Re: Seq scans roadmap - Mailing list pgsql-hackers

From CK Tan
Subject Re: Seq scans roadmap
Date
Msg-id AD9630BA-9A0F-408C-A7E1-D799722986AC@greenplum.com
Whole thread Raw
In response to Re: Seq scans roadmap  (Heikki Linnakangas <heikki@enterprisedb.com>)
List pgsql-hackers
The patch has no effect on scans that do updates. The  
KillAndReadBuffer routine does not force out a buffer if the dirty  
bit is set. So updated pages revert to the current performance  
characteristics.

-cktan
GreenPlum, Inc.

On May 10, 2007, at 5:22 AM, Heikki Linnakangas wrote:

> Zeugswetter Andreas ADI SD wrote:
>>> In reference to the seq scans roadmap, I have just submitted a  
>>> patch that addresses some of the concerns.
>>>
>>> The patch does this:
>>>
>>> 1. for small relation (smaller than 60% of bufferpool), use the  
>>> current logic 2. for big relation:
>>>     - use a ring buffer in heap scan
>>>     - pin first 12 pages when scan starts
>>>     - on consumption of every 4-page, read and pin the next 4-page
>>>     - invalidate used pages of in the scan so they do not force out  
>>> other useful pages
>> A few comments regarding the effects:
>> I do not see how this speedup could be caused by readahead, so  
>> what are
>> the effects ?
>
> I was wondering that as well. We'd really need to test all the  
> changes separately to see where the improvements are really coming  
> from.
>
> Also, that patch doesn't address the VACUUM issue at all. And using  
> a small fixed size ring with scans that do updates can be  
> devastating. I'm experimenting with different ring sizes for COPY  
> at the moment. Too small ring leads to a lot of WAL flushes, it's  
> basically the same problem we have with VACUUM in CVS HEAD.
>
> -- 
>   Heikki Linnakangas
>   EnterpriseDB   http://www.enterprisedb.com
>




pgsql-hackers by date:

Previous
From: Heikki Linnakangas
Date:
Subject: Re: Seq scans roadmap
Next
From: Neil Conway
Date:
Subject: Planning large IN lists