Docstoc

Semantic grid From Concepts to Implementation

Document Sample
Semantic grid From Concepts to Implementation Powered By Docstoc
					       Semantic grid
From Concepts to Implementation



               Nguyen Thanh Vu
               Hoang Song Cam Thach
               Cu Nguyen Phuong Ha
    Outline

       Introduction
       Semantic Web
       S-OGSA
       Implementation ( e-Science & myGrid )




2
    What is the Semantic Gird?

       An extension of the current Grid in which information
        and services are given well-defined and explicitly
        represented meaning, so that it can be shared and
        used by humans and machines, better enabling them
        to work in cooperation.




3
    Why we need the Semantic Grid?

     “It is a truth universally acknowledged, that
     an application in possession of good
Grid middleware, must be in want of meaningful
     metadata.”
                                 -- prof. C. Goble

         Semantic


4
    Why we need the Semantic Grid?

       Example: To illustrate, consider if a machine‟s operating
        system is described as “SunOS” or “Linux.” To query for a
        machine that is “Unix” compatible, a user either has to:

    1. Explicitly incorporate the Unix compatibility concept into the
      request requirements by requesting a disjunction of all Unix-
      variant operating systems, e.g., (OpSys=“SunOS” ||
      OpSys=“Linux”), or
    2. Wait for all interesting resources to advertise their operating
      system as Unix as well as either Linux or SunOS, e.g.,
      (OpSys=“SunOS,” “Unix”), and then express a match as set-
      membership of the desired Unix value in the OpSys value set,
      e.g., hasMember(OpSys, “Unix”).


5
    Why we need the Semantic Grid?

       Example (cont)
         Apply Semantics…
        - Knowledge base: “SunOS and Linux are types of Unix
         operating system”
        - Request: “Need the Unix compatibility OS”




6
    Semantic Web
        Current Web ( WWW )
        - Is a huge library of interlinked documents that are transferred
         by computers and presented to people.
        - Anyone can contribute to it.
        - Quality of information or even the persistence of documents
         cannot be generally guaranteed.
        - Contains a lot of information and knowledge, but machines
         usually serve only to deliver and present the content of
         documents describing the knowledge.
        - People have to connect all the sources of relevant information
         and interpret them themselves.
                        Machine can Process the content
                                    But
                      Machine can’t Understand content
7
    Semantic Web

       Definition
         The Semantic Web is an extension of the current
        web in which the semantics of information and
        services on the web is defined, making it possible for
        the web to understand and satisfy the requests of
        people and machines to use the web content.
                                    --- Tim Berners-Lee




8
    Semantic Web

       Definition ( cont )
        Semantic web is an effort to enhance current web
        so that computers can process the information
        presented on WWW, interpret and connect it, to help
        humans to find required knowledge




9
     Semantic Web

        Semantic Web is a project that should
         provide a common framework that allows
         data to be shared and reused across
         application, enterprise, and community
         boundaries.
        Is led by World Wide Web Consortium
         (W3C).


10
     Semantic Web Architecture (1)

                      URI (Uniform Resource
                       Identifier) is a string of a
                       standardized form that
                       allows to uniquely identify
                       resources.
                      Unicode is a standard of
                       encoding international
                       character sets and it allows
                       that all human languages
                       can be used (written and
                       read) on the web using one
                       standardized form.


11
     Semantic Web Architecture (2)

                         XML ( Extensible
                          Markup Language)
                          layer makes sure
                          that there is a
                          common syntax
                          used in the semantic
                          web.


