| From: | "Luke Lonergan" <llonergan(at)greenplum(dot)com> |
|---|---|
| To: | "Mark Kirkwood" <markir(at)paradise(dot)net(dot)nz>, "Neil Conway" <neilc(at)samurai(dot)com> |
| Cc: | "PostgreSQL-patches" <pgsql-patches(at)postgresql(dot)org>, "Eng" <eng(at)intranet(dot)greenplum(dot)com>, "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Subject: | Re: Avg performance for int8/numeric |
| Date: | 2006-11-25 01:38:30 |
| Message-ID: | C18CDE96.E77E%llonergan@greenplum.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers pgsql-patches |
So, if I understand this correctly, we're calling Alloc and ContextAlloc 10
times for every row being summed?
There are approx 10M rows and the profile snippet below shows 100M calls to
each of those.
- Luke
On 11/24/06 4:46 PM, "Mark Kirkwood" <markir(at)paradise(dot)net(dot)nz> wrote:
> time seconds seconds calls s/call s/call name
> 14.42 2.16 2.16 100002977 0.00 0.00 AllocSetAlloc
> 9.08 3.52 1.36 20000000 0.00 0.00 add_abs
> 5.54 4.35 0.83 10000000 0.00 0.00 slot_deform_tuple
> 5.41 5.16 0.81 60001673 0.00 0.00 AllocSetFree
> 4.34 5.81 0.65 10000000 0.00 0.00 construct_md_array
> 4.21 6.44 0.63 20000003 0.00 0.00 make_result
> 3.54 6.97 0.53 10000000 0.00 0.00 numeric_add
> 3.27 7.46 0.49 30000003 0.00 0.00 set_var_from_num
> 3.00 7.91 0.45 100002652 0.00 0.00 MemoryContextAlloc
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Luke Lonergan | 2006-11-25 01:44:53 | Re: [PATCHES] Avg performance for int8/numeric |
| Previous Message | Mark Kirkwood | 2006-11-25 00:48:39 | Re: [PATCHES] Avg performance for int8/numeric |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Luke Lonergan | 2006-11-25 01:44:53 | Re: [PATCHES] Avg performance for int8/numeric |
| Previous Message | Mark Kirkwood | 2006-11-25 00:48:39 | Re: [PATCHES] Avg performance for int8/numeric |