cox by lanyuehua

VIEWS: 11 PAGES: 26

									    Grid Enabled Optimisation and
    Design Search for Engineering
             (GEODISE)

               Prof Simon Cox
            Southampton University

 3rd Annual Workshop on Linux Clusters for Super Computing
National Supercomputer Centre, Linköpings Universitet, Sweden

          http://www.geodise.org
Grid Enabled Optimisation and Design
 Search for Engineering (GEODISE)
                   Southampton, Oxford and Manchester
Simon Cox- Grid/ W3C Technologies         BAE Systems- Engineering
   and High Performance Computing
   Global Grid Forum Apps Working Group   Rolls-Royce- Engineering
Andy Keane- Director of Rolls Royce/      Fluent- Computational Fluid Dynamics
   BAE Systems University Technology
   Partnership in Design Search and       Microsoft- Software/ Web Services
   Optimisation
                                          Intel- Hardware
Mike Giles- Director of Rolls Royce
   University Technology Centre for       Compusys- Systems Integration
   Computational Fluid Dynamics
Carole Goble- Ontologies and DARPA        Epistemics- Knowledge Technologies
   Agent Markup Language (DAML) /         Condor- Grid Middleware
   Ontology Inference Language (OIL)
Nigel Shadbolt- Director of Advanced
   Knowledge Technologies (AKT) IRC


                      http://www.geodise.org
           The GEODISE Team ...
   Richard Boardman       Andy Keane
   Sergio Campobasso      Marc Molinari
   Liming Chen            Juri Papay
   Mike Chrystall         Graeme Pound
   Simon Cox              Nicola Reader
   Mihai Duta             Angus Roberts
   Clive Emberey          Mark Scott
   Hakki Eres             Tony Scurr
   Matt Fairman           Nigel Shadbolt
   Carole Goble           Paul Smart
   Mike Giles             Barry Tao
   Ian Hartney            Jasmin Wason
   Tracey Hunt            Fenglian Xu
   Zhuoan Jiao            Gang “Luke” Xue
Design
Structural Optimisation
                   Design Challenges
    Modern engineering firms are global and distributed
How to … ?

… improve design environments             CAD and analysis tools, user
… cope with legacy code / systems      interfaces, PSEs, and Visualization

… produce optimized designs                  Optimisation methods

… integrate large-scale systems in a   Management of distributed compute
  flexible way                               and data resources

                                       Data archives (e.g. design/ system
… archive and re-use design history
                                                    usage)

                                           Knowledge repositories &
… capture and re-use knowledge
                                       knowledge capture and reuse tools.

                “Not just a problem of using HPC”
                  The Grid Problem
“Flexible and secure sharing of resources among
  dynamic collections of individuals within and
              across organisations”
   Resources = assets, capabilities, and knowledge
       Capabilities (e.g. application codes, analysis tools)
       Compute Grids (PC cycles, commodity clusters, HPC)
       Data Grids
       Experimental Instruments
       Knowledge Services
       Virtual Organisations
       Utility Services
     Grid middleware mediates between these resources
  GEODISE                                                      Engineer
                                                                            Reliability
                                                                            Security
                                                                              QoS
                                                           GEODISE
                                                                                               Visualization
                                                           PORTAL
                      Knowledge
                      repository       Session
  Ontology for                        database
  Engineering,                   Traceability
Computation, &
Optimisation and                                          OPTIMISATION
 Design Search

                                                     OPTIONS                                Globus, Condor, OGSA
                                                      System
                                                                 Optimisation
                                                                   archive
                          APPLICATION
                            SERVICE                                                   COMPUTATION
                           PROVIDER               Licenses
   Intelligent                                    and code                                                       Intelligent
  Application                                                                                                    Resource
   Manager         CAD System             Analysis                                    Parallel machines           Provider
                    CADDS                  CFD                                            Clusters
                     IDEAS                 FEM                                  Internet Resource Providers
                      ProE                 CEM                                           Pay-per-use
                   CATIA, ICAD
                                                        Geodise will provide grid-based seamless access to an intelligent knowledge
                                            Design        repository, a state-of-the-art collection of optimisation and search tools,
                                            archive    industrial strength analysis codes, and distributed computing & data resources
                  GEODISE Demo
  (1) Security Infrastructure
Authentication & Authorisation