12
     Semantic Web Architecture (3)

                         RDF stands for Resource
                          Description Framework.
                         RDF is a graphical formalism
                          ( + XML syntax + semantics)
                           – for representing
                              metadata
                           – for describing the
                              semantics of information
                              in a machine- accessible
                              way
                         Provides a simple data
                          model based on triples
                          subject-predicate-object
13
     RDF Data model

        Statements are <subject, predicate, object> triples:
          –   <Joe, hasFamilyName,Smith >
        Can be represented as a graph.
        Statements describe properties of resources
        A resource is any object that can be pointed to by a URI:
        Properties themselves are also resources (URIs)




14
     RDF Syntax

        RDF has an XML syntax that has a specific meaning:
         - Every Description element describes a resource
         - Every attribute or nested element inside a Description is a
         property of that Resource
         - We can refer to resources by URIs




15
     RDF – Example

     English Statement:
     http://www.example.org/index.html has a creation-date whose value is August 16, 1999

     Triple representation:
     ex:index.html       exterms:creation-date               "August 16, 1999"

     RDF Graph representation:




16
     RDF – Example (cont)

     RDF/XML syntax:




17
     Semantic Web Architecture (4)

                         RDFS (RDF Schema)
                          is extending RDF
                          vocabulary to allow
                          describing taxonomies
                          of classes and
                          properties.



18
     RDFS ( cont…)

        RDF does not give any special meaning to vocabulary such as
         subClassOf or type (supporting OO-style modelling).
        RDF Schema extends RDF with a schema vocabulary that allows
         you to define basic vocabulary terms and the relations between
         those terms
         –   Class, type, subClassOf,
         –   Property, subPropertyOf, range, domain
         –   it gives “extra meaning” to particular RDF predicates and resources
         –   this “extra meaning”, or semantics, specifies how a term should be
             interpreted.


19
     Semantic Web Architecture (5)

                        OWL stands for Web
                         Ontology Language.
                        OWL is a language
                         derived from description
                         logics.
                        OWL provides additional
                         standardized
                         vocabulary.
                        OWL provide reasoning
20                       support
     Semantic Web Architecture (6)

                         RIF/SWRL: rule
                          languages are being
                          standardized for the
                          semantic web.
                         Provide rules beyond
                          the constructs
                          available from RDFS
                          & OWL.

21
     Semantic Web Architecture (7)

                         SPARQL stands for
                          Simple Protocol And
                          RDF Query Language.
                         SPARQL is used to
                          query RDF data as
                          well as RDFS and
                          OWL ontologies with
                          knowledge bases.

22
     S-OGSA

        Why
        What
        How
        Design Principles
        S-OGSA
        Conclusions and future works
        Reference
        Q&A
23
     Why Semantic Grid ?

        Currently, Grid metadata is generated and
         used in an ad hoc fashion , represented in
         different formats.
         –   Its hard to share
         –   Its hard to reuse
         –   Its hard to reinterpret
        Semantic Grid is an extension of the Grid
         increases interoperability and greater
         flexibility
24
     What is Semantic Grid

        An extension of the Grid
        Rich metadata is exposed and handled
         explicitly, shared, and managed via Grid
         protocols




25
     What is Semantic Grid

        The Semantic Grid uses metadata to
         describe information in the Grid.
        Turning information into something more
         than just a collection of data means
         understanding the context, format, and
         significance of the data.
        Therefore:
         –   Understand information
         –   Discovery and reuse
26
     Semantic?

        Semantic = metadata + meaning
        Metadata explicitly exposed as a first class object in
         a machine processable form.
        Controlled vocabularies or knowledge models (aka
         Ontologies) for describing metadata in a machine
         processable form.
        Schemas for structuring metadata in a machine
         processable form.
        Rules over metadata.

     Possibly using Semantic Web technologies
     For people and machines
27
     Design Principles for a Reference
     Semantic Grid Architecture

        Parsimony
          – lightweight
          – minimize the impact on legacy Grid infrastructure
            and tooling.
        Extensibility
        Uniformity (of the mechanisms)
         –   manageability of S-OGSA entities
         –   Have both stateless and stateful Grid services like
             OGSA

28       –   S-OGSA services are OGSA-observant Grid services.
     Design Principles for a Reference
     Semantic Grid Architecture

        Diversity
         –   Mixed ecosystem of Grid and Semantic Grid
             services
                 Services ignorant of semantics
                 Services aware of semantics but unable to process them
                 Services aware of semantics and able to process (part
                  of) them




29
     Design Principles for a Reference
     Semantic Grid Architecture

        Heterogeneity (of semantic
         representation)
         –   Any resource‟s property may have many different
             semantic descriptions
         –   captured (or not) in different representational
             forms (text, logic, ontology, rule).




