This prompted me to check my indexing and it was
mostly OK, but I found that adding an index on RateUsername improves the speed
of this query by a factor of about 4. I think we're back to pretty much OK
now.
If it gets bad, I'll ask for help with my
SQL.
-- Phil Holmes
----- Original Message -----
Sent: Tuesday, July 10, 2012 6:36
PM
Subject: Re: Regression test rater
Also if you want to give more details about your database, e.g., which SQL
implementation, how the tables are organised, and the code you tried, maybe
someone here can improve it.
On Jul 10, 2012 4:54 PM, "Trevor Daniels" < address@hidden> wrote:
Phil Holmes wrote Tuesday, July 10, 2012 12:03
PM
> Kind-of fixed. The way the files are presented is aimed
at ensuring no-one > rates a regtest more than once, and that they get
the least-rated files > presented to them in a random order. The
only way I seem to be able to get > this to work is with nested SQL
statements, and this is quite slow. The > alternative would be to make
it simpler, so that users simply get files > which they haven't rated,
with no ordering apart from that. However, the > downside of
this is that we may get lots of files with 4 ratings, but some >
remain with only 1 until we've done the lot. Let me know what you'd
prefer, > fellow raters.
I'm getting load times of just less
than 10 sec fairly consistently. This seems quite
acceptable.
Trevor _______________________________________________ lilypond-user
mailing list address@hidden https://lists.gnu.org/mailman/listinfo/lilypond-user
|