> I would be nice to get a feel for how much performance loss would be incurred in
> maintaining the index flags against possible performance gains for getting the data back
> out again.
I guess the real question is, why maintain index flags and not simply
drop the index entry altogether?
A more interesting case would be to have the backend process record
index tuples that it would invalidate (if committed), then on commit
send that list to a garbage collection process.
It's still vacuum -- just the reaction time for it would be much
In response to
pgsql-performance by date
|Next:||From: Gary Doades||Date: 2004-04-29 19:23:19|
|Subject: Re: planner/optimizer question|
|Previous:||From: Josh Berkus||Date: 2004-04-29 18:21:51|
|Subject: Re: Wierd context-switching issue on Xeon|