Notion of Equivalent Relational Databases (RDBs)?

  • Thread starter Thread starter WWGD
  • Start date Start date
  • Tags Tags
    Equivalent
Click For Summary
SUMMARY

The discussion centers on the concept of determining the equivalence of Relational Databases (RDBs). Key methods for assessing equivalence include comparing database snapshots using CRC (Cyclic Redundancy Check), analyzing schema structures at various levels, and ensuring proper data mapping during migrations. The conversation highlights the importance of tools and methodologies, such as ETL (Extract, Transform, Load) processes, particularly in contexts like banking mergers where standardized schemas are crucial. The need for comprehensive industry-specific schemas, as provided by companies like IBM, is emphasized for successful database integration.

PREREQUISITES
  • Understanding of Relational Database concepts and terminology
  • Familiarity with Cyclic Redundancy Check (CRC) for data integrity
  • Knowledge of ETL (Extract, Transform, Load) processes
  • Basic principles of schema design and relational algebra
NEXT STEPS
  • Research methods for comparing database schemas using tools like SchemaSpy
  • Learn about data migration strategies and best practices in ETL processes
  • Explore the use of relational algebra for database equivalence analysis
  • Investigate industry-specific schemas provided by vendors like IBM for banking
USEFUL FOR

Database administrators, data engineers, and IT professionals involved in database migration and integration projects, particularly in sectors like banking and finance.

WWGD
Science Advisor
Homework Helper
Messages
7,801
Reaction score
13,105
Hi all,
RDB:=Relational Database
I need for this to determine if there is a standard, " reasonable" notion for when two RDBs are considered to be equivalent ( other than the obvious relabeling and reordering of fields). OF course, we assume that the Functional Dependency graphs are unique/fixed, up to graph isomorphism. I did a search, but it seems like anything useful is behind a pay wall.
 
Last edited:
Computer science news on Phys.org
This is a hard question to answer.

On the lowest level, comparing database saves using CRC or something similar would be one way. It would basically say if two snapshots of a database were identical.

At a higher level, using the whole database schema comparing schema to schema, table to table, indexes to indexes at the user level would be another. It assumes the names of schema, tables, columns... are the same. This might show two databases implemented on different vendor platform are the same.

At the highest level, comparing schema design to design and how things are modeled to see if any data is left out by one of the databases where names of things are different but still represent the same things.

Usually what happens is a company say a bank wants to migrate to a new database system. They would insure that existing data is properly mapped into the new schema and that nothing needed is lost in the migration.

Companies like IBM, would offer comprehensive industry specific schemas for banking that banks would use if a bank merger was planned so they would now have a common database for the new organization. The vendor would also supply the necessary ETL tools to migrate the data and its relationships over to th new system.
 
Last edited:
  • Like
Likes   Reactions: WWGD
Thanks, Jedi, how about using Relational Algebra? And, what is CRC?
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
1K
  • · Replies 95 ·
4
Replies
95
Views
8K
  • · Replies 110 ·
4
Replies
110
Views
20K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 26 ·
Replies
26
Views
6K
  • · Replies 1 ·
Replies
1
Views
400
  • · Replies 21 ·
Replies
21
Views
7K
  • · Replies 51 ·
2
Replies
51
Views
5K
  • · Replies 37 ·
2
Replies
37
Views
5K
  • · Replies 335 ·
12
Replies
335
Views
41K