Data modeling 101
?
2 likes?906 views
In my customer meetings I am often asked about the importance of Data Modelling. In this presentation we will look at the Top 5 reasons of why Data Modelling is important, including the difference between Data Modelling and Data Architecture.
Data modeling 101
- 1. EMBARCADERO TECHNOLOGIESEMBARCADERO TECHNOLOGIES Data Modeling 101
- 2. EMBARCADERO TECHNOLOGIES Top 5 Reasons Why Modeling is Important? 1. Faster, Error-free Database Analysis & Design 2. Improved Data Quality 3. Standards and Reuse 4. Traceability of Data Structures 5. Communication and Collaboration
- 3. EMBARCADERO TECHNOLOGIES 1. Faster, Error-free Database Analysis & Design
- 4. EMBARCADERO TECHNOLOGIES SQL Dev Environment 5
- 6. EMBARCADERO TECHNOLOGIES 2. Improved Data Quality
- 7. EMBARCADERO TECHNOLOGIES Where does data quality go wrong? 1. Developers building systems with only their requirements in mind 2. Database validation code not enforcing proper business rules 3. Improper data entry by end users
- 8. EMBARCADERO TECHNOLOGIES ? What happens? 1. Models become out-of-date and no one uses them 2. Reports from out-of-date models are useless 3. No understanding what has changed When modeling goes bad…
- 9. EMBARCADERO TECHNOLOGIES ? How do they do it? 1. Use existing models to create new systems 2. Keep models current with databases 3. Ultimately drive changes from the model 4. Communicate designs often 5. Automate the process When modeling works well…
- 10. EMBARCADERO TECHNOLOGIES 3. Standards and Reuse
- 11. EMBARCADERO TECHNOLOGIES Common Challenges with Standards ? No standards in place, i.e., you want it you build it ? Lots of standards in place, i.e., no standard standard in place ? You have a standard(s) in place, but no one knows about them ? Varchar(30) ? Null ? Varchar(20) ? Not Null ? Varchar(max) ? Null ? Varchar(100) ? Not Null Customer Management Finance Marketing Licensing
- 12. EMBARCADERO TECHNOLOGIES Improving Standards ? Important thing is herding the cats and understanding the variances ? Have a central place people can review, edit and define standards ? Think about enterprise level standards, but still maintain "project-level" standards where applicable Impact Analysis / Inheritance / Where Used Repository
- 13. EMBARCADERO TECHNOLOGIES 4. Traceability of Data Structures
- 14. EMBARCADERO TECHNOLOGIES Data Modeling vs. Data Architecture
- 15. EMBARCADERO TECHNOLOGIES Data Modeling ? Smaller in scale, usually departmental ? One off project with a specific use case ? Limited audience
- 16. EMBARCADERO TECHNOLOGIES Data Architecture ? Focused on enterprise level needs ? Holistic across many projects with an eye on reuse ? Wide audience with many levels of abstraction
- 19. EMBARCADERO TECHNOLOGIES Discovery of Undocumented Data gain understanding of data sources through reverse engineering and visualization sales and customer databases Data Architect Marketing Finance & Accounting E-commerce access understanding benefit Reverse Engineering & Infer domains
- 20. EMBARCADERO TECHNOLOGIES 5. Communication and Collaboration
- 21. EMBARCADERO TECHNOLOGIES In a decentralized system the benefits are only available to an individual or small group responsible for the model databases no access no understanding no benefit access understanding benefit Data Architect dbas, dev, business users, analysts, etc
- 22. EMBARCADERO TECHNOLOGIES In a centralized system the benefits can extend to every modeler, developer, dba, and business analyst databases access understanding benefit access understanding benefit dbas, dev, business users, analysts, etc Data Architect
- 23. EMBARCADERO TECHNOLOGIES Levels of abstraction Contextual ? Identify subject areas Conceptual ? Define the meaning of things in the organization Logical ? Describe the logical representation of properties Physical ? Represent the coding language on a specific development platform Instance ? Represent the live system and its whereabouts
- 25. EMBARCADERO TECHNOLOGIES Communicate with Executives
- 26. EMBARCADERO TECHNOLOGIES Why Data Model? Data modeling enables you to: ? Follow principles of object reuse (inclusion) by making structures as applicable across enterprise ? Provide graphical/lexical documentation of the business area under examination ? Permits analysis of data structures in isolation of business processes ? Identify areas of agreement and contention between business and technical parties ? Evolve an enterprise data foundation/architecture which underpins any Data Governance, MDM and Data Quality strategy 27
- 27. EMBARCADERO TECHNOLOGIES Additional Resources ? http://docs.embarcadero.com/ – Read Me file – Quick Start Guide (Includes Installation and Tutorial Information) – User Guide – New Features Guide ? http://edn.embarcadero.com/ – White Papers – Articles – Blogs – Videos – On-line Events
- 28. EMBARCADERO TECHNOLOGIES Technical Support ? Contact Details – Form: http://support.embarcadero.com/forms/ – Phone: +44 (0) 1628 684499