Re: 9.3 Pre-proposal: Range Merge Join - Mailing list pgsql-hackers

From Stefan Keller
Subject Re: 9.3 Pre-proposal: Range Merge Join
Date
Msg-id CAFcOn2-sYxaT3m5AxPAtX4A1qq0RK1TufOpPn0DRDir5q_xrCw@mail.gmail.com
Whole thread Raw
In response to Re: 9.3 Pre-proposal: Range Merge Join  (Jeff Davis <pgsql@j-davis.com>)
Responses Re: 9.3 Pre-proposal: Range Merge Join
List pgsql-hackers
Hi Jeff

2013/1/18 Jeff Davis <pgsql@j-davis.com>:
> On Thu, 2013-01-17 at 21:03 +0100, Stefan Keller wrote:
>> Hi Jeff
>
>> I'm perhaps really late in this discussion but I just was made aware
>> of that via the tweet from Josh Berkus about "PostgreSQL 9.3: Current
>> Feature Status"
>>
>> What is the reason to digg into spatial-joins when there is PostGIS
>> being a bullet-proof and fast implementation?
>>
>
> Hi Stefan,
>
> You are certainly not too late.
>
> PostGIS uses the existing postgres infrastructure to do spatial joins.
> That mean it either does a cartesian product and filters the results, or
> it uses a nested loop with an inner index scan.
>
> That isn't too bad, but it could be better. I am trying to introduce a
> new way to do spatial joins which will perform better in more
> circumstances. For instance, we can't use an inner index if the input
> tables are actually subqueries, because we can't index a subquery.
>
> Regards,
>         Jeff Davis

Sounds good.
Did you already had contact e.g. with Paul (cc'ed just in case)?
And will this clever index also be available within all these hundreds
of PostGIS functions?

Regards, Stefan



pgsql-hackers by date:

Previous
From: Heikki Linnakangas
Date:
Subject: Passing connection string to pg_basebackup
Next
From: wang chaoyong
Date:
Subject: Re: [HACKERS] How to hack the storage component?