Re: Recommendation to run vacuum FULL in parallel

On 2019-04-03 19:42:03 +0400, rihad wrote:
> > And future updates can reuse it, too (an update is very similar to an
> > insert+delete).
>
> Hm, then it's strange our DB takes 6 times as much space compared to freshly
> restored one (only public schema is considered).

This is indeed strange if you accumulated that much bloat gradually (as
you wrote). It is much less strange if you did some massive
reorganisations in the past (In one case I witnessed, changes had to be
made to almost every value in 4 or 5 columns of a large table. So the
person doing the updates first issued an update on the first column,
checked that the result looked plausible, then issued an update on the
second column, and so on. The result was of course massive bloat).

hp

--
_ | Peter J. Holzer | we build much bigger, better disasters now
|_|_) | | because we have much more sophisticated
| | | hjp(at)hjp(dot)at | management tools.
__/ | http://www.hjp.at/ | -- Ross Anderson <https://www.edge.org/>