Обсуждение: Too frequent message of pgbench -i?

Поиск
Список
Период
Сортировка

Too frequent message of pgbench -i?

От
Tatsuo Ishii
Дата:
Currently pgbench -i prints following message every 10k tuples created.
        fprintf(stderr, "%d tuples done.\n", j);

I think it's long time ago when the frequency of message seemed to be
appropriate because computer is getting so fast these days and every
10k message seems to be too often for me. Can we change the frequency
from 10k to 100k? Or should we make the frequency be adjustable from a
command line? Or even better, the frequency should be changed
automatically according to scale factor?
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp


Re: Too frequent message of pgbench -i?

От
Heikki Linnakangas
Дата:
On 20.06.2012 11:04, Tatsuo Ishii wrote:
> Currently pgbench -i prints following message every 10k tuples created.
>
>             fprintf(stderr, "%d tuples done.\n", j);
>
> I think it's long time ago when the frequency of message seemed to be
> appropriate because computer is getting so fast these days and every
> 10k message seems to be too often for me. Can we change the frequency
> from 10k to 100k? Or should we make the frequency be adjustable from a
> command line? Or even better, the frequency should be changed
> automatically according to scale factor?

Or print the message every n seconds, instead of every n lines. I think 
a new command-line option would be overkill. I would be fine with just 
bumping it to every 100k lines.

So +1 on doing something about it, not sure what.

--   Heikki Linnakangas  EnterpriseDB   http://www.enterprisedb.com


Re: Too frequent message of pgbench -i?

От
Robert Haas
Дата:
On Wed, Jun 20, 2012 at 4:04 AM, Tatsuo Ishii <ishii@postgresql.org> wrote:
> Currently pgbench -i prints following message every 10k tuples created.
>
>                        fprintf(stderr, "%d tuples done.\n", j);
>
> I think it's long time ago when the frequency of message seemed to be
> appropriate because computer is getting so fast these days and every
> 10k message seems to be too often for me. Can we change the frequency
> from 10k to 100k?

+1 for doing it that way.  I have an old patch lying around for that.
It's one line, which seems to be about as much work as the problem
justifies.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Вложения

Re: Too frequent message of pgbench -i?

От
Tatsuo Ishii
Дата:
> On Wed, Jun 20, 2012 at 4:04 AM, Tatsuo Ishii <ishii@postgresql.org> wrote:
>> Currently pgbench -i prints following message every 10k tuples created.
>>
>>                        fprintf(stderr, "%d tuples done.\n", j);
>>
>> I think it's long time ago when the frequency of message seemed to be
>> appropriate because computer is getting so fast these days and every
>> 10k message seems to be too often for me. Can we change the frequency
>> from 10k to 100k?
>
> +1 for doing it that way.  I have an old patch lying around for that.
> It's one line, which seems to be about as much work as the problem
> justifies.

+1. Your patch looks good.
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp


Re: Too frequent message of pgbench -i?

От
Robert Haas
Дата:
On Thu, Jun 21, 2012 at 3:44 AM, Tatsuo Ishii <ishii@postgresql.org> wrote:
>> On Wed, Jun 20, 2012 at 4:04 AM, Tatsuo Ishii <ishii@postgresql.org> wrote:
>>> Currently pgbench -i prints following message every 10k tuples created.
>>>
>>>                        fprintf(stderr, "%d tuples done.\n", j);
>>>
>>> I think it's long time ago when the frequency of message seemed to be
>>> appropriate because computer is getting so fast these days and every
>>> 10k message seems to be too often for me. Can we change the frequency
>>> from 10k to 100k?
>>
>> +1 for doing it that way.  I have an old patch lying around for that.
>> It's one line, which seems to be about as much work as the problem
>> justifies.
>
> +1. Your patch looks good.

Thanks, committed.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company