| From: | Kohei KaiGai <kaigai(at)kaigai(dot)gr(dot)jp> |
|---|---|
| To: | Jan Wieck <jan(at)wi3ck(dot)info> |
| Cc: | Kouhei Kaigai <kaigai(at)ak(dot)jp(dot)nec(dot)com>, Merlin Moncure <mmoncure(at)gmail(dot)com>, "pgsql-hackers(at)postgreSQL(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: DBT-3 with SF=20 got failed |
| Date: | 2015-06-11 14:28:52 |
| Message-ID: | CADyhKSU3Sh37Hh8dgqCuv6tU2J-A17trgOeRBbK6TyjpUB6CMA@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
2015-06-11 23:20 GMT+09:00 Jan Wieck <jan(at)wi3ck(dot)info>:
> On 06/11/2015 09:53 AM, Kouhei Kaigai wrote:
>>>
>>> curious: what was work_mem set to?
>>>
>> work_mem=48GB
>>
>> My machine mounts 256GB physical RAM.
>
>
> work_mem can be allocated several times per backend. Nodes like sort and
> hash_aggregate may each allocate that much. You should set work_mem to a
> fraction of physical-RAM / concurrent-connections depending on the
> complexity of your queries. 48GB does not sound reasonable.
>
Smaller number of max_connections and large work_mem configuration are
usual for typical OLAP workloads.
Even if configuration is not reasonable, it is not a right error message.
People cannot understand how to fix it.
psql:query21.sql:50: ERROR: invalid memory alloc request size 1073741824
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tomas Vondra | 2015-06-11 14:33:17 | Re: DBT-3 with SF=20 got failed |
| Previous Message | Robert Haas | 2015-06-11 14:28:48 | Re: DBT-3 with SF=20 got failed |