You may also find that organizing the data into the order that it is most likely to be retrieved (lastname order for customers, for instance) or in the order of the index that is most often used, can really help on a network. (Like a clustered index in SQL).

You can do this say with a customer table by setting the index to LASTNAME and then doing a COPY TO a new table. This will copy the file in INDEX order and then you can do the reindexing.

If the table is in a dbc you might want to do something like
COPY TO TEMP (not as a free table)
DELETE ALL
PACK
APPEND FROM TEMP

This puts the data into index order and makes retrieve much faster. You would only need to do this once a month or so.



>Hello,
> I actually just deleted all the tags last night because we had this speed issue for little over a week now. I was saying may be I need to modify the reindexing program instead of issuing a reindex command I am going to delete all the tags for each table and recreate all the tags each night. I guess that will be a good idea. thanks for all your inputs.
>sherry