Re: [HACKERS] Parallel bitmap heap scan - Mailing list pgsql-hackers

From Dilip Kumar
Subject Re: [HACKERS] Parallel bitmap heap scan
Date
Msg-id CAFiTN-vSvv3VGm9vUqz2uYX62QsJqxsHsJSiMnvj=VpFr2HO1A@mail.gmail.com
Whole thread Raw
In response to Re: [HACKERS] Parallel bitmap heap scan  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
On Wed, Mar 8, 2017 at 6:42 PM, Robert Haas <robertmhaas@gmail.com> wrote:
> I don't think I understand exactly why this system might be prone to a
> little bit of extra prefetching - can you explain further?
Let me explain with an example, suppose there are 2 workers
prefetching jointly, lets assume
prefetch_target is 3, so for example when both the workers prefetch 1
page each then prefetch_page count will be 2, so both the workers will
go for next prefetch because prefetch_page is still less than prefetch
target, so again both the workers will do prefetch and totally they
will prefetch 4 pages.

> I don't think it will hurt anything as long as we are talking about a small
> number of extra prefetches here and there.

I completely agree with this point and I mentioned in the mail so that
it don't go unnoticed.

And, whatever extra prefetch we have done we will anyway use it unless
we stop the execution because of some limit clause.

>
>> Fixed
>
> Committed 0001.

Thanks


-- 
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com



pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: [HACKERS] port of INSTALL file generation to XSLT
Next
From: Tom Lane
Date:
Subject: Re: [HACKERS] Removing #include "postgres.h" from a couple of headers