Re: RE: Re: [ADMIN] v7.1b4 bad performance - Mailing list pgsql-hackers

From Vadim Mikheev
Subject Re: RE: Re: [ADMIN] v7.1b4 bad performance
Date
Msg-id 387428168.982939317658.JavaMail.root@web444-ec.mail.com
Whole thread Raw
List pgsql-hackers
> > > It may be that WAL has changed the rollback
> > > time-characteristics to worse than pre-wal ?
> >
> > Nothing changed ... yet. And in future rollbacks
> > of read-only transactions will be as fast as now,
> > anyway.
> 
> What about rollbacks of a bunch uf inserts/updates/deletes?
>
> I remember a scenario where an empty table was used
> by several backends for gathering report data, and
> when the report is done they will rollback to keep
> the table empty.
>
> Should this kind of usage be replaced in the future by 
> having backend id as a key and then doing delete by that 
> key in the end ?

Isn't it what we have right now?
But I believe that in future we must remove
modifications made by aborted transactions
immediately, without keeping them till vacuum.
So - yes: rollback of read-write transactions
will take longer time.

Vadim

-----------------------------------------------
FREE! The World's Best Email Address @email.com
Reserve your name now at http://www.email.com




pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: WAL does not recover gracefully from out-of-disk-space
Next
From: Sezai YILMAZ
Date:
Subject: Re: [BUGS] Turkish locale bug