table splinting

  • Hello Experts,

    I have a 24 GB database, which has a table candidate_resumes, table size is 20 GB, table has a column CanResumes ( image datatype) which stores doc or pdf files. i guess this column is heaviest in the table. i am thinking to keep that column separate in a table with a PK-FK Relation of candidate_resumes table . And the table of separated column i want to store in a different filegroup.

    need your suggetion as table size reduced by doing this, will it give any improvement in performance. or if any other suggetion u have so please give me to implement. like drawback or benefits

    please help

    thank you

  • Performance improvement would be noticed in any queries against the table that involve SELECT *. You need to consider too how often that column is being queried as if it's going to be quite often, then you're not really saving much on performance as now you have to do a JOIN between the table several times.

    If the image column is not being used that often though, then you'll be better off splitting it into its own table and doing the PK/FK constraint.

  • ya thats the main reason we dont use that column so much.

    bt it holds around 10 gb size of data from 20 gb table size.

    some fix procedure are thr which fetch that column,

    so i guess splitting it wud be gud becos rest of column come in use frequently of that table.

  • i have splited tht column in a separate table bt in the server i have only one drive so i cant make filegroup in a diff drive, bt i have kept both tables in the same drive in the same file group, if ill keep tables in the same drive bt in different filegroup will it help in performance????

Viewing 4 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic. Login to reply