On 16 February 2017 at 05:24, Seki, Eiji <seki.eiji@jp.fujitsu.com> wrote:
> Simon Riggs <simon(at)2ndquadrant(dot)com> wrote:
>> Please persuade us with measurements that allowing this impact on
>> ANALYZE would really improve performance at least in your case, and
>> also examine the effect of this on the accuracy and usefulness of the
>> gathered statistics.
>
> I explain results of the test that Haribabu mentioned in [1].
Thanks for the tests. I can see the performance is affected by this.
> Please let me know if you need any further information.
...you didn't comment at all on the accuracy and usefulness of the
gathered statistics, when the sample is biased towards non-updated
data.
I'm wondering whether this should be an additional table-level option.
--
Simon Riggs http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services