SlideShare uma empresa Scribd logo
1 de 15
Paving the way to HPC as a Service
Heavy Duty Abaqus Structural Analysis Using HPC in the Cloud –
Challenges, Solutions, Lessons Learned, & Recommendations
Frank Ding
May 16, 2013
What is the UberCloud HPC Experiment?
 Started in mid-2012 as a voluntary effort
 Demonstrate the potential of HPC in the Cloud
 Uncover and overcome the obstacles
 Now with over 450 participants Round 3 is in progress
 Approaching 80 teams as of today
 Please submit your project ideas at the end of this webinar!
2
About the presenter
 Frank Ding, Engineering Analysis & Computing
Manager, Simpson Strong-Tie, HPC Experiment
Advocate,An early adopter of HPC for simulation-driven
design, a huge fan of Linux-based cluster computing
 Thanks to members of the HPC ExperimentTeam #47
 Matt Dunbar, SIMULIA 3DS
 Steve Hebert & Rob Sherrard, Nimbix
 Sharan Kalwani, DataSwing
 Antonio Arena & Cynthia Underwood, NICE Software
 Dennis Nagy, BeyondCAE
3
Simpson Strong-Tie
 A leader in structural systems research, testing and innovation
 Product Lines
 Connectors for light frame construction
 Fasteners and fastening systems
 Lateral systems
 Anchoring systems
 Fiber reinforcing materials
4
Realistic Simulation Requires HPC
 High-fidelity modeling
 Highly nonlinear materials
 Multi-physics
 Fracture and cracking
 Progressive damage and failure
5
Why HPC in the Cloud?
 Current HPC Cluster – 4-node 32 cores total
 Nehalem-based Xeon
 InfiniBand DDR
 HPC in the Cloud
 Lack of in-house HPC resources
 Capacity surge – large jobs speedup
 Capacity overflow jobs – large number of jobs
 Project tested in Round 2 HPC Experiment
 Concrete anchor bolt tension capacity
 1.9 millions of DOF’s
 11.5 hours runtime (32-core)
6
Cloud-Based HPCWorkflow
 ProjectWorkflow
 Pre-processing on the local workstation at the user end
 Abaqus input file loaded to the data staging point thru SFTP
 Abaqus job submitted to the compute cloud thru Nimbix web portal
 Job monitoring thru Nimbix dashboard plus email notification of job status
 Post-processing using remote visualization tool NICE DCV
7
Barriers and Challenges
 Data movement limited by internet bandwidth
 Use remote visualization for post-processing
 End user side internet bandwidth issue
 Team member time schedule (voluntary effort)
 Team member expertise gap
 Meet the project deadline
8
Benefits
 HPC in the Cloud solutions require multi-vendor support
 ISV – 3DS SIMULIA provided Abaqus license
 Cloud infrastructure and service provider – Nimbix HPC Cloud for CAE
 Remote visualization – NICE Desktop CloudVisualization(DCV)
 End-user applications
 HPC Experiment provides a collaboration platform
 Form a team based on end-user application requirement
 Basecamp.com to support team communication
 Third-party solution providers invited to the team if needed
9
Lessons Learned & Recommendations
 End point internet bandwidth variability is the top barrier
 Some workflow details have been identified to improve end user
experience
 Service provider is recommended to provide connection bandwidth
testing tool to the end users
 Remote visualization using NICE DCV is a good platform if a
stable/consistent internet bandwidth is available. Results accessible
anywhere when needed.
 HPC Experiment is a great platform to test your project on the Cloud
HPC
10
Ready for your project!
 Round 3 is in progress and will conclude at the end of June
 Anyone can create a project
 And it is not too late for Round 3 (and beyond)
 Please fill out the form at the end of this webinar
11
Q&A session notes
 What bandwidth is needed in your opinion?
 FD: Roughly a consistent bandwidth over 5Mbps is required
for a smooth response for dynamic 3D model manipulation.
 What is the largest memory node/core on the cloud
forAbaquspurposes?
 FD: I use 4GB/core for my local HPC, but I don’t know Nimbix
Compute Cloud’s configuration, but I would recommend the
same.
Q&A session notes
 Do you use .pbs to submit in batch?
 FD: No, I use Nimbix Compute Cloud job submission web portal,
which talks to theTorque load manager.
 What sort of interconnect types did you use / try?
 FD:We used GigE in the Round 1 and InfiniBand in Round 2.
InfiniBand performs better.
 Can you talk about the Memory required for your project?
 FD:The job was solved by Abaqus/Explicit, which does not have
intensive memory requirement like Abaqus/Standard does. Usually
I go with 4GB/core.
Q&A session notes
 You mentioned that Internet speed (or lack of) was a major
factor.What was the maximum internet speed available to
you for the project?
 FD:The maximum I ever tested was 10 Mbps on a 20 Mbps
pipe.
 Is the NICE visualization using a secure channel (encryption)?
 FD:Yes, encryption using the standard AES algorithm (128 or
256-bit)
ThankYou
http://www.hpcexperiment.com
http://www.cfdexperiment.com
http://www.compbioexperiment.com
http://www.bigdataexperiment.com

