A Linked Fusion of Things, Services, and Data to Support a Collaborative Data Management Facility
0 likes322 views
This document discusses linking together data, services, and things to support a collaborative data management facility for a wind characterization scientific study. It proposes using semantic technologies like REST, Linked Open Data, Linked Services, and concepts from the Internet of Things. The approach aims to seamlessly link the study's instruments, services, activities, and data to gain insights and make everything accessible and discoverable for researchers. It leverages existing open-source and commercial tools and illustrates how a linked knowledge environment can support search and discovery across components for both facility operations and scientists using the study results.
A Linked Fusion of Things, Services, and Data to Support a Collaborative Data Management Facility
- 1. A Linked Fusion of Things, Services, and Data to Support a Collaborative Data Management Facility EG STEPHAN, TO ELSETHAGEN, AS WYNNE, C SIVARAMAN, MC MACDUFF, LK BERG, WJ SHAW Pacific Northwest National Laboratory Richland, WA October 21, 2013 1
- 2. Presentation Overview ! Purpose ! Brief History of Linked Approaches ! Wind Characterization Scientific Study Use Case ! The Benefits of “Thinking Linked” October 21, 2013 2
- 3. Purpose ! Illustrate a pragmatic approach to use of semantic technologies and approaches to seamlessly link together a scientific research study using concepts from: ! ! ! ! REST Linked Open Data Linked Services Internet of Things ! Gain additional insights to advances and leading edge approaches in the Internet of Things community October 21, 2013 3
- 4. A Few Quick Definitions ! Things – any device (electronic or non-electronic) pertinent to the study. ! Services – automated machine processes and tracked human activities that may impact things or data. ! Data – physical streams, file containers, and directory trees representing data resulting from the study. ! Knowledge – any metadata, thesauri terms or RDF vocabulary used to describe syntactic data structures or semantically represent things, services or data. October 21, 2013 4
- 5. State of Linked Technology A Decade Ago: An Application Developers Perspective ! The world of linked communities emerge from different focus areas. ! IoT community focused on innovating approaches to connect things to the internet. ! Web community inter-connected information and media. ! Semantic web community focused solely on portraying knowledge as a web of interrelated terms and concepts. ! Service-oriented approaches solved business problems ! Developers tended to favor a particular approach ! In legacy operational scenarios how do you overcome information silos? ! Production mindset versus prototype functional dependability ! How do you protect your long term investments at the same time? October 21, 2013 5
- 6. Representational State Transfer (ReST) ! It’s a design principle ! Anything of perceived importance is a “resource”: ! E.g. data, knowledge, services ! Any resource is addressable by Uniform Resource Identifier (URI) ! Any resource can linked to other resources. ! Using this development model: the approach of linking resources emerged ! API relies on simple protocol stack based on HTTP. Richardson, Leonard, and Sam Ruby. RESTful web services. O'Reilly, 2008. J. Clerk Maxwell, A TreaFse on Electricity and MagneFsm, 3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68–73. October 21, 2013 6
- 7. Linked Open Data ! Is also a design principle with many overlapping characteristics that REST supports. ! Its main focus to make data: ! ! ! ! Addressable Accessible Easily integrated with other data Easy integrated with knowledge ! W3C introduced 5 star rating system for users to describe the level of detail data exhibited. Heath, Tom, and ChrisFan Bizer. "Linked data: Evolving the web into a global data space." Synthesis lectures on the semanFc web: theory and technology 1.1 (2011): 1-‐136. October 21, 2013 7
- 8. Emerging Community Vocabularies ! The good news of the semantic web is knowledge can be literally described in any way ! – without some constraints from a scientific study perspective this can also be the bad news. ! Fortunately there are well established and emerging RDF vocabulary communities willing to share ! By using community vocabularies we allow: ! The study semantics to emerge and change over time ! Manually align these semantics to community vocabularies so that evolving scientific knowledge can be navigated and cross referenced. ! From a domain perspective this can provide a place for scientists to contribute and share own domain community thesauri and vocabularies October 21, 2013 8
- 9. Example Vocabularies Purpose Type Source Data CitaFons Vocabulary Dublin Core People InteracFons Vocabulary Friend of a Friend (FOAF) OrganizaFon Vocabulary W3C ORG Services, Things Vocabulary Minimum Service Model Services, Things Vocabulary Universal Services DescripFon Language Vocabulary Management Vocabulary VOAF Vocabulary Stability Vocabulary VS Climate and ForecasFng Domain CF Provenance Vocabulary PROV October 21, 2013 9
