYS Fluent)
Smooth setup of environment and sw
‼ Appropriate cloud licensing required
‼ Network bandwidth not good for graphics
‼ Customized sw needs to be recompiled
Team 54: Pool Plant Simulation (ANSYS CFX)
Ability to easily burst into the Cloud
Accelerated file transfer and 3D graphics
‼ Cost of the commercial CFD licenses
Ease of use
Good remote visualization
‼ File uploading time
‼ Stress test with multiple users required
Team 56: Axial Fan Simulation (ANSYS Fluent)
Courtesy: Wim Slagter, ANSYS

17. TEAM 118: Coupling in-house FE code
with ANSYS Fluent CFD in the Cloud
 End user - Marius Swoboda, Hubert Dengg,
Rolls-Royce Deutschland
 Software Provider:Wim Slagter, René Kapa,
ANSYS
 Cloud Provider: Matthias Reyer, CPU 24/7
 Team Expert: Alexander Heine, CPU 24/7
 Team Mentor: Wolfgang Gentzsch, UberCloud

18. Team 118: Temperature predictions
for jet engine components
CFD Model of a High Pressure Turbine Interstage Cavity
HPT Rotor 1
HPT Rotor 2
Nozzle Guide Vane
Seal Carrier
Nozzle Inlet
(Mass Flow Inlet)
Nozzle Outlet
(Pressure Outlet)
Seal Outlet
(Pressure Outlet)
Annulus Outlet
(Pressure Outlet)
© Rolls-Royce The Jet Engine

19. Team 118: Temperature predictions
for jet engine components
 Jet engine high pressure turbine assembly
 Transient aero-thermal analysis
 FEA/CFD coupling achieved through iterative loop
with exchange of information between the FEA
and CFD at each time step,
 Ensuring consistency of temperature
& heat flux on the coupled interfaces
between metal and fluid domains
Temperature contours for
a Jet Engine Component

20. Team 118:
The aim of this experiment
 To couple ANSYS Fluent with in-house FE code.
 Done by extracting heat flux profiles from Fluent
model and applying FE model. FE model provides
metal temperatures in the solid domain.
 Conjugate heat transfer needs a lot of computing,
especially when 3D-CFD-models with
more than 10 mio cells are required.
 Using cloud resources is beneficial
regarding computing time. Contours of heat flux

21. Team 118:
Benefits of CAE in the Cloud
 Keep on using your workstation for daily design
while using Cloud resources for bigger jobs
 An HPC system at your finger tip, on demand
 Pay per use (cost savings by reducing CAPEX)
 Scaling resources up and down (business flexibility)
 Low risk by working with multiple providers.
 Maintaining control: Cloud provider was around
the corner

22. The ProblemToday:
Crowded and ineffective cloud ‘market’
Supply
Cloud providers
ISVs
Consultants
Demand
Engineers
Scientists
Data analysts
.
.
.
.
.
Complexity
Data
Transfer
SecurityLicensing
Uncertain
Cost
Roadblocks

23. Solution:
The UberCloud Marketplace
Supply
Cloud providers
ISVs
Consultants
…
Demand
Engineers
Scientists
Data analysts
UberCloud
Marketplace

24. UberCloud marketplace, sample

25. Builder
Launcher
Controller
ISV DataTools
Stackable units with tools (ex: encryption), ISV application codes (ex: OpenFOAM).
Just add your own codes and data.
Run anywhere with UberCloud Run Time.
Scale up or down the compute power as needed.
Collect granular usage data, logs.
Monitor, alert, report.
Any
Workstation
Any Cluster Any Cloud
Run Time Run Time Run Time
Containers: Build once, run anywhere

26. Builder
Launcher
Controller
ISV DataTools
Stackable units with tools (ex: encryption), ISV application codes (ex: OpenFOAM).
Just add your own codes and data.
Run anywhere with UberCloud Run Time.
Scale up or down the compute power as needed.
Collect granular usage data, logs.
Monitor, alert, report.
Any
Workstation
Any Cluster Any Cloud
Run Time Run Time Run Time
Containers: Build once, run anywhere

27. Builder
Launcher
Controller
ISV DataTools
Stackable units with tools (ex: encryption), ISV application codes (ex: OpenFOAM).
Just add your own codes and data.
Run anywhere with UberCloud Run Time.
Scale up or down the compute power as needed.
Collect granular usage data, logs.
Monitor, alert, report.
Any
Workstation
Any Cluster Any Cloud
Run Time Run Time Run Time
Containers: Build once, run anywhere

28. Builder
Launcher
Controller
ISV DataTools
Stackable units with tools (ex: encryption), ISV application codes (ex: OpenFOAM).
Just add your own codes and data.
Run anywhere with UberCloud Run Time.
Scale up or down the compute power as needed.
Collect granular usage data, logs.
Monitor, alert, report.
Any
Workstation
Any Cluster Any Cloud
Run Time Run Time Run Time
Containers: Build once, run anywhere

29. Containers:
Reducing / Removing Cloud Challenges
CAE Cloud Challenges UberCloud *)
Security 
Portability 
Compliance 
DataTransfer 
Standardization 
Software Licenses 
Resource Availability 
Transparency of Market 
Cost & ROITransparency 
No Cloud Expertise Needed 
*)When UberCloud is fully developed one year from now

30. It’s your turn now 
 Download 2013 Compendium of case studies
Download 2014 Compendium of case studies
 Register at TheUberCloud.com
 Register for The UberCloudVoice newsletter
 Check The UberCloud Marketplace

31. www.nafems.org
Thank You !
Please register at
TT
http://www.TheUberCloud.com