Transcription
Oracle Spatial and GraphOverview of New Graph Features
"THE FOLLOWING IS INTENDED TO OUTLINE OUR GENERAL PRODUCTDIRECTION. IT IS INTENDED FOR INFORMATION PURPOSES ONLY, ANDMAY NOT BE INCORPORATED INTO ANY CONTRACT. IT IS NOT ACOMMITMENT TO DELIVER ANY MATERIAL, CODE, OR FUNCTIONALITY,AND SHOULD NOT BE RELIED UPON IN MAKING PURCHASING DECISION.THE DEVELOPMENT, RELEASE, AND TIMING OF ANY FEATURES ORFUNCTIONALITY DESCRIBED FOR ORACLE'S PRODUCTS REMAINS ATTHE SOLE DISCRETION OF ORACLE."2Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Program Agenda Background: Graph Concepts Use Cases Newest Graph Features in Oracle Database 12c Performance, Scalability, Manageability3Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Oracle Spatial and GraphMature, Proven Graph Database CapabilitiesGraph Features Network Data Model graph W3C RDF Semantic graph4Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Network Data Model:Graph model to represent physical and logical networksKey Features Explicitly stores and maintainsconnectivity Attributes at link and node level Java API to perform Analysis in memory Multiple Cost Support in Path Analysis Traveling salesman, spanning tree,shortest path, sub-path, within cost,nearest neighbors5Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
GRAPH CONCEPTSAPPLIED TO THE ENTERPRISE6Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
RDF Graph Use CasesSemanticMetadata LayerText Mining &Entity Analytics Unified content metadatafor federated resources Validate semantic andstructural consistency Find related content &relations by navigatingconnected entities “Reason” across entitiesSocial MediaAnalysis7 Analyze social relationsusing curated metadata- Blogs, wikis, tweets,video- Calendars, IM, voiceCopyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Modeling Entity Relationships as graph The basic unit of information (fact) is represented as subject,predicate, object “triple” Originally created to encode metadata such as ‘author’, ‘date’, etc.for web resources. Recently, it has become popular to relate things in the real-worldsuch as people, places, concepts etc. Triples together form a “graph”, connecting pieces of datasubjectpredicateobject:John8Copyright 2013, Oracle and/or its affiliates. All rights reserved.:Mary:friendOf:ageInsert Information Protection Policy Classification from Slide 12 of the corporate presentation template“32” xsd:decimal
Relational to Graph ModelingPathologicalChangeNeuronhasis located PathologicalAgentinhibitsinhibitsReceptoris located in9Copyright 2013, Oracle and/or its affiliates. All rights reserved.inhibitsDrugChannelCourtesy: SenseLab, Yale UniversityInsert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Enterprise Metadata ModelingMid-Tier ServerGraph Metadata MappingApplication 2Application 1Application 3SPARQL BenefitsShared Ontologies– Existing relational data stays in placeand corresponding applications do notneed to changeSQLRDF GraphInventory GraphSales Graph– Use of virtual mapping eliminatessynchronization issuesHR SchemaInventory SchemaSales SchemaHR DatabaseInventory DatabaseSales Database– Common vocabulary helps with dataintegration issuesDatabase Server10Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Linked Data in EnterpriseAccess & Presentation LayerIndexSemantic Graph model(W3C RDF Metadata Model)Data ServersEvent ServerHadoop ApplianceContent MgmtBI ServerData WarehouseData Sources / TypesHuman SourcedMachine Generated Data11Copyright 2013, Oracle and/or its affiliates. All rights reserved.Social MediaInformationSubscription ServicesInsert Information Protection Policy Classification from Slide 12 of the corporate presentation templateTransaction Systems
Benefits of RDF Graph Metadata Management Open, flexible, unifying enterprise metadata model Incorporates ALL business content (RDBMS, ECM, Big Data, events) Highly expressive rules for managing data policies and governance Enables users to link, analyze, and discover information horizontally, asshared services, across a variety of enterprise and Web sources12Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
USE CASES13Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Industries Have Already Adopted the ConceptIndustries Life Sciences Finance Media Networks & Communications Defense & Intelligence Public Sector14Copyright 2013, Oracle and/or its affiliates. All rights reserved.Hutchinson3G AustriaInsert Information Protection Policy Classification from Slide 12 of the corporate presentation templateThomson Reuters
Eli Lilly and CompanyOracle Spatial and Graph: RDF Graph Metadata RepositoryObjectives Unified vocabulary for scientificinvestigation Easier, more complete investigations“[This technologyP] provides improved insightinto our business by bringing together relatedinformation from diverse data sources,”J. Phil BrooksInformation Consultant, Eli Lilly and CompanySolution Integrate patient records, chemicalstructures, biological sequences &pathways, images, scientific papersP View related data as a graph Traverse graphs to discover relationships,search for a term, or browse ontologies15Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Allied Nation Intelligence ServiceOracle Spatial and Graph: Social AnalysisObjectivesBenefits Profile suspects through telephone, email Standards-based tools: W3C RDF & SPARQLand social network communications Produce “data products” for analysts Semantic tagging for 600 TB / 10b triples graphSolution New discovery on 100 million triples / month RDF Graph modeling of the social network: Top-secret , compartmented