Building RDF- and Schema-Based Peer-to-Peer Systems / University - PDF document

Building RDF- and Schema-Based Peer-to-Peer Systems / University of Hannover Wolfgang Nej dl Germany L3S

Overview Relevant L3S Proj ect Background � Motivation � Proj ect Background - PADLR, Edutella, et al S chema-Based Peer-to-Peer Networks � Characteristics and Building Blocks � Resource Description Framework (RDF) and RDF S chema � Edutella Query S ervice / RDF Query Exchange Language RDF-QEL � Semantic Web Inferencing � S ubscriptions � Efficient Routing / HyperCuP & S uper-Peers � Integration of new Peers / Clustering � Distributed Query Processing in P2P Networks � Mediation S ummary and Conclusions Wolfgang Nej dl 14/ 10/ 03 2

Motivation Distributed Peer-to-Peer Infrastructures for the S emantic Web S emantic Web Metadata How can we use distributed S tandards for describing (learning) resources in a (learning) resources and users personalized way? Personalized Environments in the Adaptive Web Wolfgang Nej dl 14/ 10/ 03 3

PADLR and Edutella Personalized Access to Distributed Learning Repositories (www.learninglab.de/ english/ proj ects/ padlr.html) Most important (CS) modules - Peer-to-Peer Infrastructure (incl. Clients and Providers) - Courseware Watchdog and Metadata Extraction - Personalization and Personalized Queries based on Metadata - Web-(Learning-) Services PADLR Participants (Stanford, Hannover, Karlsruhe, Stockholm, Uppsala) + Edutella Participants (Vienna, Berlin, Darmstadt, etc.) Wolfgang Nej dl 14/ 10/ 03 4

Edutella: Goal and Approach S pecify and implement a RDF-based meta-data infrastructure for P2P networks Developed as part of the open source peer-to-peer proj ect JXTA edutella.j xta.org > 60 contributors from various institutions Wolfgang Nej dl 14/ 10/ 03 5

Knowledge Web 6 14/ 10/ 03 Heterogeneity Wolfgang Nej dl Languages Dynamics Scalabity EU/FP6 NoE KnowledgeWeb Semantic Web Services

EU/FP6 NoE PROLEARN Working towards � innovative elearning resources � interoperable elearning resources and systems � sustainable elearning infrastructures and processes for S MEs Wolfgang Nej dl 14/ 10/ 03 7

EU/FP6 NoE REWERSE Reasoning on the Web with Rules and S emantics Develop reasoning languages for advanced Web systems Test these languages on adaptive Web systems and Web-based decision support systems Bring these languages to the level of pre-standards S elected applications for proof-of-concept purposes Personalized Web systems Web-based decision support Towards a Bioinformatics S emantic Web Wolfgang Nej dl 14/ 10/ 03 8

Overview Relevant L3S Proj ect Background � Motivation � Proj ect Background - PADLR, Edutella, et al S chema-Based Peer-to-Peer Networks � Characteristics and Building Blocks � Resource Description Framework (RDF) and RDF S chema � Edutella Query S ervice / RDF Query Exchange Language RDF-QEL � Semantic Web Inferencing � S ubscriptions � Efficient Routing / HyperCuP & S uper-Peers � Integration of new Peers / Clustering � Distributed Query Processing in P2P Networks � Mediation S ummary and Conclusions Wolfgang Nej dl 14/ 10/ 03 9

Schema-Based Peer-to-Peer Networks User-definable schemas Database Systems Schema-based P2P Systems S tructured schemas A MOS II A NY RDBMS C HATTY W EB schema- O BJECT G LOBE Query language C ONCEPT B ASE E DUTELLA based TSIMMIS O NTO B ROKER P IAZZA T UKWILA D IRECT C ONNECT fixed G NUTELLA Decentralized control schema/ N APSTER K A Z A A keywords Node autonomy P - GRID P2P Systems Transient peers CAN S elf organization key CHORD local distributed peer-to-peer (system list not complete) Wolfgang Nej dl 14/ 10/ 03 10

Building Blocks Flexible S chema Language � to describe complex and heterogeneous resources in the P2P network Expressive Query Language � to retrieve data from heterogeneous data stores Efficient Network Topology � to allow appropriate routing algorithms Mediation Facilities � to integrate and combine (possibly heterogeneous) information Wolfgang Nej dl 14/ 10/ 03 11

RDF / RDF Schema for Describing Distributed Resources Basic Formalisms for the S emantic Web � URIs to identify resources � Combine resources and annotate resources with attributes, using <S ubj ect, Property, Value> Tuples � Graph as basic model, easy to translate to logic facts � RDFS allows us to define the RDF vocabulary used (classes and attributes), and thus to represent simple semantic models � Possible extensions towards more expressive semantic descriptions, e.g. description logic (DAML+OIL / OWL) Using RDF / RDFS in the P2P context � Distributed annotations for distributed resources � Flexible schema definitions, which can be uniquely identified and combined, as well as extended by additional properties Wolfgang Nej dl 14/ 10/ 03 12

