Docstoc

PowerPoint - The Semantic Web meets the Grid

Document Sample
PowerPoint - The Semantic Web meets the Grid Powered By Docstoc
					The Semantic Web meets
               the Grid

                                        Carole Goble
                            University of Manchester
                                                 UK
                              carole@cs.man.ac.uk

                             Co-chair Grid Global Forum
                          Semantic Grid Research Group
Invited talk, SWFAT, Nara, Japan 12th March 2003
Roadmap


 Semantic Web – a short reprise
 The Grid and its evolution
 The Semantic Grid ??
  Grid Applications
  Grid Infrastructure
 Semantic Grid projects
 Closing remarks


                 Invited talk, SWFAT, Nara, Japan 12th March 2003
Take home message
 At the highest level the Grid and the Semantic Web
  appear to share common goals
 The Grid is metadata driven middleware
    An application for Semantic Web technologies and tools
 Semantics on the Grid
    Applications – lots of (obvious) attention
 Semantics for / in the Grid
    Infrastructure – little attention (see Decker P2P)
 Semantic Open Grid services
    Stateful, Transient, lifetime management
 Grid is about scale, flexibility and performance
    Are the technologies of the Semantic Web up to it?
 Overcome suspicions and discipline boundaries
    Those wacky AI guys! Those Grid people only care about ftp!
                    Invited talk, SWFAT, Nara, Japan 12th March 2003
The Semantic Web
“The Semantic Web is an extension of the current Web in
  which information is given a well-defined meaning, better
  enabling computers and people to work in cooperation. It is
  the idea of having data on the Web defined and linked in a
  way that it can be used for more effective discovery,
  automation, integration and reuse across various
  applications. The Web can reach its full potential if it
  becomes a place where data can be processed by
  automated tools as well as people”
               From the W3C Semantic Web Activity statement

“computational agents require machine-readable descriptions
   of the content and capabilities of web accessible resources.
   These descriptions must be in addition to the human-
   readable versions of that information. “
                                                               From the OWL Guide
                   Invited talk, SWFAT, Nara, Japan 12 March 2003
                                              th
Semantic Web as connective fabric for the Web

                                                                                     <daml:ObjectProperty rdf:ID="photograph">
                                                    Event:title                      <rdfs:domain rdf:resource="#Picture"/>
                                                                                     <rdfs:range rdf:resource= …#person"/>
                                                                                     </daml:ObjectProperty>
                      Event:WebPage


                           < > rdf:type photo:Photograph,                                 <s:Class
                                                                                          rdf:about="http://www.semanticweb.org/o
                              Photo:File http://…/images#image1,                          ntologies/swrc-onto-2000-09-
                              Photo:topic :event1#event:speaker.                          10.daml#Conference">
                                                                                          <s:comment>
                            Event1 a Event:event;                                         describes a generic conceptabout events
                                                                                                </s:comment>
                              date “May 7-11”,                                            <s:subClassOf
                              speaker http://…#timbl.html                                 rdf:resource="http://www.semanticweb.or
                              Title “WWW 2002…”                                           g/ontologies/swrc-onto-2000-09-
                                                                                          10.daml#Event"/>
                                                                                          <a:disjointFrom
                           TimBL rdf:type w3c-ont:person;                                 rdf:resource="http://www.semanticweb.or
                             name “Tim Berners-Lee”                                       g/ontologies/swrc-onto-2000-09-
                             …                                                            10.daml#Workshop"/>
                                                                                          <a:restrictedBy
                                                                                          rdf:resource="http://www.semanticweb.or
                                                                                          g/ontologies/swrc-onto-2000-09-
                                                                                          10.daml#genid18"/>



                       <rdf:Description rdf:about="http://www.w3.org/2001/03/earl/0.95#Person">
                       <rdf:type rdf:resource="http://www.w3.org/2000/01/rdf-schema#Class"/>
                       <rdfs:subClassOf rdf:resource="http://www.w3.org/2001/03/earl/0.95#Assertor"/>
(Hendler, Maryland)    </rdf:Description>
                        Invited talk, SWFAT, Nara, Japan 12th March 2003
 Bringing Semantics to the Web
                                         Semantic Web vision




 Semantic Web
  technologies and
  tools

               Invited talk, SWFAT, Nara, Japan 12th March 2003
   Putting Semantics in the Web?
 Web services
   Web-accessible programs and
    devices.
   perform functions, which can be
    anything from simple requests to
    complicated business
    processes!
   will transform the Web from a
    collection of information to a
    distributed device of
    computation
 Web services require a
  semantic-driven description
 Semantics as a component
                                                             [Fensel et al, 2002]
  of the infrastructure. SWFAT, Nara, Japan 12
                    Invited talk,                th   March 2003
Semantic                     Web
Web                         Services
Services




           Semantic
             Web

             Invited talk, SWFAT, Nara, Japan 12th March 2003
                                        The Grid
             Isn‟t it just high performance
               computing for high energy
                                 physicists?



Invited talk, SWFAT, Nara, Japan 12th March 2003
The Collaboratory Concept

 In 1989, William Wulf, then with the U.S.
  National Science Foundation, defined a
  collaboratory as
  "a center without walls, in which the
  nation's researchers can perform their
  research without regard to geographical
  location, interacting with colleagues,
  accessing instrumentation, sharing data
  and computational resources, and
  accessing information in digital libraries."
              Invited talk, SWFAT, Nara, Japan 12th March 2003
 The Grid Problem                                       (Foster, Kesselman, Tueke)


“flexible, secure, coordinated resource sharing among
  dynamic collections of individuals, institutions, and
  resources - what we refer to as virtual organizations."




