Custom fields schema

[Kevin Benton]

Myk – I’m concerned about some of the numbers I’m seeing from the result
times you posted.  I saw a number of queries at 50+ and assuming that’s in
seconds, to me, that just doesn’t cut it.  Please correct me if I'm wrong.

I need queries to run in under 1 second the first time they're run
consistently.  For example, in Real, on 1 str, you show the first query time
was 2.4 seconds, and the distributed query ran 56.2 seconds!  1 str bzlike
even in Real ran 17.8 seconds and from looking at the ref, it looks like
there was no index available.

>From what I can see, the fields as columns performance was *much* better
than the fields as data performance (myk-result-3.html).  This suggests to
me that this particular implementation is not going to work here.  I'm not
saying that a FAR method can not be developed that will perform within
expectations, but from the data presented, this particular method doesn't
look like it will suit my needs.

I am writing statuscount.cgi - a completely new reporting package for
Bugzilla to show my management the information they need out of Bugzilla to
help them effectively direct development.  I expect this report will be run
about one to five times a week.  This report currently executes eight
distinct queries of which six are complex.  So, if the queries are running
an average of 50+ seconds each first-time, it's going to take on the order
of five to six minutes for the manager to get the information they're
looking for.  That will prevent my managers from using my reports.  On the
other hand, if my report takes less than ten seconds, I will be doing okay.
When I'm reasonably close to being done with this reporting software, I will
release it to the Bugzilla community.  I've already got at least two non-AMD
testers evaluating it for me now.

Granted, I'm assuming that we're migrating to handling all the fields in a
unified way regardless of whether or not they're legacy or new.  Again, as I
said, please correct me if I'm wrong.  I just don't see this particular
implementation as viable.  Again, I really don't care which way we go - FAR
or FAC as long as it is easy to understand, performs within expectations,
and is scalable.  It seems to me that FAC is easier to grasp in my mind, but
if you or anyone else can make a solid case for FAR, great!

---
Kevin Benton
Perl/Bugzilla Developer
Advanced Micro Devices
 
The opinions stated in this communication do not necessarily reflect the
view of Advanced Micro Devices and have not been reviewed by management. 
This communication may contain sensitive and/or confidential and/or
proprietary information.  Distribution of such information is strictly
prohibited without prior consent of Advanced Micro Devices.  This
communication is for the intended recipient(s) only.  If you have received
this communication in error, please notify the sender, then destroy any
remaining copies of this communication.