The UberCloud
Paving the way to
High Performance Computing as a Service

International Conference on Utility and Cloud Computing
December 9 – 12, Dresden, Germany
Wolfgang Gentzsch and Burak Yenier
What is High Performance Computing ?
Modeling the World on the computer …

t   U      U  U   0

F  ma

i t   H 

E  mc 2

… in order to allow for Virtual Product Development …
2
What is High Performance Computing ?
… or for scientific insights
Combustion

Environment

Cosmology
Climate

3
HPC is needed to stay competitive
The digital manufacturing engineer has
several options to use High Performance
Computing (HPC):
 HPC on the Desktop: over 90% of engineers
 HPC on the Server: about 5% of engineers
 HPC as a Service: in the Cloud; less than 1%
Workstations: limited capacity
Low-end workstations and PCs are important for
daily design and development work, but
 50+ % of users are dissatisfied with their computing

capacity*
 Too slow, e.g. jobs run over night or a whole week
 Too small, detailed geometry and physics don’t fit into
memory
 Number of jobs are limited which affects the quality of
the final result
*

Source: http://www.compete.org/
Servers: expensive and complex
 For SMEs buying and using large scale HPC systems is

expensive and complex …
HPC as a Service: benefits & challenges
HPC as a Service (in the Cloud) offers flexibility, business
agility, scaling up and down, pay-per-use, OPEX instead
CAPEX, but
 It’s a new business and working paradigm
 Security, privacy, trust in service provider
 Intellectual property
 Software Licensing
 Heavy data transfers
UberCloud Poll about Roadblocks

The UberCloud, June 13, 2013
The UberCloud HPC Experiment
An open voluntary collaborative community
Objective:
Making HPC as a Service available, for everybody, on demand
How?
For SMEs and their engineering applications
to explore the end-to-end process
of using remote computing resources,
as a service, on demand, at your finger tip,

and learning how to resolve the roadblocks.
The End-User’s Benefits
 Free HPC Experiment, on-demand access to hardware, software,

and expertise, with a one stop resource access experience
 No hunting for resources in the complex emerging cloud market
 Professional match-making of end-users with service providers
 Perfectly tuned end-to-end, step-by-step process to HPC Cloud
 Lowering barriers & risks for frictionless entry into HPC Cloud
 Crowd sourcing: End-Users are building relationships with other

community members who actively contribute to improvements
How does the Experiment work?
 End-User joins the experiment

 Software Vendor joins
 We select a Team Expert
 We suggest a Resource Provider

 Team is ready to go
 … 22 steps on Basecamp’s virtual team office
 Finally, writing the Case Study
Where are we with the experiment
 Started August 2012: today 700+ participating organizations

and individuals
 Participants are from 66 countries
 Round 4 started September 1: already 42 new teams

 124 teams have been formed in Rounds 1-4
 Registration at:

www.hpcexperiment.com
www.cfdexperiment.com
www.compbioexperiment.com
www.bigdataexperiment.com
UberCloud community website
 With social network and crowdsourcing features
 Forums, Q&A, discussions, newsletters,…
 Feature stories: HPCwire, Desktop Engineering, Bio-IT…
 UberCloud University with free and paid lectures
 UberCloud HPC Experiment free trial service
 UberCloud Exhibit services directory
 List of currently ongoing team projects, their status, and

organizations involved
 List of upcoming conferences with ‘meet me there’ button
Finally: The UberCloud Marketplace
 Crowdsourcing and social networking: provide the

community web platform to discover and try HPCaaS,
address pain points, facilitate adoption, harness collective
intelligence
 Marketplace for computing-related services: 20+ million

engineers, scientists, and their service providers to list,
discover, try, and sell/purchase
 IaaS: HPC centers and public cloud providers
 SaaS: open source and commercial technical software
 Expertise: specialized HPC, software, technical know-how
Cloud Computing Reality Check

 UberCloud Compendium

sponsored by Intel

 25 selected use-cases from

60 teams in Rounds 1 & 2
 Google:
“ubercloud compendium”

17
Team 2: Simulating new probe design
for a medical device
Front End + 2 GPU Solvers In Action
Team 2: Team members
HPC Expert:

Electromagnetic Simulation

End User: Anonymous
Team 8: Multiphase flows within the
cement and mineral industry
Drying moist particles with hot gas
Team 8: Team members
Ingo Seipp
and Team

