You could also look into a filtered index that perhaps only covers dates earlier than a certain point in time where regular performance wouldn't be hindered. But Gavin is absolutely right otherwise.
On Jan 5, 2014 5:22 PM, "Sergey Konoplev" <gray.ru@gmail.com> wrote:
>
> On Sun, Jan 5, 2014 at 2:19 PM, Gavin Flower
> <GavinFlower@archidevsys.co.nz> wrote:
> > On 06/01/14 11:08, Sergey Konoplev wrote:
> > [...]
> >
> >> An index might be considered as useless when there were no idx scans for
> >> the significantly long period. However it might be non-trivial to define
> >> this period. Eg. one have a query building an annual report that uses this
> >> index and the period here is one year.
> >
> > [...]
> >
> > An index only used by an annual report, should possibly be only created
> > prior to the report run & dropped immediately afterwards - why carry its
> > overhead for the bulk of the year?
>
> I fully agree. This is the matter of implementation.
>
> --
> Kind regards,
> Sergey Konoplev
> PostgreSQL Consultant and DBA
>
> http://www.linkedin.com/in/grayhemp
> +1 (415) 867-9984, +7 (901) 903-0499, +7 (988) 888-1979
> gray.ru@gmail.com
>
>
> --
> Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-general