Mais conteúdo relacionado

Semelhante a Heavy duty Abaqus structural analysis using HPC in the cloud

Uber cloud at ucc dresden dec 2013
Uber cloud at ucc dresden dec 2013Uber cloud at ucc dresden dec 2013
Uber cloud at ucc dresden dec 2013
Wolfgang Gentzsch
 
Scalable and Distributed DNN Training on Modern HPC Systems
Scalable and Distributed DNN Training on Modern HPC SystemsScalable and Distributed DNN Training on Modern HPC Systems
Scalable and Distributed DNN Training on Modern HPC Systems
inside-BigData.com
 

Semelhante a Heavy duty Abaqus structural analysis using HPC in the cloud (20)

UberCloud - From Project to Product
UberCloud - From Project to ProductUberCloud - From Project to Product
UberCloud - From Project to Product
 
The UberCloud - From Project to Product - From HPC Experiment to HPC Marketpl...
The UberCloud - From Project to Product - From HPC Experiment to HPC Marketpl...The UberCloud - From Project to Product - From HPC Experiment to HPC Marketpl...
The UberCloud - From Project to Product - From HPC Experiment to HPC Marketpl...
 
UberCloud at ucc dresden
UberCloud at ucc dresdenUberCloud at ucc dresden
UberCloud at ucc dresden
 
International Conference on Utility and Cloud Computing December 9 – 12, Dres...
International Conference on Utility and Cloud Computing December 9 – 12, Dres...International Conference on Utility and Cloud Computing December 9 – 12, Dres...
International Conference on Utility and Cloud Computing December 9 – 12, Dres...
 
Uber cloud at ucc dresden dec 2013
Uber cloud at ucc dresden dec 2013Uber cloud at ucc dresden dec 2013
Uber cloud at ucc dresden dec 2013
 
Learn more about the tremendous value Open Data Plane brings to NFV
Learn more about the tremendous value Open Data Plane brings to NFVLearn more about the tremendous value Open Data Plane brings to NFV
Learn more about the tremendous value Open Data Plane brings to NFV
 
Irati goals and achievements - 3rd RINA Workshop
Irati goals and achievements - 3rd RINA WorkshopIrati goals and achievements - 3rd RINA Workshop
Irati goals and achievements - 3rd RINA Workshop
 
How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...
How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...
How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...
 
