Survey on NoSQL integration
Download as PPTX, PDF1 like969 views
This document summarizes several papers on integrating NoSQL databases. It discusses different approaches to integration such as global-as-view and local-as-view. Most solutions presented use a global-as-view approach and expose a unified REST API. The papers cover domains like healthcare, biodiversity, and graph matching. The BigDAWG system is highlighted as the most complete approach, federating access across different database models in a scalable way.
Survey on NoSQL integration
- 1. A survey on NoSQL database integration Luiz Henrique Zambom Santana Prof. Dr. Ronaldo dos Santos Mello Profa. Dra. Carina Dorneles
- 2. Agenda • Background • NoSQL • Global vs. Local • Model • Related Works • Comparison • Taxonomy • Conclusions
- 4. Background: NoSQL Sadalage e Fowler, 2012 (http://martinfowler.com/books/nosql.html) Not only SQL Nathan Marz, 2014 (http://www.slideshare.net/nathanmarz/runaway-complexity-in-big-data-and-a-plan-to-stop-it) Relational databases will be a footnote in history
- 6. Background: Global-as-view Vs. Local-as-view ● GAV ○ mapping from entities in the mediated schema to entities in the original sources ● LAV ○ mapping from entities in the original sources to the mediated schema ● The latter approach requires more sophisticated inferences to resolve a query on the mediated schema, but makes it easier to add new data sources to a (stable) mediated schema.
- 8. Dey, Akon, Alan Fekete, and Uwe Röhm. "Scalable transactions across heterogeneous NoSQL key-value data stores." Proceedings of the VLDB Endowment 6.12 (2013): 1434-1439. • VLDB Endowment • Qualis A1 • Impact Factor 1.568 • Why it is important? • Seminal • Transactions •“Weak” global-as-view
- 9. Zhang, Duo, Benjamin Rubinstein, and Jim Gemmell. "Principled graph matching algorithms for integrating multiple data sources." (2014). • IEEE Transactions on Knowledge and Data Engineering (TKDE) • Qualis A1 • Impact factor 2.067 • Why it is important? • Graph matching algorithms • Entity resolution • Shows that integration is far more complicated in NoSQL applications •Local-as-view
- 10. Da Silva, Daniel L., et al. "A Computational Framework for Integrating and Retrieving Biodiversity Data on a Large Scale." Big Data (BigData Congress), 2014 IEEE International Congress on. IEEE, 2014. • IEEE International Congress on Big Data • No Qualis (yet) • Impact factor • Why it is important? • Integrating and Retrieving Biodiversity Data •Global-as-view •Resembles the Lambda Architecture
- 11. Kiran, V. K., and R. Vijayakumar. "Ontology based data integration of NoSQL datastores." Industrial and Information Systems (ICIIS), 2014 9th International Conference on. IEEE, 2014. • 2014 9th International Conference on Industrial and Information Systems (ICIIS) • Qualis B1 • Why it is important? • Intermediate model • Global-Local-as-view • Information extraction may require sourcing data from multiple data sources, establishing relationship among them and querying across these data sources together.
- 12. Kaur, Karamjit, and Rinkle Rani. "Managing Data in Healthcare Information Systems: Many Models, One Solution." Computer 3 (2015): 52-59. • IEEE Computer • Qualis A1 • Impact fator 1.443 •Global-as-view • Why it is important? • Because healthcare data comes from multiple, vastly different sources, databases must adopt a range of models to process and store it. A polyglot-persistent framework combines relational, graph, and document data models to accommodate information variety
- 13. Duggan, Jennie, et al. "The BigDAWG Polystore System." ACM SIGMOD Record 44.2 (2015): 11- 16. • SIGMOD Record • Qualis A1 • Impact Factor 1.05 • Global-as-view • A polystore architecture designed to unify querying over multiple data models. •“No one size fits all”
- 14. Duggan, Jennie, et al. "The BigDAWG Polystore System." ACM SIGMOD Record 44.2 (2015): 11-16. • Why it is important? • Twitter guys and Stonebraker • Deals with the entire complexity • Introduces the Island abstraction • Model cast between the DBMS
- 15. Taxonomy
- 16. Comparativo Year Main author Summary NoSQL Taxonomy 2013 Dey Transactional access Key/Value Schema unification > Poliglot 2014 Zhang Graph match Graph Schema unification > Unified Language 2014 Da Silva Biodiversity databases integration Document Applicational integration > CAP 2014 Kiran Ontology as canonical model Column-oriented Schema unification > Unified Language 2015 Kaur Medical Virtually any of database Applicational integration > CAP 2015 Duggan BigDAWG Virtually any of database Federation > Indepent access
- 17. Conclusions • The problem is real •Important for many fields • Most of the solutions uses Global-as-View • Most of the solutions exposes a REST API as unified access • Many works cites also SQL and NoSQL integration • Concerns • The solution have to scalable • The solution cannot be difficult to setup • BigDAWG is the most complete approach