| From: | Simon Riggs <simon(at)2ndquadrant(dot)com> |
|---|---|
| To: | Gregory Stark <stark(at)enterprisedb(dot)com> |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Bulk loading performance improvements |
| Date: | 2008-02-26 15:36:16 |
| Message-ID: | 1204040176.4252.332.camel@ebony.site |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Tue, 2008-02-26 at 15:25 +0000, Gregory Stark wrote:
> "Simon Riggs" <simon(at)2ndquadrant(dot)com> writes:
>
> > 1.41 7.25 0.13 10000029 CacheInvalidateHeapTuple
> >
> > which together account for more than 50% of CPU.
>
> The second column is cumulative percentage (the third column is "self" time).
> So together this function and all the items above it account for 7.25% of CPU.
I think you have read my profile incorrectly. Only certain lines of the
profile have been included in my assessment, based upon whether the
number of calls is the minimum.
The first column is the important one: those can be added together,
which is all I have done.
--
Simon Riggs
2ndQuadrant http://www.2ndQuadrant.com
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Robert Lor | 2008-02-26 15:42:29 | Re: Proposed changes to DTrace probe implementation |
| Previous Message | Andrew Dunstan | 2008-02-26 15:31:40 | Re: [COMMITTERS] pgsql: Link postgres from all object files at once, to avoid the |