“When the network is as fast as the computer's internal
 links, the machine disintegrates across the net into a set
 of special purpose appliances”                  (George Gilder)
                   Invited talk, SWFAT, Nara, Japan 12th March 2003
Why Grids?
 A biochemist exploits 10,000 computers to screen 100,000
  compounds in an hour;
 A biologist combines a range of diverse and distributed
  resources (databases, tools, instruments) to answer
  complex questions;
 1,000 physicists worldwide pool resources for petaop
  analyses of petabytes of data
 Civil engineers collaborate to design, execute, & analyze
  shake table experiments
 Climate scientists visualize, annotate, & analyze terabyte
  simulation datasets
 An emergency response team couples real time data,
  weather model, population data
 A multidisciplinary analysis in aerospace couples code and
  data in four companies
                                                        From
                 Invited talk, SWFAT, Nara, Japan 12th March 2003Steve   Tuecke 12 Oct. 01
Enabler of 21st Century Science




          Invited talk, SWFAT, Nara, Japan 12th March 2003
CERN: Large Hadron Collider (LHC)
Raw Data: 1 Petabyte / sec
Filtered 100Mbyte / sec = 1 Petabyte / year = 1 Million CD ROMs




                                             CMS Detector
                   Invited talk, SWFAT, Nara, Japan 12th March 2003
What is the Grid?
 Trusted resource sharing & coordinated problem
  solving in dynamic, multi-institutional virtual
  organizations
 On-demand, ubiquitous access to computing, data,
  and all kinds of services
 New capabilities constructed dynamically and
  transparently from distributed services
 No central location, No central control, No existing
  trust relationships, Little predetermination
 Uniformity for pooling resources
 Virtual pools of resources: databases, clusters,
  instruments, applications….
 Virtual configurations Nara, Japan 12 March 2003
                 Invited talk, SWFAT,
                                th
                  Biogrid system
 SCORE
 Management Station
                                                                         SCORE
                                                                         Management Station
                                                  Myrinet-2000




                                                                                              Flat Neighborhood networks
Connected to
Grid system3            Grid system 1
                        Express5800/ISS for PC-Cluster
                        Xeon2.2G x 8 + Management node1



1000Base-SX


                                                                          Grid system 2
                                                                          NEC Blade Server
           1000Base-T x 12                                                78node(156CPU)



                                                   Data Grid Disk
                                                   Express5800/140Ra-4 x3
                              Invited talk, SWFAT, Nara, Japan 12th March 2003
             NovartisGri
                                                               A Grid vs
                 d
                               BioSimGrid
                                                               The Grid
MouseGrid

                                                           A Grid of resources, not just




                                               Logical
                                                           compute resources but
                                                           databases, digital libraries,
                                                           instruments, workflows,
        Grid Middleware                                    documents …


                                                           These configurations are
      Gigabit IP Network

                                                Physical
                                                           dynamic
                                                           Resources discovered,
                                                           combined, used and
 Node       Node
                                                           disbanded as and when
                                    Node                   needed or available.
Geographically     Node
(e.g. UKGrid)
                   Invited talk, SWFAT, Nara, Japan 12th March 2003
So What is Grid Infrastructure?
 Broadly deployed services in support of fundamental
  collaborative activities
   Formation & operation of virtual organizations
   Authentication, authorization, discovery, …
 Services, software, and policies enabling on-demand
  access to critical resources
   Computers, databases, networks, storage, software
     services,…
 Operational support for 24x7 availability
 Integration with campus and commercial infrastructures



               Invited talk, SWFAT, Nara, Japan 12th March 2003
 Datagrid perspective
                                    ~PBytes/sec
                                                                                                               1 TIPS is approximately 25,000
                                                       Online System          ~100 MBytes/sec                  SpecInt95 equivalents

                                                                                   Offline Processor Farm
         There is a “bunch crossing” every 25 nsecs.
                                                                                          ~20 TIPS
         There are 100 “triggers” per second
                                                                                                        ~100 MBytes/sec
         Each triggered event is ~1 MByte in size

                                                      ~622 Mbits/sec
                                                                         Tier 0               CERN Computer Centre
                                       or Air Freight (deprecated)

Tier 1
         France Regional                   Germany Regional                  Italy Regional                     FermiLab ~4 TIPS
             Centre                            Centre                           Centre
                                                                                                                              ~622 Mbits/sec


                                                           Tier 2            Caltech                   Tier2
                                                                                              Tier2 Centre Centre        Tier2 Centre
                                                                                                               Tier2 Centre
                                                                             ~1 TIPS            ~1 TIPS ~1 TIPS ~1 TIPS ~1 TIPS
                                            ~622 Mbits/sec


                               Institute
                                       Institute Institute       Institute
                              ~0.25TIPS                                                       Physicists work on analysis “channels”.
                                                                                              Each institute will have ~10 physicists working on one or more
     Physics data cache
                                                ~1 MBytes/sec                                 channels; data for these channels should be cached by the
                                                                                              institute server
                                                                Tier 4
                   Physicist workstations


                                               www.ppdg.net www.eu-datagrid.org
                        www.griphyn.orgSWFAT, Nara, Japan 12 March 2003
                                 Invited talk,                                                th
Grid application evolution

                  Functional Genomics                   Smaller scale data, less
                  Oceanography                          machine computational
                  Biodiversity                             intensive, complex
                  Earth Science                       heterogeneous applications,
                  Neuroscience                         complex semantics, many
                                                                 people




  Large scale data, large
   number of machines,
                                     High Energy Physics
  expensive computation,
  simple semantics, small
    numbers of people


                     Invited talk, SWFAT, Nara, Japan 12th March 2003
Status of Grid Computing?

