Best Practice for new tables with other tables common columns

• amns

  SSCrazy

  Points: 2628

  More actions

  November 2, 2015 at 2:34 am

  #301994

  Hi!

  I have this scenario:

  1 - table A with several columns (around 20), which usually are all filled with not NULL values;

  2 - I need to save more columns related with the same table A, but in this second situation, there are only a few (2,3) columns in common and there are more 2,3 different columns that are filled, leaving the other 20 columns with NULL values;

  My question is:

  3 - What is the best practice or the solution with best performance to add this new columns:

  a. add this new columns to table A, and when rows are inserted, all of the other columns (around 20) will have NULL values but I will have to query one table regarding the same 'subject'

  b. create a new table (B) with the distinct columns and when rows are inserted, all the tables rows will be filled with values, but I will have to query 2 tables regarding the same 'subject'

  Thanks!

• Jason A. Long

  SSC-Insane

  Points: 23786

  More actions

  November 2, 2015 at 8:30 am

  #1837111

  Based on your vague description option B would be the preferred way to go.

• amns

  SSCrazy

  Points: 2628

  More actions

  November 3, 2015 at 6:34 am

  #1837341

  Thanks for your opinion.

  I also think that a second table is better than using the same table keeping sevral rows with NULL values.