- 10. Linked Services ! Developed as the next wave of service technology based on three ideas: ! Base it on on similar linked open data principles ! Make service annotations available in the web of data ! Create services to support the web of data ! Linked Services made services: ! Addressable ! Accessible ! Self-describing with knowledge using a common RDF vocabulary that: ! generalized the concept of service ! acknowledged varying technical implementations. B Pedrinaci, Carlos, and John Domingue. "Toward the Next Wave of Services: Linked Services for the Web of Data." J. ucs 16.13 (2010): 1694-‐1719. October 21, 2013 10
- 11. Internet of Things ! Interest in semantically linking electronic and non-electronic devices to the Web of Data. ! Recently, the Linked Services community made the connection to the IoT community referencing things like services using the same RDF vocabulary. Mandler, Benny, Fabio Antonelli, Robert Kleinfeld, Carlos Pedrinaci, David Carrera, Alessio Guglioca, Daniel Schreckling et al. "COMPOSE-‐-‐A Journey from the Internet of Things to the Internet of Services." In Advanced InformaFon Networking and ApplicaFons Workshops (WAINA), 2013 27th InternaFonal Conference on, pp. 1217-‐ 1222. IEEE, 2013. October 21, 2013 11
- 12. Reference Facility for Offshore Renewable Energy Use Case ! Make the planned instrument measurements taken at a reference facility accessible to a range of users, which includes atmospheric researchers and energy planners who perform wind energy studies. NREL Artist’s Conception October 21, 2013 12
- 13. Use Case Goals ! Rely on off the shelf proven and reliable operational technology ! Support management and production of high integrity research study: ! ! ! ! Instrument mentors Operations Maintenance DMF administrators ! Provide consumer researchers with accessible, searchable, and discoverable credible results October 21, 2013 13
- 14. Managing Results from the Scientific Study October 21, 2013 14
- 15. Leveraging Operational Off the Shelf Tools Capability Government Open Source Commercial CollecFon and Transfer Ingest REST Processing REST Data Management Interfaces Knowledge Management File APIs, REST AcFvity Tracking* Access File APIs (FTP) REST, SQL, Web REST SemanFc Web *AcFvity tracking includes automated processes and human acFviFes supporFng: change management, relevant events, quality assurance, problem reporFng October 21, 2013 15
- 16. Atmosphere Radiation Measurement ARM is a global observaFon network—and associated data infrastructure— providing diverse measurements to the climate research community. Deploy measurement instrumentaFon Provide infrastructure to collect, process, preserve, and discover data Operate with excellence October 21, 2013 Provide high-‐quality data to support global climate research 16
- 17. Challenges ! Operational quality of existing components needed to be preserved while at the same time meeting the needs of the RFORE DMF. ! In some cases current state of the art had to be scaled down with new components to form a new solution. ! E.g. Activity Tracking ! Solution needed to make sure knowledge about instruments, services, and measurement data were accessible to both operations and scientists relying on the study. October 21, 2013 17
- 18. Thinking Linked: Linked Knowledge Environment (LiKE) ! Common Address Layer ! Based on linked design principles ! Every resource will have a unique and meaningful name (URI) ! Address mappers will be required to resolve local identifiers to URI. ! Cataloging ! Although every resource will be uniquely identifiable in a meaningful way, additional referential technical information is required to help describe each in greater detail. ! Catalogs are regarded as the linchpin of the linked fusion concept since they possess the logic for users and machines to search, discover, and reason semantic and syntactic information relating to things, services, and data. ! Knowledge Synchronized to Support: ! ! ! ! Facility, instruments, and measurements Activities and services Data Vocabulary management October 21, 2013 18
- 19. Where Things, Services, Data, Knowledge Live Things Data, Services, Things Services Services Data Knowledge October 21, 2013 19
- 20. Search and Discovery Services ! Relies on a seamlessly linked knowledge model for registered services, things, and data ! Supporting cross component searches ! Support for operational staff conducting the study and researchers relying on data ! Reliance on community vocabularies What Comes First, the OWL or the Bean? . Eric G Stephan, Todd O Elsethagen, KersFn Kleese van Dam, Laura Riihimaki. figshare. hcp://dx.doi.org/10.6084/m9.figshare.790738 October 21, 2013 20
- 21. Future Things of Interest ! IoT Vocabularies ! Quality assurance and provenance of things ! Soft error detection in IoT October 21, 2013 21
- 22. Questions? ! Email: eric.stephan@pnnl.gov October 21, 2013 22