Re: a fast bloat measurement tool (was Re: Measuring relation free space)

On Wed, Sep 24, 2014 at 2:26 PM, Abhijit Menon-Sen <ams(at)2ndquadrant(dot)com>
wrote:
>
> Hi Amit.
>
> Thanks for your comments, and I'm sorry it's taken me so long to
> respond.

No issues.

> At 2014-08-03 11:18:57 +0530, amit(dot)kapila16(at)gmail(dot)com wrote:

> > 7.
> > HeapTupleSatisfiesVacuumNoHint()
> > a. Why can't we use HeapTupleSatisfiesVisibility() with dirty snapshot
> > as we use for pgstattuple?
>
> Heavier locking. I tried to make do with the existing HTS* functions,
> but fastbloat was noticeably faster in tests with the current setup.

I am not sure for the purpose of this functionality, why we need to
use a different HTS (HeapTupleSatisfiesVacuum) routine as compare
to pgstat_heap(). Unless you have some specific purpose to achieve,
I think it is better to use HeapTupleSatisfiesVisibility().

> Maybe I'll call it not_too_slow_bloat().

How about pgfaststattuple() or pgquickstattuple() or pgfuzzystattuple()?

With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com