  1995 – 2000+: “Grid book” gave a
   comprehensive view of the state of the art
  Important infrastructure and middleware
   efforts initiated
    Globus, Legion, Condor, SRB, etc.



  2000+: Beginnings of a Global Grid
    Evolution of the Global Grid Forum
    Some projects evolving to de facto
     standards (e.g. Globus, Condor)

                  Invited talk, SWFAT, Nara, Japan 12th March 2003
But aren‟t the Grid guys still
              trapped in ftp?
            Deep impact of ontologies
      Ontologies infiltrate infrastructure




 Invited talk, SWFAT, Nara, Japan 12th March 2003
Today‟s Grid
 A Single System Image                   Security, certification,
 Transparent wide-area                    single sign-on
                                           authentication, AAA
  access to large data
                                                 Grid Security
  banks                                           Infrastructure,
 Transparent wide-area                   Data access,Transfer &
  access to applications on                Replication
  heterogeneous platforms                        GridFTP, Giggle
 Transparent wide-area                   Computational resource
  access to processing                     discovery, allocation and
  resources                                process creation
                                                 GRAAM, Unicore,
                                                  Condor-G

                 Invited talk, SWFAT, Nara, Japan 12th March 2003
               Grid Evolution
               1st generation
                                                             • Computationally intensive
                                                             • File access/transfer
                                                             • Bag of various heterogeneous
                                                             protocols & toolkits
Increased functionality,




                                                             • Recognises internet, Ignores Web
                                                             • Academic teams
    standardization




                           X.509,
                           LDAP,       Globus Toolkit                      Legion, Condor, Unicore …
                           FTP, …
                                       Defacto standards
                           Custom      GGF: GridFTP, GSI
                           solutions

                                                                                Time
                                         Invited talk, SWFAT, Nara, Japan 12th March 2003 (based   on Foster, 2003)
                                                           • Data intensive -> knowledge intensive
               Grid Evolution                              • Open services-based architecture
                                                           • Recognises Web services
               2nd generation                              • Global Grid Forum
                                                           • Industry participation


                                                                                               App-specific
Increased functionality,




                                                                                                Services
    standardization




                                                                       Open Grid
                                       Web services
                                                                      Services Arch
                                                               GGF: OGSI, …
                           X.509,                             (+ OASIS, W3C)
                           LDAP,       Globus Toolkit Multiple implementations,
                           FTP, …                         including Globus Toolkit
                                        Defacto standards
                           Custom      GGF: GridFTP, GSI
                           solutions

                                                                                Time
                                         Invited talk, SWFAT, Nara, Japan 12th March 2003 (based   on Foster, 2003)
Open Grid Services Architecture
 Present Grid Architecture                 OGSI extends Web Services
  is a services architecture                   Transient Service
                                                Instances
 Implemented using Web
                                               Service State
  Services Technology
                                               Lifetime management
 OGSA will provide
    Naming /Authorization /
     Security / Privacy
                                            Defines fundamental (WSDL)
    Higher level services:                  interfaces and behaviors that
     Workflow, Transactions,                 define a Grid Service
     Data Mining,Knowledge
     Discovery,…                               Required + optional
                                                 interfaces = WS “profile”
 Exploiting Synergy:                       Defines WSDL extensibility
  Commercial Internet with                   elements
  Grid Services                                E.g., serviceType (a group
                                                 of portTypes)
                   Invited talk, SWFAT, Nara, Japan 12th March 2003
Grid service configurations are:
 Dynamic and volatile. A consortium of services
  (databases, sensors, compute servers)
  participating in a complex analysis may be
  switched in and out they become available or
  cease to be available;
 Ad-hoc. Service consortia have no central
  location, no central control, and no existing trust
  relationships;
 Large Hundreds of services could be
  orchestrated at any time;
 Potentially long-lived. A simulation could take
  weeks.

                Invited talk, SWFAT, Nara, Japan 12th March 2003
 Transient Service Instances
“Web services” address discovery &
 invocation of persistent services
  Interface to persistent state of entire enterprise
In Grids, must also support transient service
 instances, created/destroyed dynamically
  Interfaces to the states of distributed activities
  E.g. workflow, video conf., dist. data analysis
Significant implications for how services are
 managed, named, discovered, and used

                Invited talk, SWFAT, Nara, Japan 12th March 2003
Open Grid Service Architecture:
“where are the OGSI services”

  Technical specifications
    Open Grid Services Infrastructure is complete
    Security, data access, Java binding, common
      resource models, etc., etc., in the pipeline
  Implementations and compliant products
    OGSA-based Globus Toolkit v3, pyGlobus, …
    IBM, Avaki, Platform, Sun, NEC, Oracle, …
  Rich set of service defns & implementations
    Starting on OGSI-compliant services
    OGSA-DAI