30
     Design Principles for a Reference
     Semantic Grid Architecture

        Enlightenment
         –   minimal impact on adding explicit semantics to
             current Grid entities
         –   Grid entities should not break if consume and
             process Grid resources but cannot consume and
             process associated semantics
         –   Grid entities can incrementally acquire, lose and
             reacquire explicit semantics during their lifetime


31
     S-OGSA

        Defined by
         –   Information model                         Model
                 New entities           provide/                expose
                                         consume
         –   Capabilites
                 New functionalities
                                        Capabilities           Mechanisms
         –   Mechanisms                                use
                 How it is delivered




32
     S-OGSA


    How to provide:
      –   Just give the semantic metadata to those services
      –   Or we can have the semantic services by SOGSA own.




33
     S-OGSA

        There are no big differences…
         –   if the service can understand semantic (e.g., they
             support semantic API), then itself can be a S-
             OGSA service.




34
     S-OGSA


    A Grid usually consist of several different services
     by OGSA:
      –   VO management service
      –   Resource discovery and Management service
      –   Job Management service
      –   Security service
      –   Data Management service
    The S-OGSA should (will) provide the metadata
     +semantic services to those services.
35
     S-OGSA

        The Solution:
         –   Attached the semantic to Grid entities.
         –   Binding them together by semantic binding
             service.
         –   Normal grid services can be “semantic” by the
             semantic binding service.




36
     S-OGSA Model. Semantic
     Bindings




37
                 S-OGSA
                          Application 1           Application N
 Semantic-OGSA




                      Security                   Optimization
       OGSA




                                                                            Data
                                                                Semantic Provisioning
                                                                     Services
                                    Execution




                                                                           Semantic binding
                                   Management
                                                                 Semantic
                                                               Ontology                       Metadata




                                                  Knowledge
                                                                Provisioning
                                                                  Services
                      Resource                                Reasoning                       Annotation
                     management
                                                Information
                                                Management

                      Infrastructure Services



38
     S-OGSA




39
      S-OGSA Model and Capabilities
                                    Annotation              Metadata                                    WebMDS
                                     Service                 Service
Ontology
                                                         Is-a
 Service
                                                                           Is-a                         OGSA-DAI
              Knowledge                Semantic Binding                            Grid Service
               Service                Provisioning Service
Reasoning                   Is-a
 Service
                   Is-a                                  Is-a                             Is-a            CAS

              Knowledge   1..m       Semantic Provisioning                  1..m
                Entity                     Service                                  Grid Entity


                   Is-a                                  uses                                             SAML
Ontology                                                                                  Is-a             file

              Knowledge                 Semantic aware
               Resource                   Grid Service                             Grid Resource
                                                                                                          DFDL
 Rule set                                         1..m             1..m                                    file
                                       produce




                                                         consume




                                                  0..m             0..m                                   JSDL
                                                                                                           file
                                       Semantic Binding
                             0..m                                         0..m

40     Knowledge                                 Semantic Grid
                                                                                   Is-a
                                                                                                 Grid
     S-OGSA Model and Capabilities
        Grid Entities
         –   Resources and services
        Knowledge Entities
         –   Grid Entities that represent or could operate with some form
             of knowledge (e.g ontologies, rules, knowledge bases …)
        Semantic Bindings
         –   entities associatie of a Grid Entity with one or more
             Knowledge Entities




41
     S-OGSA Model and Capabilities

        Semantic Grid Entities (all entitites in the binding
         model)
        Semantic Provisioning Services
                 provisioning and management of explicit semantics and
                  its association with Grid entities
                 creation, storage, update, removal and access of
                  different forms of knowledge and metadata
         –   Knowledge provisioning services
                 ontology services , reasoning services .
         –   Semantic binding provisioning services
                 metadata services, annotation services .
42
     S-OGSA Model and Capabilities