Scaling Streaming - Concepts, Research, Goals
Scaling Streaming - Concepts, Research, GoalsScaling Streaming - Concepts, Research, Goals
Scaling Streaming - Concepts, Research, Goals
 
Private Network Project for Colleges
Private Network Project for CollegesPrivate Network Project for Colleges
Private Network Project for Colleges
 
UberCloud HPC Experiment Introduction for Beginners
UberCloud HPC Experiment Introduction for BeginnersUberCloud HPC Experiment Introduction for Beginners
UberCloud HPC Experiment Introduction for Beginners
 
Global Media Exchange: Work of the ETC By Christine Thomas of Dolby
Global Media Exchange: Work of the ETC By Christine Thomas of DolbyGlobal Media Exchange: Work of the ETC By Christine Thomas of Dolby
Global Media Exchange: Work of the ETC By Christine Thomas of Dolby
 
OFC 2014 Dinesh Dutt
OFC 2014 Dinesh DuttOFC 2014 Dinesh Dutt
OFC 2014 Dinesh Dutt
 
Busy Polling: Past, Present, Future
Busy Polling: Past,      Present, FutureBusy Polling: Past,      Present, Future
Busy Polling: Past, Present, Future
 
2017 dagstuhl-nfv-rothenberg
2017 dagstuhl-nfv-rothenberg2017 dagstuhl-nfv-rothenberg
2017 dagstuhl-nfv-rothenberg
 
Scalable and Distributed DNN Training on Modern HPC Systems
Scalable and Distributed DNN Training on Modern HPC SystemsScalable and Distributed DNN Training on Modern HPC Systems
Scalable and Distributed DNN Training on Modern HPC Systems
 
NCMS UberCloud Experiment Webinar .
NCMS UberCloud Experiment Webinar .NCMS UberCloud Experiment Webinar .
NCMS UberCloud Experiment Webinar .
 
Accelerating TensorFlow with RDMA for high-performance deep learning
Accelerating TensorFlow with RDMA for high-performance deep learningAccelerating TensorFlow with RDMA for high-performance deep learning
Accelerating TensorFlow with RDMA for high-performance deep learning
 
ICEOTOPE & OCF: Performance for Manufacturing
ICEOTOPE & OCF: Performance for Manufacturing ICEOTOPE & OCF: Performance for Manufacturing
ICEOTOPE & OCF: Performance for Manufacturing
 
Research challenges in Reconfigurable Computing
Research challenges in Reconfigurable ComputingResearch challenges in Reconfigurable Computing
Research challenges in Reconfigurable Computing
 

Último

Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Victor Rentea
 

Último (20)

Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
 
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf
 
Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyArtificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : Uncertainty
 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
MS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsMS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectors
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamDEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
 
ICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesICT role in 21st century education and its challenges
ICT role in 21st century education and its challenges
 
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
 
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
 
Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..
 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
Platformless Horizons for Digital Adaptability
Platformless Horizons for Digital AdaptabilityPlatformless Horizons for Digital Adaptability
Platformless Horizons for Digital Adaptability
 
Six Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal OntologySix Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal Ontology
 
WSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering DevelopersWSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering Developers
 