End-User

CFX
Team 26: Simulating Stent
Deployment
Using SIMULIA’s Abaqus/Standard and Remote Viz
Software from NICE to run CAE on SGI Cyclone™
 Assessment of a fictitious balloon expandable stent design
 deployment, physiological pulsatile loading
 and both axial and radial compression.
Team 26- Team members
 End User: Anonymous: Global Designer and Manufacturer of Sterile

Medical Products
 CAE Software Provider: Matt Dunbar, Chief Architect, SIMULIA
 Remote Viz Software: NICE Desktop Cloud Visualization (DCV)
 HPC/CAE Expert: Scott Shaw, Senior Applications Engineer, SGI
 Resource Provider: SGI Cyclone. Tony DeVarco, Senior Manager for

Strategic Partners and Cloud Computing at SGI. Eugene Kremenetsky,
Systems Engineering Technical Lead at SGI
 Team Mentor: Gregory Shirin from the HPC Experiment team
Challenges and Solutions
 Information security, privacy: protecting the users intellectual

property, guarding raw data, processing models, resulting
information: document security requirements, and select the
right provider
 Internet too slow for heavy data transfer: don’t ship every
result, just the important ones; use remote visualization; if
necessary, fedex the data over-night via USB hard drive
 Incompatible software licensing models: ISVs have to develop
compatible on-demand software licensing models
Challenges and Solutions, cont’d
 Reliability & availability of resource providers: seeking info on

reliability and availability of each vendor before partnering
 Lack of easy registration and administration: originally HPC
resources are not designed for the masses: use automated rules
based instant decision making capabilities
 Costs: pay-per-use billing can result in unpredictable cost,
project can easily run out of budget: use automated, policy
driven monitoring of usage and billing
 High expectations, disappointing results: we are in transition
to the cloud computing paradigm: Set goals that are
incrementally better than your current capabilities.
Why join?
 HPC as a Service is the next big thing, benefits are obvious

 HPC is complex; together it is easier to tackle the complexity
 Low entry barrier to HPC as a Service through an experiment
 Learning by doing; experimenting with no risk; no failure
 Becoming an active part of this growing community
 Exploring the end-to-end process and learning how this fits

into your research and/or business direction in the near future
Thank You
http://www.hpcexperiment.com
http://www.cfdexperiment.com
http://www.compbioexperiment.com
http://www.bigdataexperiment.com