                 Invited talk, SWFAT, Nara, Japan 12th March 2003
                                              The Grid
                                              Technology
                                              Repository
                                    Community repository
                                    Clearing house for
                                     service definitions,
                                     code, documentation
                                    Encourage
             http://gtr.globus.org   collaboration & avoid
                                     redundant work
                                    OGSA-DAI
International advisory committee: Ian Foster (Chair), Malcolm Atkinson,
John Brooke, Fabrizio Gagliardi, Dennis Gannon, Wolfgang Gentzsch, Andrew
Grimshaw, Keith Jackson, Gregor von Laszewski, Satoshi Matsuoka, Jarek
Nabrzyski, Bill St. Arnaud, Jay Unger Nara, Japan 12th March 2003
                          Invited talk, SWFAT,
Challenging Technical Requirements
 Dynamic formation and management of virtual
  organizations
 Online negotiation of access to services: who,
  what, why, when, how
 Configuration of applications and systems able
  to deliver multiple qualities of service
 Autonomic management of distributed
  infrastructures, services, and applications
 Management of distributed state as a
  fundamental issue
               Invited talk, SWFAT, Nara, Japan 12th March 2003
Grid Services


                                                             Grid
                       Web                                   services
                      Services




                                              Grid


          Invited talk, SWFAT, Nara, Japan 12th March 2003
                    What is the Grid?
             A semantic web perspective




Invited talk, SWFAT, Nara, Japan 12th March 2003
Realizing a Service-Oriented
Architecture: How Do I
 Create, name, manage, discover services?
 Render resources, data, sensors as services?
 Negotiate service level agreements?
 Express & negotiate policy?
 Organize & manage service collections?
 Establish identity, negotiate authentication?
 Manage VO membership & communication?
 Compose services efficiently?
 Achieve interoperability?
                                                       (Foster, Argonne Labs, 2003)
               Invited talk, SWFAT, Nara, Japan 12th March 2003
Semantic Grid: the gap
 A gap between grid computing endeavours and
  the vision of Grid computing
  high degree of easy-to-use and seamless automation
  flexible collaborations and computations on a global
   scale.
 To support the full richness of the grid computing
  vision we need both grid and semantic web
  (technologies)
 Knowledge explicitly asserted & explicitly used.
 The Semantic Grid http://www.semanticgrid.org
                Invited talk, SWFAT, Nara, Japan 12th March 2003
Richer semantics
                   Semantic                         Semantic
                     Web                              Grid

                   Classical                          Classical
                     Web                                Grid

                                More computation
                                                    Source: Norman Paton
                     Invited talk, SWFAT, Nara, Japan 12th March 2003
Grid is … metadata based middleware
 Metadata drives the Grid.
 Metadata links and interoperates the
  components of the Grid.
 Metadata services are the Semantic Grid.
 At ALL level of the Grid (e.g. resource brokering,
  load balancing, provenance, trust, workflow,
  context, PSM, database schema and on and on
  and on).
 The technologies developed by the Semantic
  Web for metadata are important.

               Invited talk, SWFAT, Nara, Japan 12th March 2003
                                         Source: Keith Jeffery

Invited talk, SWFAT, Nara, Japan 12th March 2003
Metadata take 1: Grid applications
 Representing the knowledge and operational know-how
  of the domain.
    a query or workflow; the annotation of results, workflows,
     database entries and parameters of analyses with personal
     notes, provenance data describing sources and derivation paths
     of information, and explanations or claims; linking in silico and „at
     the bench‟ experimental components (literature, notes, code,
     databases, intermediate results, sketches, images, workflows,
     the person doing the experiment, the lab they are in, the final
     paper); descriptions of people, labs, literature, tools
    knowledge about the domain such as molecular biology or
     astronomy which can be contradictory, incomplete so for forth,
     just as with any resource on the web.

 Indistinguishable from the Semantic Web


                     Invited talk, SWFAT, Nara, Japan 12th March 2003
Science must “use the links”
  Web

                                                                            PVT
 Burkitt‟s Lymphoma                                                             Rearrangement of a DNA
                                                                                sequence homologous
                                                                                to a cell-virus junction
                                                                                fragment in several Moloney
                                                                                murine leukemia
                                                                                virus-induced rat thymomas
                                                                                                    PubMed




   Semantic Web             Oncogene(MYC):
                             Found_In_Organism(Human).
                             Gene_Has_Function(Transcriptional_Regulation).
                             Gene_Has_Function(Gene_Transcription).
                             In_Chromosomal_Location(8q24).
                             Gene_Associated_With_Disease(Burkitts_Lymphoma).

                                                                                 PVT
 Burkitt‟s Lymphoma                                                             Rearrangement of a DNA
                                                                                sequence homologous
                                                                                to a cell-virus junction
                                                                                fragment in several Moloney
                      8q24                                                      murine leukemia
                      PVT1                                                      virus-induced rat thymomas