        Semantically Aware Grid Services
         –   Be able to consume Semantics Bindings and
             being able to take actions based on knowledge
             and metadata .
         –   Sample Actions :
               Metadata aware authorization of a given identity by a VO Manager
                  service .
                  Execution of a search request over entries in a semantic resource
                  catalogue .
                  Incorporation of a new concept in to an ontology hosted by an
                  ontology service .
                  Reduction of an annotated scientific data set to a smaller subset by a
43                scientist.
     S-OGSA Mechanisms

        Treating Knowledge Entities and Semantic
         Bindings as Grid Resources
         –   Common Information Model (CIM) Resource Model
         –    Grid Entities : class CIM-ManagedElement in the CIM
             Model.
         –   Knowledge Entities : class S-OGSA-KnowledgeEntity
         –   S-OGSA-SemanticBinding:Semantic Binding, the
             association between a Grid Entity (CIM-ManagedElement)
             and a Knowledge Entity (S-OGSA-KnowledgeEntity).


44
     S-OGSA Mechanisms




45
     S-OGSA Mechanisms

        S-Stateful Services: mechanisms for the
         delivery of Semantic Bindings for
         resources
         –   Based on Web Services Resource Framework
             (WSRF)




46
     Retrieving and Querying Semantic Bindings of Resources
          Query/Retrieval Result
      4
                                                           Metadata                         Ontology
                          Metadata Retrieval/Query         Service                           Service
                    3
                                 Request


                                                     5 Obtain schema for Semantic
                                                                Bindings




             Metadata                Semantic Binding Ids Retrieval Request
             Seeking                                             1
              Client
                                                                                                   Resource
                                                                                                   Specific

                                                                                              Lifetime

                                                               State/properties/metadata
                                                                              access port
     • A Feta ODE-SGS,              2                                                                               Resource
     OWL-S, WSMO                   Semantic Binding Ids
     service desc
     • FOAF Profile                                       • Deliver Metadata pointers                     Service




                                                                                                   ...
     •….                                                  through resource properties
                                                          • Zero impact on existing protocols



47
     Conclusions and future works

        Extensions to current Grid models to deal with
         flexible forms of explicit metadata
         –   The central component : Semantic Binding
        Define a set of services (Semantic Provisioning
         Services) that play an important role in the exposure,
         delivery and generation of metadata
         –   ontology management and reasoning services, metadata
             services and annotation services.
         The actual mechanisms to be used for treating the
         new components as Grid entities and for delivering
         them as part of existing Grid service frameworks.
48
     Conclusions and future works

        Design principles :
         –   The Semantic Grid is the Grid.
         –   The Semantic Grid has a spectrum of
             Semantic Capabilities.
         –   Painless migration to the Semantic Grid.
         –   Semantic Grid lifecycle.
         –   Multiple semantics.



49
     Conclusions and future works

        Challenges :
         –   Technical
                 architectural or theoretical foundations, the maturity of
                  Semantic and Grid technologies,
                 improving the performance of creating and retrieving
                  semantically-encoded metadata
         –   Operational
                 gathering and maintaining the semantic content
         –   Sociological and political
                 legal, security and privacy implications of clearly
                  exposed metadata and automated reasoning
50
     Q&A




51
     Implementation

        e-Science
        myGrid




52
     e-Science

        ‘e-Science is about global collaboration in key
         areas of science, and the next generation of
         infrastructure that will enable it.’
        ‘e-Science will change the dynamic of the way
         science is undertaken.’
                                   John Taylor, DG of UK OST
         ‘[The Grid] intends to make access to computing
         power, scientific data repositories and
         experimental facilities as easy as the Web makes
         access to information.’
                                              Tony Blair, 2002
53
     UK e-Science Grid




54
     UK e-Science Initiative

        $180M Programme over 3 years
        $130M is for Grid Applications in all areas of
         science and engineering
         –   Particle Physics and Astronomy (PPARC)
         –   Engineering and Physical Sciences (EPSRC)
         –   Biology, Medical and Environmental Science
        $50M ‘Core Program’ to encourage
         development of generic ‘industrial strength’
         Grid middleware
55
     Some UK e-Science Projects

