Re: Batch Billing Performance.

pgreenfi@ssax.com wrote:

>
> The SQL optimizer takes many aspects of your database into consideration
> when deciding how to access the data.  The key to this statement is "your
> data".  It does not matter whether you have "out of the box" BPCS or not.
> The optimizer considers your data.  It also depends heavily on the level of
> OS/400 and the database PTF's you have loaded on your system.

Peter I appreciate your reply.
I am extremely interested in your point that the optimiser considers your data.

I take this to mean the number of records in a file.
You cannot surely mean that I use product code XXX and you use product code
YYY.

Assuming file size is the issue.  SQL works fine if you have only a few
thousand records in your files.
For this trivial case of small data files one could have coded (in pre SQL
days)
every file as sequential read and still got some sort of system.
This is of course ridiculous which is why ISAM (Indexed sequential Access
Method) was invented.

But surely that is the point.  Users have large files with lots of data. You
cannot make assumptions that
some files will be small so that an index is not required. And what is the
problem of having indexes
over small files.

On the other hand to have too many indexes so that the data can be accessed in
any way possible is also
out of the question, but that is exactly what UNCONTROLLED use of SQL ensures.

What I am saying is that there must be a judgment and a balance on every
logical file created and
the same applies with every SQL statement that is coded otherwise you end up
with a system that needs
neverending care and attention as to with index needs to be created next.

+---
| This is the BPCS Users Mailing List!
| To submit a new message, send your mail to BPCS-L@midrange.com.
| To subscribe to this list send email to BPCS-L-SUB@midrange.com.
| To unsubscribe from this list send email to BPCS-L-UNSUB@midrange.com.
| Questions should be directed to the list owner: dasmussen@aol.com
+---