security for data Find & label “same-as” relationshipspeople, groups and places of interest Inferencing & graph analytics discoverrelationships among individuals & meaningof pseudonyms, aliases, codes, terminology16Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Cisco WebEx SocialGraph for Enterprise CollaborationObjectivesBenefits Social connectivity and collaboration Unifies metadata model - forum, blog, wiki, etc.through semantic enablement Connect knowledge silos Tagging media documents, pictures, blogs, etc.Solutionto user-defined and/or enterprise vocabularies. Validates tag semantic/structural consistency Persistent unified graph metadata model Concepts tagged with unique meaning Find related content & groups bynavigating connected entities,recommendations17Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Novartis Institutes for BioMedical Research (NIBR)Project MetastoreInsert your companylogo hereBRIEF ORGANIZATIONAL OVERVIEWNIBR is the global pharmaceutical organization for Novartiscommitted to discovering innovative medicines to treatdiseases with high unmet medical need6000 scientists, physicians, business professionals worldwideBUSINESS CHALLENGES / OPPORTUNITIESScientific knowledge portal does not provide ability to: Link database information on genes, proteins, metabolicpathways, compounds, ligands, etc. to original sources. Increase productivity for accessing, sharing, searching,navigating, cross-linking, analyzing internal /external dataSOLUTIONProvide a semantic integration layer on existing relational tables: Rich domain-specific terminology (biology, chemistry andmedicine) containing 1.6 M terms Terminology Hub: 8 GB of referential data that crossreferences between data repositories.18Copyright 2013, Oracle and/or its affiliates. All rights reserved.BUSINESS BENEFITS REALIZED BY ORACLE SOLUTION Performance: met design goals for comparable performanceto legacy application with enhanced usability Better Analysis: enables discovery of unknown relationshipsbased on the meaning (the semantics) of the data Flexible Data Modeling: supports discovery and allows easychanges to incorporate new kinds of data and relationships Manageability: RDF triple store benefits from OracleDatabase functionality: backup, security, replication,P.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
RDF GRAPH DATABASEFEATURES19Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Oracle Database 12c Spatial and Graph orm &Load, QueryApplications &Modeling Tools& InferenceAnalysis ToolsR2RMLProtégé for OracleJena for OracleRSS, emailOther DataFormatsSesame for OracleNLP Extraction(partners) RDF/OWL DataManagement SQL & SPARQL Query OWL Inferencing Semantic Rules Scalability & Security Semantic IndexingDataSources20Copyright 2013, Oracle and/or its affiliates. All rights reserved. Java, HTTP access JSON output Graph visualization(Cytoscape) Map (GIS)Visualization Oracle BusinessIntelligence (OBIEE) Oracle BPMOracle Database 12cInsert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Oracle Database 12c RDF Triple StoreLeverages Oracle Manageability: RAC & Exadata scalability Compression & partitioning SQL*Loader direct path load Parallel load, inference, query High Availability Triple-level label security Ladder based inference Choice of SPARQL, SQL, or Java Native inference engine Enterprise Manager21Copyright 2013, Oracle and/or its affiliates. All rights reserved.Load /Storage Native RDF graph data store Manages tens of billions of triples Relational to RDF mappingQuery SPARQL-Jena/Joseki, Sesame SQL/graph query, B-tree indexing Ontology assisted SQL queryReasoningAnalytics RDFS, OWL2 RL, EL , SKOSUser-defined rulesIncremental, parallel reasoningUser-defined inferencingPlug-in architecture Semantic indexing framework Integration with OBIEE, Oracle REnterprise Oracle Data MiningInsert Information Protection Policy Classification from Slide 12 of the corporate presentation template
SPARQL and “SPARQL in SQL” ArchitectureHTTPJavaStandard SPARQL EndpointEnhanced with query management controlJena APIJena AdapterSesame APISesame AdapterSPARQL-to-SQLTranslation LogicSQL22SEM MATCHrewritable table functionCopyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
NEWEST GRAPH FEATURES INORACLE SPATIAL AND GRAPHFOR ORACLE DATABASE 12C23Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
SPARQL 1.1 Query Support 2440 new query functions/operators: IF, COALESCE, STRBEFORE, REPLACE, ABS,Aggregates: COUNT, SUM, MIN, MAX, AVG, GROUP CONCAT, SAMPLESubqueriesValue Assignment: BIND, GROUP BY Expressions, SELECT ExpressionsNegation: NOT EXISTS, MINUSCopyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
RDB2RDF: Modeling Relational Data as a GraphRelational to RDF Modeling W3C Standard Specification Oracle Spatial and Graph 12c canrepresent relational schema as graph view Integrate content from distributed sources Federate distributed databases Apply SPARQL queries on tables, views,SQL query results No duplication of data and storage25Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
SPARQL 1.1 UpdateCapabilities of SPARQL Update Insert triples into an RDF Graph Delete triples from an RDF Graph Load an RDF Graph Clear an RDF Graph Create a new RDF Graph Drop an RDF Graph Copy, move or add the content of one RDF Graph to another Perform a group of update operations as a single action26Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