        GRIDPP (PPARC)            Climateprediction.com (NERC)
        ASTROGRID (PPARC)         Oceanographic Grid (NERC)
        Comb-e-Chem (EPSRC)       Molecular Environmental Grid (NERC)
        DAME (EPSRC)              NERC DataGrid (NERC + OST-CP)
        DiscoveryNet (EPSRC)      Biomolecular Grid (BBSRC)
        GEODISE (EPSRC)           Proteome Annotation Pipeline (BBSRC)
        myGrid (EPSRC)            High-Throughput Structural Biology
        RealityGrid (EPSRC)        (BBSRC)
                                   Global Biodiversity (BBSRC)



56
     Some UK e-Science Projects

        Biology of Ageing (BBSRC + MRC)
        Sequence and Structure Data (MRC)
        Molecular Genetics (MRC)
        Cancer Management (MRC + PPARC)
        Clinical e-Science Framework (MRC)
        Neuroinformatics Modeling Tools (MRC)

        Interdisciplinary Research Collaborations „Grand
         Challenge‟
          – Advanced Knowledge Technologies
          – Medical Images and Signals
          – Equator
57        – DIRC (Dependability)
     Content

        e-Science
        myGrid
         –   Context
                 Workflows, repository, registry and provenance
         –   Concept services
         –   Using concepts
                 Discovering workflows and services
                 Workflow composition support
                 Discovering and linking experimental components
                 Linking provenance logs
         –   Remarks

58
     myGrid




        EPSRC UK e-Science pilot project
        Open Source Upper Middleware for
         Bioinformatics
        Knowledge-driven Middleware for data intensive
         in silico experiments in biology
        (Web) Service-based architecture -> OGSA Grid
         services
        Targeted at Tool Developers, Bioinformaticians
         and Service Providers
         http://www.mygrid.org.uk


59
     Data intensive bioinformatics




60
     Graves Disease
     Autoimmune disease of the thyroid




61
     Workflows as in silico experiments


                         Freefluo workflow enactment engine
                           –   WSFL
                           –   Scufl
                         Workflow discovery
                           – Finding workflows that others
                              have done, and that I have done
                              myself
                         Workflow creation
                           – Finding classes of services
                           – Guiding service composition
                           – We don‟t do automated
                              composition
                         Dynamic workflow enactment
                          service discovery and invocation
                           – Choose services instances
                              when running workflow
                         User involvement
63
     FreeFluo and Taverna
     environments

        Freefluo workflow
         enactment
         engine
         –   WSFL
         –   Scufl
        Taverna
         development
         environment




64
         Investigation = set of experiments +
         metadata

        Experimental design components
          –   workflow specifications; query
              specifications; notes describing objectives;
              applications; databases; relevant papers;
              the web pages of important workers,

        Experimental instances that are records
         of enacted experiments
          –   data results; a history of services invoked
              by a workflow engine; instances of services
              invoked; parameters set for an application;
              notes commenting on the results

        Experimental glue that groups and links
         design and instance components
          –   a query and its results; a workflow linked
              with its outcome; links between a workflow     •Life Science IDs & URIs
              and its previous and subsequent versions; a    •RDF-based annotations
              group of all these things linked to a
              document discussing the conclusions of the     •DAML+OIL -> OWL ontologies
65            biologist
     Experiment life cycle




67
         Sharing info  Sharing meaning

     Metadata
        Data describing the content and
         meaning of resources and services.
        But everyone must speak the same
         language…
     Terminologies
        Shared and common vocabularies
        For search engines, agents,
         curators, authors and users
        But everyone must mean the same      • A common vocabulary of
         thing…                               terms
     Ontologies                               • Some specification of the
        Shared and common understanding      meaning of the terms
         of a domain                          • A shared understanding for
        Essential for search, exchange and
68       discovery                            people and machines
     myGrid   Service Stack