   (2) Define Geometry to                Nacelle Design
           optimise

                                                   Axisymmetric
                                   3D                  (2D)




(3) Sample Objective function
  to build Response Surface
             Model

                        Grid Computing
                GEODISE Demo

(4) Optimise over Response
       Surface Model




(5) Grid Database Query and
  Postprocessing of Results
Coming next…




    Grid Computing using (.NET)
           web services



                  http://www.geodise.org
Knowledge Portal
Knowledge Acquisition
Coming next…


       Knowledge Technologies
   a) Annotating Design Optimisation Workflows
             using OntoMat-Annotizer
               b) Knowledge Portal



                         http://www.geodise.org
            Workflow Requirements
   Flexibility
      Customise the workflow and its components initially
      Compose a work flow via drag & drop activity node component into
       an editor panel
      Link to knowledge and other services
   Monitoring
      Interact with the workflow or its components during simulation
      Job status
      Resource usage
   Maintainable
      Modify & re-use the workflow either in a GUI or in a human readable
       file
   Usability
      Easy to use by engineering users
Workflow Tools: IBM
Workflow Tools: Microsoft
Grid Enabled Optimisation and
Design Search for Engineering
         (GEODISE)

       Prof Simon Cox
    Southampton University
   http://www.geodise.org
            Workflow Requirements
   Flexibility
      Customise the workflow and its components initially
      Compose a work flow via drag & drop activity node component into
       an editor panel
      Link to knowledge and other services
   Monitoring
      Interact with the workflow or its components during simulation
      Job status
      Resource usage
   Maintainable
      Modify the workflow either in a GUI or in a human readable file
   Usability
      Easy to use by engineering users
                             Matlab
   Matlab and toolboxes for mathematical computation,
    analysis, visualization, and algorithm development:
“MATLAB is an intuitive language and a technical computing environment.
     It provides core mathematics and advanced graphical tools for data
  analysis, visualization, and algorithm and application development. With
     more than 600 mathematical, statistical, and engineering functions,
    engineers and scientists rely on the MATLAB environment for their
                         technical computing needs.”
                                                    (www.mathworks.com)
   Industry usage
      Aerospace
      Defense
      Automotive
      Biotech
      Pharmaceutical
      Medical                          CROSS PLATFORM/ OS
                          Matlab
   Flexibility
     Integrated Development Environment
     Provides full GUI Support and can integrate code
      from Fortran, C, Java, & Web/ Grid Services
   Monitoring
       Command line/ batch mode
   Maintainable
       Matlab Scripts readable and editable by Engineers
   Usability
       Already in widespread use by Engineers
        Grid Computing from Matlab
   Globus commands (based on Java CoG v0.9.13)
     gd_CreateProxyCertVisual
     gd_BatchSubmit
     gd_JobStatus
     gd_GetFile (GSIftp v0.9.14)
     gd_PutFile (GSIftp v0.9.14)
   Database commands
     gd_archive
     gd_query
     gd_retrieve
Coming next…



   Grid-enabled Mesh Generation
      and Computational Fluid
             Dynamics


                 http://www.geodise.org
Coming next…




    Database Archive, Query and
             Retrieve



                  http://www.geodise.org
    The future of design optimisation
Design Optimisation needs integrated services
 Design improvements driven by CAD tools coupled to
   advanced analysis codes (CFD, FEA, CEM etc.)
 On demand heterogeneous distributed computing and
   data spread across companies and time zones.
 Optimization “for the masses” alongside manual search
   as part of a problem solving environment.
 Knowledge based tools for advice and control of process
   as well as product.
   Geodise will provide grid-based seamless access to an
        intelligent knowledge repository, a state-of-the-art
      collection of optimisation and search tools, industrial
     strength analysis codes, and distributed computing and
                          data resources
                  Conclusions
   Grid is not just about computing
   Grid must be provided to users in a familiar
    environment
   Services need semantics
   Workflow is an important issue

   … Agnostic about operating system

								
To top