i've 3 tables where 4 million rows deleted , to regain the pseudo deleted rows etc i think there is a need to reorg the table

can any one let me know if my ablve & below understanding is correct

per the info in ibm link i need to run reorgchk if the out put in the last column reorg shows a (*) star mark i need to go for reorg for that table.

my question is after running reorg next time if i run reorg the orgchk again showing * .

is there any other parameter / information i can check to know if the orgchk have helped. ie before & after running the orgchk to justify if it helped?

thanka

Using reorg has 3 main advantages. First it will free up space that is no longer being used, It will fix all overflows (rows that had to be moved to another page due to data changes), and lastly it can order (cluster) the data based on some index.

If you have runstats done before and after the reorg, you should see differences in syscat.tables for NPAGES, FPAGES, and OVERFLOW columns for the table. For the clustering, you should see improvements in some queries.

Andy

This article may be helpful: getting a handle on reorg

...if you are discussing DB2 for mainframes.

thanks for the reply -

i'm wondering if we take a backup restore the same db will take care of the above 3..? (may be my understanding is wrong..)....
if it wont take care of the above why ?

Is there any other type?

No, backup/restore will not do the same as reorg. All backup does is copy the database pages as they exist and write them to the backup image. All restore does is take that database pages as the exist in the backup image and lay them down in a database. Nothing really changes structurally with the pages. That is what reorg does.

Andy

A backup/restore does not change anything in the data layout on the pages. It just creates/restores a copy. So if you take a backup before a REORG, you will restore the non-reorg'd database again.