Heavy duty Abaqus structural analysis using HPC in the cloud

  • 1. Paving the way to HPC as a Service Heavy Duty Abaqus Structural Analysis Using HPC in the Cloud – Challenges, Solutions, Lessons Learned, & Recommendations Frank Ding May 16, 2013
  • 2. What is the UberCloud HPC Experiment?  Started in mid-2012 as a voluntary effort  Demonstrate the potential of HPC in the Cloud  Uncover and overcome the obstacles  Now with over 450 participants Round 3 is in progress  Approaching 80 teams as of today  Please submit your project ideas at the end of this webinar! 2
  • 3. About the presenter  Frank Ding, Engineering Analysis & Computing Manager, Simpson Strong-Tie, HPC Experiment Advocate,An early adopter of HPC for simulation-driven design, a huge fan of Linux-based cluster computing  Thanks to members of the HPC ExperimentTeam #47  Matt Dunbar, SIMULIA 3DS  Steve Hebert & Rob Sherrard, Nimbix  Sharan Kalwani, DataSwing  Antonio Arena & Cynthia Underwood, NICE Software  Dennis Nagy, BeyondCAE 3
  • 4. Simpson Strong-Tie  A leader in structural systems research, testing and innovation  Product Lines  Connectors for light frame construction  Fasteners and fastening systems  Lateral systems  Anchoring systems  Fiber reinforcing materials 4
  • 5. Realistic Simulation Requires HPC  High-fidelity modeling  Highly nonlinear materials  Multi-physics  Fracture and cracking  Progressive damage and failure 5
  • 6. Why HPC in the Cloud?  Current HPC Cluster – 4-node 32 cores total  Nehalem-based Xeon  InfiniBand DDR  HPC in the Cloud  Lack of in-house HPC resources  Capacity surge – large jobs speedup  Capacity overflow jobs – large number of jobs  Project tested in Round 2 HPC Experiment  Concrete anchor bolt tension capacity  1.9 millions of DOF’s  11.5 hours runtime (32-core) 6
  • 7. Cloud-Based HPCWorkflow  ProjectWorkflow  Pre-processing on the local workstation at the user end  Abaqus input file loaded to the data staging point thru SFTP  Abaqus job submitted to the compute cloud thru Nimbix web portal  Job monitoring thru Nimbix dashboard plus email notification of job status  Post-processing using remote visualization tool NICE DCV 7
  • 8. Barriers and Challenges  Data movement limited by internet bandwidth  Use remote visualization for post-processing  End user side internet bandwidth issue  Team member time schedule (voluntary effort)  Team member expertise gap  Meet the project deadline 8
  • 9. Benefits  HPC in the Cloud solutions require multi-vendor support  ISV – 3DS SIMULIA provided Abaqus license  Cloud infrastructure and service provider – Nimbix HPC Cloud for CAE  Remote visualization – NICE Desktop CloudVisualization(DCV)  End-user applications  HPC Experiment provides a collaboration platform  Form a team based on end-user application requirement  Basecamp.com to support team communication  Third-party solution providers invited to the team if needed 9
  • 10. Lessons Learned & Recommendations  End point internet bandwidth variability is the top barrier  Some workflow details have been identified to improve end user experience  Service provider is recommended to provide connection bandwidth testing tool to the end users  Remote visualization using NICE DCV is a good platform if a stable/consistent internet bandwidth is available. Results accessible anywhere when needed.  HPC Experiment is a great platform to test your project on the Cloud HPC 10
  • 11. Ready for your project!  Round 3 is in progress and will conclude at the end of June  Anyone can create a project  And it is not too late for Round 3 (and beyond)  Please fill out the form at the end of this webinar 11
  • 12. Q&A session notes  What bandwidth is needed in your opinion?  FD: Roughly a consistent bandwidth over 5Mbps is required for a smooth response for dynamic 3D model manipulation.  What is the largest memory node/core on the cloud forAbaquspurposes?  FD: I use 4GB/core for my local HPC, but I don’t know Nimbix Compute Cloud’s configuration, but I would recommend the same.
  • 13. Q&A session notes  Do you use .pbs to submit in batch?  FD: No, I use Nimbix Compute Cloud job submission web portal, which talks to theTorque load manager.  What sort of interconnect types did you use / try?  FD:We used GigE in the Round 1 and InfiniBand in Round 2. InfiniBand performs better.  Can you talk about the Memory required for your project?  FD:The job was solved by Abaqus/Explicit, which does not have intensive memory requirement like Abaqus/Standard does. Usually I go with 4GB/core.
  • 14. Q&A session notes  You mentioned that Internet speed (or lack of) was a major factor.What was the maximum internet speed available to you for the project?  FD:The maximum I ever tested was 10 Mbps on a 20 Mbps pipe.  Is the NICE visualization using a secure channel (encryption)?  FD:Yes, encryption using the standard AES algorithm (128 or 256-bit)