70
     W3C Ontology and Metadata
     languages
        OWL (and DAML+OIL)
         –   The Web Ontology Language OWL
         –   Family of languages: OWL Lite, OWL DL & OWL Full
         –   OWL DL = DAML+OIL
         –   Expressive language for describing concepts, relationships, constraints and
             axioms
         –   Sound and complete, and efficient, reasoning over expressions to infer
             relationships between concepts rather than assert them (including the
             hierarchy).
         –   OWL is W3C Candidate recommendation.
        RDF
         –   Resource Description Framework
         –   W3C language for describing metadata on the Web
         –   Triples (subject, predicate, object) forming graphs
         –   Associate URIs (LSIDs) with other URIs (LSIDs)
         –   Associate URIs with OWL concepts (which are URIs)
         –   RDQL
         –   Triple store RDF implementations (e.g. Jena)
71       –   http://www.w3.org/RDF
         Concept services: Ontology
         Services

        Ontology server for concept expressions
        Ontology development environments
          –   OilEd
        FaCT reasoner for inferring over concept
         expressions
        Imprecise matchmaking for best effort
         substitutability
          –   Reasoning over descriptions
          –   Generating classification structures
        Matchmaker and ranking for matching
         concept expressions
        Instance store for indexing instances of
         concept expressions in registries and
         databases
72
         Concept services: Annotation
         services

        RDF repositories
          – Jena Toolkit
          – RDF query languages RDQL
        myGrid Information Repository
          – Version 1: Relational (DB2)
          – Version 2: Federated architecture.
        Browsers for annotating objects and
         viewing annotations
        Automated tools for marking up objects
         with annotations.



73
     myGrid Information Repository

                            Stores experimental
                             components
                              –   Workflow specs as XML
                                  Scufl docs
                              –   Data
                              –   XML notes
                            Types
                              –   XML docs
                              –   Relational
                              –   RDF (like)
                            Every entry has Dublin
                             Core provenance
                             attributes
                            Every entry can have
                             (multiple) concept OWL
                             concept expressions
74                          Multiple mIRs
     Registries
        Publishes experimental components: services, workflows and
         (distributed query plans in the future?)
        Multiple & 3rd party registries
        Multiple & 3rd party metadata




75
     Using Concepts

        Controlled vocabulary for advertisements for
         workflows and services
        Indexes into registries and mIR
         –   Semantic discovery of services and workflows
         –   Semantic discovery of repository entries
        Type management for composition
         –   Semantic workflow construction: guidance and validation
        Navigation paths between data and knowledge
         holdings
         –   Semantic “glue” between repository entries
         –   Semantic annotation and linking of workflow provenance
             logs

76
     Semantic discovery – services &
     workflows
                                    Services and workflows in registry
            A registry browser       have RDF and OWL descriptions
                                    Selection by the types of inputs
                                     they use, outputs they produce,
                                     the bioinformatics tasks they
                                     perform…
                                    Querying using RDQL over RDF
                                     UDDI registry for operational
                                     metadata
                                    Matching using FaCT OWL
                                     classification for concept-based
                                     metadata




                                     A workflow wizard
77
     Workflow construction

        Outputs and inputs of
         chained services are
         compatible
          –   OWL Concept
          –   XSD Type
          –   Data Format
        Workflows are
         constructed in
         collaboration with
         Scientist
          –   No automated
              workflow creation
        Find service being
         embedded into
         Taverna by end
79       October like Geodise
         approach
     Linking objects to objects via
     concepts




80
     Reference
        Professor Carole Goble and the myGrid consortium, Knowledge-based
         Middleware for BioGrid services from the myGrid Project
        Professor Carole Goble and the myGrid consortium, The Role of
         Concepts in myGrid
        http://www.mygrid.org.uk
        http://www.semanticgrid.org
        http://www.w3.org
        An overview of S-OGSA: a Reference Semantic Grid Architecture
          –   Oscar Corcho, Pinar Alper, Ioannis Kotsiopoulos, Paolo Missier, Sean Bechhofer and
              Carole Goble School of Computer Science The University of Manchester,
              Manchester, UK
        The Semantic Grid
          –   Wei Xing1 , Marios Dikaiakos2 (1School of Computer Science University of
              Manchester, 2Department of Computer Science University of Cyprus)




81

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:5
posted:11/4/2011
language:Vietnamese
pages:77