                      Invited talk, SWFAT, Nara, Japan 12th March 2003                                PubMed
    Semantics on the Grid:
    Application Knowledge
                        Question: What ATPase superfamily proteins are found in human?
                                                                                                   Provenance
Database query                                                                                     (know-wherefrom)
                        1. P31946 from Swiss-Prot version 30, 05/11/02, 16:45 GMT, EBI server.
(know-what)             2. P42655 from Swiss-Prot version 29, 05/11/02 16:45 GMT, local copy.
                        3. P35214 and P31946 derived by a distributed query 2345.qry over
                        Enzyme and PIR.                                                            Replicas
Virtual data products   4. InterPro (no particular version) is a pattern database for protein      (know-which)
(know-how)              superfamilies and domains for GPCR’s but you need an account.
                        5. The publicly available workflow humanATPase.wf
                                                                                                   Ontology and
                        (http://www.somelab.edu/bio/carole/wf/3345.wsfl) will generate the
Workflow                                                                                            Inference
                        result from mouse data in your personal repository and you have
(know-how)                                                                                         (know-whether)
                        permission to run the services it needs. Click to run it.
                        6. The Attwood lab expertise is in nucleotide binding proteins (ATPase
Personalised profile    superfamily proteins are nucleotide binding proteins).
(know-whom-to)          7. Jones published a paper on this in Nature Genetics two weeks ago, and   Authorisation,
                        you have an account to access it on-line.                                  Authentication and
                        8. Smith in your lab asked this question yesterday and the answer he got   Accounting
Collaboration           is annotated by a commentary in his e-Log Book.                            (know-who)
& community             9. Q04917 calculated by applying the algorithm BLASTp located at NCBI
(know-where,            using data in MouseDB
 know-when)
                                                                                                   Explanation
Digital archive                                       Annotation & notes                           (know-why)
(know-which)                                          (know-that)



                                       Invited talk, SWFAT, Nara, Japan 12th March 2003
Metadata take 2: Grid infrastructure
 Driving the machinery of the Grid computing infrastructure.
    the classification of computational and data resources,
     performance metrics, job control; schema integration, workflow
     descriptions, resource brokering, resource scheduling, service
     state, event notification topics, typing service inputs and outputs,
     provenance trails; access rights to databases, personal profiles
     and security groupings; charging infrastructure …
    problem solving selection and intelligent portals;
 Knowledge permeates the Grid
    Data elements
    Service descriptions (service data elements)
    Protocols (e.g. policy, provisioning)
 More dynamic and general model than Semantic Web
    OGSA lifetime model
    OGSA Service Data Element model
                     Invited talk, SWFAT, Nara, Japan 12th March 2003
                                                                                    [Kesselman, USC]
 User and VO policy                                                                  Application Component
        models                                                                                 Models

                                       Users and Applications

           High-level
           Request                                                                  Current Request Status, Results,
          descriptions                                                                 Provenance Information

Intelligent Reasoners (matchmaking, refinement, repair, coordination, negotiation…)


 Policy Knowledge-                     Refined                    Provenance andResource Knowledge-
                                       Workflow                     Monitoring
         bases                                                                           bases

              Higher-Level Service (Virtual Data Tools, Resource Brokers)

                         Tasks                                                      Monitoring, Resources
                                                                                        knowledge
      Resource Policy                                                                 Semantic Resource
        Descriptions                                                                       Descriptions

              Basic Grid Middleware (Globus Toolkit, Condor-G, DAGMan)


                           Grid Resources (Compute, Data, Network)
                                 Invited talk, SWFAT, Nara, Japan 12th March 2003
                                                                                                     [Kesselman, USC]


                                      Community
                                        Users
  High-level specification of
 desired results, constraints,
 requirements, user policies



                                                                                                                   Resource
                                                                                          Resource                   KB
                                                                                          Indexes
                                                                Application
                                                                   KB
           Workflow                 Workflow
           Refinement                 Workflow
                                    history
  Policy                                 Workflow                                                         Simulation
                                       history
Management                              History          Replica                                            codes
                                                         Locators                                                      Other
                                   Smart Workflow                                                                      Grid
                                         Pool                                Community Distributed Resources           services
Resource         Workflow                                                    (e.g., computers, storage, network,
Matching          Repair          Workflow Manager                                  simulation codes, data)
                                                                    Policy             Policy
                                                                     KB                Information
                                                                                       Services
                                                                                                                    Other
                                                                                                                     KB

Intelligent Reasoners                                                         Pervasive Knowledge Sources




                                 Invited talk, SWFAT, Nara, Japan 12th March 2003
Resource selection: Matchmaking
 Providers and requesters describe themselves
   Syntactic description
      Structured or Semi-structured
 A Matchmaker matches compatible ClassAds (Condor)
   Match based on attribute name, simple prioritization
 Semantic matchmaking
   Inference based matching (e.g. CIM+relations)
   Automatic classification (e.g. description logic)
   Leverage domain specific ontologies



                 Invited talk, SWFAT, Nara, Japan 12th March 2003
                        Grid Interoperability Project
                        Interoperable Resource Broker
                                                  NJS
   Resource Discovery                                                                   Diagram
                                                        Delegates resource check
        Service                                                                         Of Broker
                                                                                        Architecture
                                               Broker                                                            Other
                                                                                                                 Brokers


     Unicore Broker                                                             Globus Broker
                                                  Delegates translation
                                                                                                 Uses to drive
      Lookup                                                                                     MDS search
      resources
                             Uses to
                             Drive MDS                  Translator
                             Search
                  IDB                                                                   Filter
Nodal                                    Ontology engine                                               Hierarchical
                                                                                                       Grid Search
Grid Search
                    Filter
                                         Resource Discovery
                                              Service                                      [Brooke]
                                     Invited talk, SWFAT, Nara, Japan 12th March 2003
Semantic Grid


   Semantic                                                       Grid
                              Web
   Web                                                            services
                             Services
   Services
                           Semantic
                             Grid
              Semantic
                                                 Grid
                Web


                Semantics for the Grid

                                  ??
               Invited talk, SWFAT, Nara, Japan 12th March 2003
Observations
 The Grid has been about large scale
  computation
 But the applications are also about collaboration

 Middleware has provided some computational
  interoperability
 But we now need semantic interoperability

