Thanks, your post got me to recognize something was wrong so I did
more research.  From another post on this group, I found the tip of
taking the STI column from VARCHAR(255) and making it a VARCHAR(XX)
where XX is as small as I could make it.  This, and a few other small
tweaks got me down to under 2 minutes. Epic win.

BTW, if a future help-seeker comes across this page for ways to
improve indexing performance on a production setup, make sure you read
this: 
http://www.mendable.com/thinking-sphinx-performance-index-from-slave-mysql-database/



On Oct 5, 9:22 pm, James Healy <[email protected]> wrote:
> coneybeare wrote:
> > This setup with only 20k entries takes almost 1 hour to complete.  I
> > have increased RAM to 128M, tried to follow the guide to make indexing
> > faster, but this still seems very slow.  I have worked with Xapian
> > before and on tables that were orders of magnitude larger, it was
> > faster.  Because sphinx and TS are just as good a Xapian, this leads
> > me to believe something is wrong.  As a sanity check, does this slow
> > indexing with my setup sound normal?
>
> Nope, this sounds slower than I'd expect.
>
> The first thing to check is the ID's being indexed, as per [1].
>
> Failing that, grab a copy of the slow SQL query (from
> production.sphinx.conf), prefix it with EXPLAIN and run it in mysql
> directly to see what it says. Hopefully it will point out a missing
> index or something.
>
> cheers
>
> -- James Healy <[email protected]>  Wed, 06 Oct 2010 12:20:01 +1100
>
> [1]http://freelancing-god.github.com/ts/en/common_issues.html#slow_indexing

-- 
You received this message because you are subscribed to the Google Groups 
"Thinking Sphinx" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/thinking-sphinx?hl=en.

Reply via email to