Uber cloud at ucc dresden dec 2013

  • 1.
    The UberCloud Paving theway to High Performance Computing as a Service International Conference on Utility and Cloud Computing December 9 – 12, Dresden, Germany Wolfgang Gentzsch and Burak Yenier
  • 2.
    What is HighPerformance Computing ? Modeling the World on the computer … t   U      U  U   0 F  ma i t   H  E  mc 2 … in order to allow for Virtual Product Development … 2
  • 3.
    What is HighPerformance Computing ? … or for scientific insights Combustion Environment Cosmology Climate 3
  • 4.
    HPC is neededto stay competitive The digital manufacturing engineer has several options to use High Performance Computing (HPC):  HPC on the Desktop: over 90% of engineers  HPC on the Server: about 5% of engineers  HPC as a Service: in the Cloud; less than 1%
  • 5.
    Workstations: limited capacity Low-endworkstations and PCs are important for daily design and development work, but  50+ % of users are dissatisfied with their computing capacity*  Too slow, e.g. jobs run over night or a whole week  Too small, detailed geometry and physics don’t fit into memory  Number of jobs are limited which affects the quality of the final result * Source: http://www.compete.org/
  • 6.
    Servers: expensive andcomplex  For SMEs buying and using large scale HPC systems is expensive and complex …
  • 7.
    HPC as aService: benefits & challenges HPC as a Service (in the Cloud) offers flexibility, business agility, scaling up and down, pay-per-use, OPEX instead CAPEX, but  It’s a new business and working paradigm  Security, privacy, trust in service provider  Intellectual property  Software Licensing  Heavy data transfers
  • 8.
    UberCloud Poll aboutRoadblocks The UberCloud, June 13, 2013
  • 9.
    The UberCloud HPCExperiment An open voluntary collaborative community Objective: Making HPC as a Service available, for everybody, on demand How? For SMEs and their engineering applications to explore the end-to-end process of using remote computing resources, as a service, on demand, at your finger tip, and learning how to resolve the roadblocks.
  • 10.
    The End-User’s Benefits Free HPC Experiment, on-demand access to hardware, software, and expertise, with a one stop resource access experience  No hunting for resources in the complex emerging cloud market  Professional match-making of end-users with service providers  Perfectly tuned end-to-end, step-by-step process to HPC Cloud  Lowering barriers & risks for frictionless entry into HPC Cloud  Crowd sourcing: End-Users are building relationships with other community members who actively contribute to improvements
  • 11.
    How does theExperiment work?  End-User joins the experiment  Software Vendor joins  We select a Team Expert  We suggest a Resource Provider  Team is ready to go  … 22 steps on Basecamp’s virtual team office  Finally, writing the Case Study
  • 12.
    Where are wewith the experiment  Started August 2012: today 700+ participating organizations and individuals  Participants are from 66 countries  Round 4 started September 1: already 42 new teams  124 teams have been formed in Rounds 1-4  Registration at: www.hpcexperiment.com www.cfdexperiment.com www.compbioexperiment.com www.bigdataexperiment.com
  • 13.
    UberCloud community website With social network and crowdsourcing features  Forums, Q&A, discussions, newsletters,…  Feature stories: HPCwire, Desktop Engineering, Bio-IT…  UberCloud University with free and paid lectures  UberCloud HPC Experiment free trial service  UberCloud Exhibit services directory  List of currently ongoing team projects, their status, and organizations involved  List of upcoming conferences with ‘meet me there’ button
  • 16.
    Finally: The UberCloudMarketplace  Crowdsourcing and social networking: provide the community web platform to discover and try HPCaaS, address pain points, facilitate adoption, harness collective intelligence  Marketplace for computing-related services: 20+ million engineers, scientists, and their service providers to list, discover, try, and sell/purchase  IaaS: HPC centers and public cloud providers  SaaS: open source and commercial technical software  Expertise: specialized HPC, software, technical know-how
  • 17.
    Cloud Computing RealityCheck  UberCloud Compendium sponsored by Intel  25 selected use-cases from 60 teams in Rounds 1 & 2  Google: “ubercloud compendium” 17
  • 18.
    Team 2: Simulatingnew probe design for a medical device Front End + 2 GPU Solvers In Action
  • 19.
    Team 2: Teammembers HPC Expert: Electromagnetic Simulation End User: Anonymous
  • 20.
    Team 8: Multiphaseflows within the cement and mineral industry Drying moist particles with hot gas
  • 21.
    Team 8: Teammembers Ingo Seipp and Team End-User CFX
  • 22.
    Team 26: SimulatingStent Deployment Using SIMULIA’s Abaqus/Standard and Remote Viz Software from NICE to run CAE on SGI Cyclone™  Assessment of a fictitious balloon expandable stent design  deployment, physiological pulsatile loading  and both axial and radial compression.
  • 23.
    Team 26- Teammembers  End User: Anonymous: Global Designer and Manufacturer of Sterile Medical Products  CAE Software Provider: Matt Dunbar, Chief Architect, SIMULIA  Remote Viz Software: NICE Desktop Cloud Visualization (DCV)  HPC/CAE Expert: Scott Shaw, Senior Applications Engineer, SGI  Resource Provider: SGI Cyclone. Tony DeVarco, Senior Manager for Strategic Partners and Cloud Computing at SGI. Eugene Kremenetsky, Systems Engineering Technical Lead at SGI  Team Mentor: Gregory Shirin from the HPC Experiment team
  • 24.
    Challenges and Solutions Information security, privacy: protecting the users intellectual property, guarding raw data, processing models, resulting information: document security requirements, and select the right provider  Internet too slow for heavy data transfer: don’t ship every result, just the important ones; use remote visualization; if necessary, fedex the data over-night via USB hard drive  Incompatible software licensing models: ISVs have to develop compatible on-demand software licensing models
  • 25.
    Challenges and Solutions,cont’d  Reliability & availability of resource providers: seeking info on reliability and availability of each vendor before partnering  Lack of easy registration and administration: originally HPC resources are not designed for the masses: use automated rules based instant decision making capabilities  Costs: pay-per-use billing can result in unpredictable cost, project can easily run out of budget: use automated, policy driven monitoring of usage and billing  High expectations, disappointing results: we are in transition to the cloud computing paradigm: Set goals that are incrementally better than your current capabilities.
  • 26.
    Why join?  HPCas a Service is the next big thing, benefits are obvious  HPC is complex; together it is easier to tackle the complexity  Low entry barrier to HPC as a Service through an experiment  Learning by doing; experimenting with no risk; no failure  Becoming an active part of this growing community  Exploring the end-to-end process and learning how this fits into your research and/or business direction in the near future
  • 27.