Characterization of Peers using RDFS S chema level � S upporting specific schemas: dc, lom, dcq Property level � S upporting specific properties: dc:subj ect, lom:type, dc:format Property value range � S upported ranges for specific properties, e.g. ccs:dbms for dc:subj ect Property values � S pecific attribute values, e.g. „ exercise“ for lom:type, „ en“ for dc:language Wolfgang Nej dl 14/ 10/ 03 13

RDF-QEL: RDF Query (Exchange) Language Datalog-based Query Exchange Language (RDF-QEL) � RDF QEL1: conj unctive query up to � RDF QEL5: RDF QEL4 (SQL3) + general recursion Edutella consumer see Nejdl et al: „EDUTELLA: A P2P Networking RDF query RDF QEL 1-5 Infrastructure Based on RDF“, WWW 2002 result Datalog-based ECDM � Datalog is used as the internal data model (ECDM: Edutella Common Data Model) and provided as a set of Java classes Local query � RDF is used to represent the queries transmitted between the peers repository � Wrappers for ot her RDF query languages (RQL, Edutella Provider TRIPLE, etc.) and XML query languages (like Xpath) Edutella query data flow Wolfgang Nej dl 14/ 10/ 03 14

From Querying to Reasoning World Wide Web Data as Distributed (Web) content + S emantic Web Metadata Distributed and interoperable (RDF) P2P metadata descriptions about: � Content � Relationships between the content � Learner + S emantic Web Inferencing (Logic) Programs and Rules to: � Adapt the content and Content relationships (links) � Infer new metadata Content Metadata = Declarative and Composable Web Relationships S ervices Learner Model Logic Programs siehe auch REWERS E NoE Wolfgang Nej dl 14/ 10/ 03 15

P2P and Semantic Web Inferencing: Edutella as basic infrastructure for ELENA (EU/FP5) Wolfgang Nej dl 14/ 10/ 03 16

Another Possibility: Don‘t query, subscribe S ubscriptions are a good idea, too (get the NYTimes each morning, get new teaching material on P2P topologies … ) Example: S elective Information Dissemination in P2P-DIET Instead of Queries and Answers we need � Profile forwarding � Notification forwarding / Filtering � Advertisement forwarding � Dynamicity of P2P network � storing notifications / rendezvous S ee e.g. Koubarakis et al: S elective Information Dissemination in P2P Networks: Problems and S olutions, S IGMOD Record, S pecial P2P Issue, S eptember 2003 Wolfgang Nej dl 14/ 10/ 03 17

P2P and Efficient Routing How do peer-to-peer networks scale? Requirements: � S ymmetric topology (every node is a root) � Low network diameter (small worlds property, should be O(log n)) � Limited node degrees (number of peer-connections from a node, should be O(log n)) � Load balancing of traffic � Efficient broadcast (receive broadcast messages only once) � Adaptable to dynamic number of peers Wolfgang Nej dl 14/ 10/ 03 18

HyperCuP Peer-to-Peer Topology Details: see e.g. S chlosser, S intek, Decker, Nej dl: „ HyperCuP – S haping Up Peer-to-Peer Networks“ , 2nd Intl. WS on Agents and P2P Computing, 2002 Wolfgang Nej dl 14/ 10/ 03 19

Hypercube Topology Broadcast Algorithm � Annotate messages with the “ dimension” of the peer-to-peer connection, and only forward it along “ higher” dimensions Properties � Network diameter, characteristic path length and number of nodes are O(log b N) � Fault tolerant, vertex-symmetric 6 7 0 2 2 1 1 3 2 0 Step 2 4 5 0 1 1 Step 3 2 2 8 1 0 Step 1 Wolfgang Nej dl 14/ 10/ 03 20

Super-Peer Networks Observation: Peers vary significantly in availability, bandwidth, processing power, etc. Create network backbone from highly available and powerful peers to distribute load better. S ee also Yang, Garcia-Molina: Improving S earch in P2P S ystems, Intl. Conf. on Distributed Computing S ystems, Vienna, 2002, or file sharing networks like KaZaa � Wolfgang Nej dl 14/ 10/ 03 21

Super-Peers and Routing Indices Nejdl et al. Super-Peer-Based Routing and Clustering Strategies for RDF-Based Peer-To-Peer Networks. WWW 2003 Wolfgang Nej dl 14/ 10/ 03 22

Extension to Distributed Query Processing Interleave P2P techniques and query processing � Push abstract query plans through the super peer network � S uper peers pick and expand those parts of the query plan that can be executed locally � On the fly distribution and expansion of query plans S ee Brunkhorst, Dhraief, Kemper, Nej dl, Wiesner: Distributed Queries and Query Optimization in S chema-Based P2P-S ystems, VLDB-P2P- Workshop Wolfgang Nej dl 14/ 10/ 03 23