 The old problem was lots of different computers
 The new problem is lots of different projects!
               Invited talk, SWFAT, Nara, Japan 12th March 2003
The Semantic Web meets the Grid




          Invited talk, SWFAT, Nara, Japan 12th March 2003
Semantic Grid Panel
                                    WWW2002 Semantic Grid Panel




         What do grid computing and semantic web have in
          common? Where do they differ?
         Does the Grid need the Semantic Web?
         Does the Semantic Web need the Grid?
         Where do you think it's going in 50 years?
         What is the biggest challenge we must address to
          realise the semantic grid?
       Grid meets Web
            Invited talk, SWFAT, Nara, Japan 12th March 2003
SW -> Grid Community
Global Grid Forum 5 in Edinburgh
 Semantic Grid BOF
 Ontologies and the Grid tutorial
 Semantic Web keynote




                 Dave                     Carole                Nigel

             Invited talk, SWFAT, Nara, Japan 12th March 2003
                             semgrd-rg@gridforum.org
                             http://www.gridforum.org




Invited talk, SWFAT, Nara, Japan 12th March 2003
        Semantic Grid projects
From                                                  myGrid
grid users                                           Geodise
                                                       SCEC
to                                                  CoAKTing
grid applications                                       BIRN
to                                                     GEON
grid infrastructure                                  Pegasus
                                                        GRIP

 Invited talk, SWFAT, Nara, Japan 12th March 2003
Tools to assist scientific
collaboration by integrating
intelligent meeting spaces
ontologically annotated media
 streams from online meetings
decision rationale and group
 memory capture
meeting facilitation
issue handling, planning and
 coordination support
constraint satisfaction
instant messaging/presence.
              Invited talk, SWFAT, Nara, Japan 12th March 2003
     myGrid


              myGrid   Project - Bioinformatics
     Imminent „deluge‟ of
      genomics data
     Highly heterogeneous
     Highly complex and inter-
      related
     Convergence of data and
      literature archives
1.    Database access
2.    Process enactment
3.    Provenance and notification
      services
4.    Metadata services
      Grid Services + Ontologies
                       Invited talk, SWFAT, Nara, Japan 12th March 2003
myGrid                                                          UTOPIA
                                                                              Third party applications

                          LabBook application


      Web Portal
                                                         Ontology         Shared metadata and data
                                                         Services         repositories mIR
                                  Resource
                                  annotations    Semantic-based Services
                                                                                           Inference engines
             Service & resource
             registration &
             discovery
                                                          Personalisation
                                                       e-Science Services             Change & event
Literature                                                                            notification
                            Provenance


               SoapLab
 Databases                                                             Distributed Query
                                                                       Processing
                                             Workflow
                                                        Integration Services
             Analytical
                                  Invited talk, SWFAT, Nara, Japan 12th March 2003
               Tools
 Suite                                               Specialises. All concepts are
 based on DAML-S                                     subclassed from those in the more
                                                     general ontology.


                                                      Contributes concepts to form
           Upper level                                definitions.
            ontology




 Task          Informatics            Molecular                   Publishing   Organisation
ontology         ontology          biology ontology                ontology      ontology




            Web service            Bioinformatics
             ontology                 ontology

                          Invited talk, SWFAT, Nara, Japan 12th March 2003
                                                       1. User selects values from a drop
                                                       down list to create a property based
                                                       description of their required service.
                                                       Values are constrained to provide only
                                                       sensible alternatives.
                                                                      2. Once the user has
                                                                      entered a partial
                                                                      description they submit
                                                                      it for matching. The
                                                                      results are displayed
                                                                      below.


3. The user adds
the operation to
the growing
workflow.


4. The workflow
specification is complete
and ready to match
against those in the
workflow repository.
                            Invited talk, SWFAT, Nara, Japan 12th March 2003
Role of Ontologies in myGrid

                   Service matching
                   and provisioning       Describing & Linking
                                          Provenance records                Composing and validating
                                                                            workflows and service
  Service & resource                                                        compositions & negotiations
  registration &
  discovery



                                         Ontologies                              Change & event
                                                                                 Notification topics
   Resource
   annotations



            Help                                                      Controlling contents of
                                                                      metadata and data
   Knowledge-based guidance
   and recommendation                 Schema mediation
                        Invited talk, SWFAT, Nara, Japan 12th March 2003
  GEODISE                                                      Engineer
                                                                                          (Simon Cox, 2003)
                                                                            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,
                                         Invited talk, industrial strength analysis codes, and distributed computing & data resources
                                           archive     SWFAT, Nara, Japan 12th March 2003
                       (Simon 12th March 2003
Invited talk, SWFAT, Nara, Japan Cox, 2003)
                                                                          Bertram Ludäscher

 GEON: The Geosciences Network
• Large Collaborative NSF/ITR ($11+M over 5 years)




