Re: Runtime variations during day

On Wed, Feb 13, 2013 at 2:53 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Carlos Henrique Reimer <carlos(dot)reimer(at)opendb(dot)com(dot)br> writes:
>> Will include the EXPLAIN ANALYZE again here:
>
>> -> Index Scan using pagpk_aux_mes, pagpk_aux_mes, pk_cadpag, pk_cadpag,
>> pk_cadpag, pagchavefunc00 on cadpag (cost=0.00..131521.88 rows=183322
>> width=62) (actual time=0.664..614.080 rows=32828 loops=1)
>
> BTW, I just noticed this multiple-index annotation, which means that
> either you're not using a community version of Postgres or else it's
> 8.0 or older (8.1 replaced that plan type with bitmap index scans).

I was curious about that string, never haven seen anything like it before.

That also explains why disable_cost is 1e8 rather than 1e10 (commit
b24c02ff2c2a6) in the later query/explains he posted.

Cheers,

Jeff