Re: Lossy Index Tuple Enhancement (LITE)

From: Simon Riggs <simon(at)2ndquadrant(dot)com>
To: Bruce Momjian <bruce(at)momjian(dot)us>
Cc: Simon Riggs <simon(at)2ndquadrant(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Lossy Index Tuple Enhancement (LITE)
Date: 2016-08-04 18:19:25
Message-ID: CANP8+j+mFB-OJnhFKpCzSMAQvCsG=eQov7s1ydgs09Sgfonycw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 4 August 2016 at 18:27, Bruce Momjian <bruce(at)momjian(dot)us> wrote:

>> > Also, why not use this bitmap for all indexes, not just update chains?
>>
>> I don't understand where you get this update chains thing from.
>>
>> The bitmap can apply to multiple tuples on one page, which is described.
>
> I am asking if we could use this idea for all rows on a page, not just
> HOT updated, e.g. we have five rows that have col1=4 in an index --- why
> not just use on index pointer for all of them, with a bitmap to point to
> the possible matches?

Yes. I described that...

"The same situation can occur if we have many INSERTs with same values
on the same block. This could lead to a reduction in size of the btree
for indexes with many duplicate values."

--
Simon Riggs http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Andres Freund 2016-08-04 18:20:37 Re: Pgbench performance tuning?
Previous Message Greg Stark 2016-08-04 18:15:43 Re: Pgbench performance tuning?