Скачать презентацию OGC and Grid in AU Dr Robert Woodcock Скачать презентацию OGC and Grid in AU Dr Robert Woodcock

6c918fe3b67bf287d4dfc3525527c424.ppt

  • Количество слайдов: 13

OGC and Grid in AU Dr Robert Woodcock Executive Manager, e-Science Stream Leader, Exploration OGC and Grid in AU Dr Robert Woodcock Executive Manager, e-Science Stream Leader, Exploration and Mining www. csiro. au

Application Access and Workflows Applications • Web portals • Desktop applications Pipelines: data / Application Access and Workflows Applications • Web portals • Desktop applications Pipelines: data / tool interoperability • semantics: community agreed standards • structure: GML, Geosci. ML for data interchange • syntax: XML language • service: Wx. S, mediator

Traditional Mechanical Modelling Workflow • Models (mesh + data files) are individually and laboriously Traditional Mechanical Modelling Workflow • Models (mesh + data files) are individually and laboriously constructed • The manual process is error prone • “Powerful” desktop computes several models at a time • Limitations are in the order of ~2 models per week Slide courtesy of Robert Cheung and Warren Potma • Results are manually visualised one at a time • Screenshots are manually taken and made into “movies” • Very little, if any, standardised data archiving is done. This results in potential confusion or loss of the originating conditions of the experiments, making it unrepeatable in the long term

New Refined Workflow Parameterised Geometry Creation • Parameterised template or wizard driven model geometry/mesh New Refined Workflow Parameterised Geometry Creation • Parameterised template or wizard driven model geometry/mesh creation • Boundary condition & model properties parameter sweep utilities Automated generation of visualisations Automated movie generation Automated archiving • automatically creates a “family” of model, data files based on varying a set of parameters • Inversion algorithms • determine input parameters of future iterations automatically based on the user ranking of previous results Multi-site data storage via Storage Resource Broker Slide courtesy of Robert Cheung and Warren Potma 3 D Time varying volume visualisation

The Grid Application… Service Interactions User Workflow. . . Client Login Authentication Community Infrastructure The Grid Application… Service Interactions User Workflow. . . Client Login Authentication Community Infrastructure Edit Problem Description Run Simulation Resource Registry Information Geology W. A Geology S. A Job Monitor Job Management Service Local Repository Archive Search Data Management Service Computation Geochem W. A Geochem N. S. W Physical Resource Fastflo. RT Service Escript Service HPC Repository Physical Resource

Research organisations pressing forward… Minimum accessible through NCRIS 5. 13, 5. 16: Bandwidth > Research organisations pressing forward… Minimum accessible through NCRIS 5. 13, 5. 16: Bandwidth > 1 Gbit/sec, HPC > 800 Gflops, 300 -400 TB online storage Semantic Web Semantic Grid Structured data (NCRIS 5. 13) (Geochem Test Bed) Structure (SEE Grid) Increasing Scale of Interoperability High (Current Web) (APAC Grid Projects) Classical Web Classical Grid Flat files, unstructured data, images Low Increasing scale of data and computation

Au. Scope: Earth Model for the Australian Continent Au. Scope Infrastructure System A toolkit Au. Scope: Earth Model for the Australian Continent Au. Scope Infrastructure System A toolkit for Geoscience Research and Geoscience Applications àminerals, energy, groundwater discovery and management; hazard prediction; environmental monitoring and management àMultiple workflows from observational data to user application àProgressive refinement and enrichment by a community of practice

Computing Services APAC Au. Scope Compute Services IVEC A variety of interfaces to suit Computing Services APAC Au. Scope Compute Services IVEC A variety of interfaces to suit user capabilities Software Finite Element Solvers Reactive Transport Code Fluid Flow Modelling Mechanical Modelling Terawulf TPAC AC 3 ACc. ESS APAC VPAC Hardware CSIRO Data and Knowledge Services Visualisation Tools 3 D Modelling Tools Govt Data & Knowledge Gateways APAC Data & Knowledge Gateways Industry Data & Knowledge Gateways Au. Scope Expert Non-Expert University General School Scientist Student Public Students GA SA WA SAPAC Insurance NT IVEC VIC NSW TAS QPSF AC 3 TPAC VPAC Environment Petroleum Industry Minerals Industry QLD

Au. Scope Modelling Framework Societal Need Modelling Services Natural Hazards Tsunami Vulnerability Modelling Inter Au. Scope Modelling Framework Societal Need Modelling Services Natural Hazards Tsunami Vulnerability Modelling Inter polation Sustainable Energy Storm Mine Waste. Predictive ore Mine Surge Disposal deposition design Inundation Modelling Codes Resources & Mining Mesh Numerical Basis Finite Volumes Base Scientific Concepts Chemical Reactions Reactive Transport Chemistry (Log K) Chemistry (Gibbs) Coupled Mechanics & Reactive Transport Finite Element Solver (Fastflo) Particle Rock Mechanics Earth quake Fluid dynamics Geodynamics Mantle Slab Convection subduction ESys Crustal Geodynamics Modelling code Underworld Surface Scripting environment for Visual- Modeling Finite element simulations isation Package (g. Lucifer) (E-Script) (SPModel) Finite Mantle Element Convection Finite Element Solver Code (Finley) (St. GFEM) (Citcom. S) Finite Elements Particle in Cell Finite Difference Coupled Surface Deformation Processes Fluid Flow

The benefits Benefits for CSIRO and the nation • Creation of new fields of The benefits Benefits for CSIRO and the nation • Creation of new fields of science enabled by the combination of datasets from many research disciplines. • Greater science agility by enabling the creation of larger more diverse science teams across Australia and globally • Greater access to science knowledge by the public y tor e. R osi y ep a w te Ga to ce en Sci • Greater reuse of science data assets • Creation of a data processing and distribution models to support CSIRO’s management of Australia’s national science facilities like ATNF and AAHL

Key Challenges § Data acquisition, management and storage § Large data sets (potentially 20 Key Challenges § Data acquisition, management and storage § Large data sets (potentially 20 petabytes!) § Structured, spatial scientific databases § Access to § facilities – data stores, HPC, sensors, networks § processing services – simulation, modelling, inversion, and data mining tools § Information services – discovery; structured, spatially enabled, scientific databases § applications – web portals, desktop applications § Integration and Collaboration (separation of concerns) Community of practice

What do we need? Unification § OGC and Grid: issues with service chaining and What do we need? Unification § OGC and Grid: issues with service chaining and authentication Adhoc virtual organisations § supporting secure networks over public networks High performance data transfer § appropriate protocols on appropriate networks Stateful and asynchronous versions of OGC specs § especially for large data transfers or computational tasks Toolkits § workflows, User Interfaces…deployment support Management controls and Scheduling

Name Dr Robert Woodcock Title Executive Manager, e-Science Phone +61 8 6436 8780 Email Name Dr Robert Woodcock Title Executive Manager, e-Science Phone +61 8 6436 8780 Email Robert. [email protected] au Web www. csiro. au www. seegrid. csiro. au Thank You Contact CSIRO Phone Email Web 1300 363 400 +61 3 9545 2176 [email protected] au www. csiro. au