SPARQL 1.1 Property PathsA property path is a possible route through a graph between two graph nodesW3C Property Paths Answers question, “Yes or No: does apath exist from Tim to Sam?”RayWorks withMary Extends matching of a triple pattern toTimany length path A more succinct way to write parts ofbasic graph patterns27Copyright 2013, Oracle and/or its affiliates. All rights reserved.Works withManagesJohnWorks withWorks withManagesBobWorks withInsert Information Protection Policy Classification from Slide 12 of the corporate presentation templateAliceSam
Oracle Spatial and Graph Inference Native OWL 2 EL inference support User defined inferencing Allows generation of new RDF resources Temporal reasoning, Spatial reasoning Web service callouts Ladder Based Inference Fine grained security for inference graph Performance optimization for user defined rules Integration with TrOWL*, an external OWL 2 reasoner TrOWL is a transformation based, tractable reasoner for OWL 228Copyright 2013, Oracle and/or its affiliates. All rights reserved.* http://trowl.eu/Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Jena and Sesame AdaptersPreconfigured, low query cost, ease of install & management Easy-to-use Java APIs to access Oracle database A standard-compliant SPARQL web service endpoint Data loading (RDF/XML, N-TRIPLES, N-QUADS, TriG ,Turtle) w/ long literals JSON output Oracle-specific extensions for query execution control and management Integration with OBIEE, RDF browser29Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Graph Support on Oracle NoSQLAvailable on Oracle NoSQL Database (Enterprise Edition)Graph Feature for NoSQL RDF Graph support in Oracle NoSQL DatabaseEnterprise Edition High performance Key Value store Standard access to graph data: SPARQL 1.1 Jena & Joseki SPARQL endpoint Web Services Massive horizontal scalability –petabytes of triples Support for World Wide Web Consortium(W3C) Semantic Web standards30Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
In-Database Graph and Predictive AnalyticsOracle Advanced AnalyticsProblem ClassificationSample ProblemAnomaly DetectionGiven demographic data about a set of customers, identifycustomer purchasing behavior that is significantly differentfrom the normAssociation RulesFind the items that tend to be purchased together andspecify their relationship – market basket analysisClusteringSegment demographic data into clusters and rank theprobability that an individual will belong to a given clusterFeature ExtractionGiven demographic data about a set of customers, group theattributes into general characteristics of the customersF1 F2 F3 F431Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Reporting RDF Data with Oracle BI EE Powerful BI dashboards– Visually appealing– 100% thin client Across all styles of analysis– R-OLAP, M-OLAP, Scorecards,Reporting, Collaboration, Actions Across all data sources– Federated data access– Share, collaborate, & publish32Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Performing Statistical Graph AnalyticsOracle R Enterprise Open source language Statistical computing and chart for graph data Produces publication quality plots Highly extensible with open source R packages33Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
PERFORMANCE, SCALABILITY,MANAGEABILITY34Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Performance and Scalability Scales to 100s of billions of triples (petabytes) and more- Scales linearly with Oracle database and hardware- No limitations as with other in-memory approaches Fast loading of triples– Incremental and bulk loading Parallelism is exploited– Load, Query, Inference Comparable to or faster than competing RDF graph databases35Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Manageability of RDF Semantic GraphIntegration with Oracle Database 11g/12c utilities and toolsIngest / Replicate /RecoverTune / AnalyzeManageBulk load:Tune load/ query/ inference: Control query execution: Apache Jena bulk loader Oracle external tables & SQL*Loader (Direct Path)w/ PL/SQL Bulk Load API Parallelism Btree indexing triple/quad Typed literals indexing SPARQL query hints Statistics gathering Dynamic SamplingReplicate & recover: Data Guard: physical standby Data Pump: staging tables Recovery Manager: RMAN36Copyright 2013, Oracle and/or its affiliates. All rights reserved.Analyze performance: Enterprise Manager: viewoptimizer plans, monitorexecution / resource usage in database & Jena clientCreate & monitor graphw/ SQL Developer: Semantic Network Models, virtual models Btree indexes Rule bases Entailments Security data labels Semantic index policiesInsert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Summary Oracle brings enterprise-class RDF semantic graph data management Scalable, Secure, and High Performance: load, query, inference features Supports W3C Semantic Standards Works with structured and unstructured data Enterprise-class Oracle tools can now mine insight from semantic data– OBIEE– Oracle Data Mining– Oracle R Enterprise Graph DB now available on Oracle NoSQL Database EE37Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
For More InformationOracle RDFXavier.Lopez@oracle.comoracle.com38Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
39Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
40Copyright 2013, Oracle and/or its affiliates. All rights reserved.Insert Information Protection Policy Classification from Slide 12 of the corporate presentation template
Oracle Database 12c. Oracle Database 12c RDF Triple Store LeveragesOracle Manageability: . Enterprise Manager RDFS, OWL2 RL, EL , SKOS User-defined rules . OBIEE, Oracle R Enterprise Oracle Data Mining Analytics. SPARQL and "SPARQL in SQL" Architecture Jena API Jena Adapter Sesame API Sesame Adapter Standard SPARQL .