                       Invited talk, SWFAT, Nara, Japan 12th March 2003
                                                                                                                Bertram Ludäscher
SEEK Architecture
                               AM: Analysis and Modeling System                                                                 Execution Environment
EcoGrid
                                                                                                                                      SAS, MATLAB,
provides unified access to                                                             Example of “AP0”                               FORTRAN, etc
                               Analytical Pipeline (AP)
Distributed Data Stores ,
Parameter Ontologies, &          ASx        TS1          ASy           ASz              TS2       ASr
Stored Analyses, and                                                                                                  W
runtime capabilities via the                                                    etc.                                  S
Execution Environment                                                                                                 D
                                                                                                                      L
                                    Parameters w/ Semantics                                                           /
Semantic Mediation System                                                              Data Binding
& Analysis and Modeling                                                                                               U
                                                                                                                      D            ASr
System use WSDL/UDDI to
access services within the     SM: Semantic                                              Semantic Mediation           D
                                                                                                                      I            AP0
EcoGrid, enabling              Mediation                                                   Engine
analytically driven data       System                                                                                               Invasive species
                                                                                                                                            over time
discovery and integration                         Logic Rules
                                                   ECO2-CL                     Query Processing                                  Library of Analysis
SEEK is the combination of                                                                                                       Steps, Pipelines
EcoGrid data resources and                 WSDL/UDDI                               WSDL/UDDI                                     & Results
information services,
coupled with advanced
semantic and modeling                                    ECO2
                                                                                                MC      EML   Dar                  Raw data sets
                                       Parameter            C              C                                              Wrp
capabilities                                                                                                                       wrapped
                                       Ontologies       C              C
                                                                C              C                                                   for integration
                                                       ECO2            TaxOn                   SRB      KNB Species
                                                                                                                          ...
                                                                                                                                   w/ EML, etc.




                SEEK architecture showing the interactions among the EcoGrid, Semantic Mediation System, and Analytical
                Pipelines. An Analytical Pipeline is a graph with nodes that represent computational components and arcs
                                          output parameters Nara, Japan 12 tied to a parameter ontology.
                that represent input and Invited talk, SWFAT, that are explicitly March 2003
                                                                                th
           Biomedical Informatics Research Network
           http://nbirn.net




Mediator View Definition
         Contextualization
  DERIVE CON(Result) wrt. ANATOM.
   protein_distribution(Protein, Organism,Brain_region, Feature_name, Anatom, Value)
   WHERE
   I:protein_label_image[ proteins ->> {Protein}; organism -> Organism;
       anatomical_structures ->>
   {AS:anatomical_structure[name->Anatom]}] ,                % from PROLAB
                     Query results
      NAE:neuro_anatomic_entity[name->Anatom;
                       in context                             % from ANATOM
• provided by the domain expert and mediation engineer
       located_in->>{Brain_region}],
• deductive OO language Invited talk, SWFAT, Nara, Japan 12 March 2003
                            (here: F-logic)          th
                                                                        Bertram Ludäscher
      AS..segments..features[name->Feature_name; value->Value].
SCEC Modeling Environment
                      KNOWLEDGE REPRESENTATION
                             & REASONING
                                   Knowledge Server
                             Knowledge base access, Inference
                                  Translation Services
                              Syntactic & semantic translation


                                  Knowledge Base
                           Ontologies               Pathway Models
 DIGITAL              Curated taxonomies,          Pathway templates,
LIBRARIES            Relations & constraints    Models of simulation codes           KNOWLEDGE
                                                                                     ACQUISITION
Navigation &                                                        Code           Acquisition Interfaces
  Queries                                                FSM        Repositories       Dialog planning,
 Versioning,
                                                               RDM                   Pathway construction
 Topic maps
                                                                                          strategies
                                                                    AWM              Pathway Assembly
 Mediated
                                                                                    Template instantiation,
Collections                                                            SRM            Resource selection,
 Federated
                                                                                      Constraint checking
   access

               Data
                                                                                                              Users
               Collections                           Data & Simulation
                                                     Products

                                          GRID
                                    Pathway Execution
                           Policy, Data ingest, Repository access
                                    Grid Services                                     Pathway
                        Compute & storage management, Security                        Instantiations


      Computing                                                              Storage



                    Invited talk, SWFAT, Nara, Japan 12th March 2003          [Yolanda Gil, USC/ISI]
                                                       [Yolanda Gil, USC/ISI]
   DOCKER: Publishing SHA Code
                                         User specifies:
 Web                                      Types of model parameters
Browser                                   Format of input messages
                                          Documentation
                                          Constraints
                                                                                    AS97
            DOCKER
  User           Model
                                                                    AS97
Interface     Specification                              docs          types

                                                        msg            constrs
                Wrapper
                                                             AS97
Constraint     Generation                                   ontology
Acquisition   (WSDL, PWL)                                                          SCEC
                                                                                 ontologies
                 Invited talk, SWFAT, Nara, Japan 12th March 2003
                                               [Yolanda Gil, USC/ISI]
Automatically Generates KR
Description




         Invited talk, SWFAT, Nara, Japan 12th March 2003
                                                                    [Yolanda Gil, USC/ISI]
 Pegasus:
 Planning for Execution in Grids
 Create workflow to
  create virtual data                                                                                                 Chimera

                                                                                                           (1) Abstract Workflow



  Domain specific and
                                                                                                                   (DAG)
                                                                                                                                                    (18) Results



                                                                                                          (2) Abstract DAG
                                            MCS                                           Current Sate

   generic rules                                                (3) Logical File Names
                                                                        (LFNs)
                                                                                           Generator

                                                                                                            (9) Concrete DAG
                                                                                                                             Request Manager


                                                                                                                                                       (12) DAGMan files
                                            RLS



 Map Workflow onto
                                                                (4) Physical File Names
                                                                        (PFNs)                                                                                             (15) Monitoring
                                                                                                                      (10) Concrete       (11) DAGMan files
                                                                                                                           DAG
                                            MDS
                                                                                                                                                Abstract DAG
                                                                                                                                                  reduction




