On Sun, Feb 14, 2010 at 8:25 PM, Greg Stark <stark(at)mit(dot)edu> wrote:
> So this is what I did about my two complaints earlier about the
> explain buffer patch.
> a) Changed the line description to "Total Buffer Usage" which at least
> hints that it's something more akin to the "Total runtime" listed at
> the bottom than the "actual time".
> b) Used units of memory -- I formatted them with 3 significant digits
> (unless the unit is bytes or kB where that would be silly). It's just
> what looked best to my eye.
I wasn't aware we had consensus on making this change, which I see you
committed less than an hour after posting this.
> I'm finding "hit" and "read" kind of confusing myself but don't really
> have any better idea. It's not entirely clear whether read is the
> total accesses out of which some are cache hits or if they're two
> disjoint sets.
Keep in mind these terms are taken from other parts of the system
where they existed prior to this patch. We probably want to stick
with them at this point for consistency, but in any case it's
certainly a separate discussion.
In response to
pgsql-hackers by date
|Next:||From: Robert Haas||Date: 2010-02-15 14:29:36|
|Subject: alpha4 timing (was: Speed up CREATE DATABASE)|
|Previous:||From: Magnus Hagander||Date: 2010-02-15 14:15:40|
|Subject: Re: [COMMITTERS] pgsql: Speed up CREATE DATABASE by deferring the fsyncs until after|