Hello all
Grateful for any help on this -
I have a table (definition for MAIN below) with a number of indexes
corresponding to various frequently used queries.
A SELECT statement using one of the multi-column indexes is running very
slow if it retrieves data that is not in in the index itself.
So, is it *really* the variable length row lookup that's taking all the
time
here ?
I don't see how. 18000 record is nothing and varchar searches are pretty
fast.
On top of that, you first search is on the date which would be extremelly
fast and narrow the next searches to many less
Hi Phil,
I suspect you have already answered your own question!
Warning: high ignorance factor - am currently researching indexes for myself - if I
can't figure it out for myself, you
can help me with that later...
When the SELECT can, it will answer a query from a scan-of/data-retrieval-from
I'm wondering whether this has something to do with the following words of
wisdom in the reference manual:
For tables that changes a lot you should try to avoid all VARCHAR or BLOB
columns. You will get dynamic row length as soon as you are using a single
VARCHAR or BLOB columns.
but I