  Grid resources                                                     Abstract and
                                                                   Concrete Planner
                                                                                                  Concrete Planner
                                                                                                                               (5) Full Abstract DAG     (6) Reduced Abstract DAG




                                             (7) Logical



  System state via Grid
                                          Transformations

                                             (8) Physical
                                         Transformations and
                                        Execution Environment                                      Submit File                    DAGMan
                                             Information             VDL Generator                Generator for                Submission and

   services (MDS, RLS,…)               Transformation
                                          Catalog
                                                                                                   Condor-G                      Monitoring

                                                                                                                               (13) DAG      (14) Log FIles




  Global and local                                                                                                                Condor-G/
                                                                                                                                   DAGMan


   optimization criteria


               Invited talk, SWFAT, Nara, Japan 12th March 2003
                        Grid Interoperability Project
                        Interoperable Resource Broker
                                                  NJS
   Resource Discovery                                                                   Diagram
                                                        Delegates resource check
        Service                                                                         Of Broker
                                                                                        Architecture
                                               Broker                                                            Other
                                                                                                                 Brokers


     Unicore Broker                                                             Globus Broker
                                                  Delegates translation
                                                                                                 Uses to drive
      Lookup                                                                                     MDS search
      resources
                             Uses to
                             Drive MDS                  Translator
                             Search
                  IDB                                                                   Filter
Nodal                                    Ontology engine                                               Hierarchical
                                                                                                       Grid Search
Grid Search
                    Filter
                                         Resource Discovery
                                              Service                                      [Brooke]
                                     Invited talk, SWFAT, Nara, Japan 12th March 2003
                    Closing Remarks




Invited talk, SWFAT, Nara, Japan 12th March 2003
The Grid as a killer app for SW?
 Are Grid applications Semantic Web applications?
 Grid is a real application: the emphasis is on deployment,
  high performance, a large scale, established communities
  of users.
 There isn‟t one Grid, there are collections of Grids for
  communities – might be a more tractable model for the
  Semantic Web.
 It is self-contained, with a well-defined community who
  already work with common tools and standards.
 Aspects of the Semantic Web could be applications of grid
  computing, for example in search, data mining, translation
  and multimedia information retrieval.
 It will stress Semantic Web solutions
 Reliability, scalability, performance, explanation, longevity,
  evolution …
                   Invited talk, SWFAT, Nara, Japan 12th March 2003
Summary
 Middleware enables interoperable use of heterogeneous
  computer systems
 Grid applications involve a wide range of problem driven
  pioneering and provide challenges in Information,
  Knowledge and Collaboration as well as high
  performance computation
 Semantic Web technologies should be applied now for
  machine-processable descriptions and future semantic
  interoperability AT ALL LEVELS of the Grid
    The deeper in the infrastructure the more point-to-point
     consensus can be achieved.
 Aim is to accelerate the scientific process and not just
  scientific computation – this is the reward that will
  motivate Semantic Grid

                    Invited talk, SWFAT, Nara, Japan 12th March 2003
Final remarks
 Is there such a thing as the
  semantic grid?
 There is such a thing as a grid
  with semantics                                      Semantic
                                                                       Web         Grid
                                                                      Services     services
 Overcoming boundaries between                       Web
                                                      Services   Semantic
  communities                                                        Grid
    Mutual suspicion                                        Semantic
                                                                          Grid
                                                               Web
 What can the Semantic Web
  technologies do for the Grid?                                  Semantics for the Grid
 What can the Grid can do for the
                                                                         ??
  Semantic Web?


                   Invited talk, SWFAT, Nara, Japan 12th March 2003
               Acknowledgements
                                              Bertram Ludascher
                                                     Yolanda Gil
                                                 Carl Kesselman
                                                       Ian Foster
                                                     Jim Hendler
                                                 David De Roure
                                                      Simon Cox
                                                   Steve Tuecke
                                                    John Brooke
                                                    Dieter Fensel
                                                    Bill Johnston
Invited talk, SWFAT, Nara, Japan 12th March 2003
http://www.isi.edu/~stefan/SemPGRID/




                   Invited talk, SWFAT, Nara, Japan 12th March 2003
Web Semantics: Science, Services and Agents on the
World Wide Web
                               http://authors.elsevier.com/
                               The Semantic Web
                               Knowledge Technologies
                               Ontology
                               Agents
                               Databases
                               Semantic Grid and Peer-to-Peer Technology
                               Information Retrieval
                               Language Technology
                               Human-Computer Interaction
                               Knowledge Discovery
                               Web Standards
                               eBusiness
                               eCommunity
                               Knowledge Management
                               eLearning
                               Digital Libraries
                               e-Science
               Invited talk, SWFAT, Nara, Japan 12th March 2003
Workshop on e-Services and the
Semantic Web at WWW2003




           Invited talk, SWFAT, Nara, Japan 12th March 2003
 Intl. Semantic Web Conference

 Semantic Web
  Technologies
  for Searching
  and Retrieving
  Scientific Data
  workshop in
  ISWC 2003.




                Invited talk, SWFAT, Nara, Japan 12th March 2003
For More Information
 Grid Book (new edition due soon)
   www.mkp.com/grids
 Survey + research articles
   www.mcs.anl.gov/~foster
 The Globus Project™
   www.globus.org
 Global Grid Forum
   www.gridforum.org
   www.gridforum.org/ogsi-wg
Semantic Grid
   www.semanticgrid.org


                      Invited talk, SWFAT, Nara, Japan 12th March 2003

				
DOCUMENT INFO