Thread: What is the extent of FDW join pushdown support in 9.5?

What is the extent of FDW join pushdown support in 9.5?

From
Peter Eisentraut
Date:
The 9.5 release notes contain this promising but cryptic item:

- Allow foreign data wrappers and custom scans to implement join
pushdown (KaiGai Kohei)

As phrased, this seems to mean, "it can be done, but we haven't done
it".  But there is no link to any documentation that explains how to do
this.  The commit that appears to have added this feature touches
postgres_fdw, but neither the documentation nor the tests show anything
about this feature, and I haven't been able to reproduce anything that
would seem to indicate anything like this is at work.

So what is the actual extent of this feature?



Re: What is the extent of FDW join pushdown support in 9.5?

From
Kouhei Kaigai
Date:
> The 9.5 release notes contain this promising but cryptic item:
> 
> - Allow foreign data wrappers and custom scans to implement join
> pushdown (KaiGai Kohei)
> 
> As phrased, this seems to mean, "it can be done, but we haven't done
> it".  But there is no link to any documentation that explains how to do
> this.  The commit that appears to have added this feature touches
> postgres_fdw, but neither the documentation nor the tests show anything
> about this feature, and I haven't been able to reproduce anything that
> would seem to indicate anything like this is at work.
> 
> So what is the actual extent of this feature?
>
It says these enhancement on interface allows extensions to implement
join operation in their own way (including remote join in case of FDW),
however, enhancement of postgres_fdw is not yet upstreamed.

So, how about the description below? It focuses on interface enhancement.

- Allow extensions to implement join in their own way using foreign data wrappers and customer scan interface.

Thanks,
--
NEC Business Creation Division / PG-Strom Project
KaiGai Kohei <kaigai@ak.jp.nec.com>


Re: What is the extent of FDW join pushdown support in 9.5?

From
Peter Eisentraut
Date:
On 9/15/15 10:02 PM, Kouhei Kaigai wrote:
>> The 9.5 release notes contain this promising but cryptic item:
>>
>> - Allow foreign data wrappers and custom scans to implement join
>> pushdown (KaiGai Kohei)
>>
>> As phrased, this seems to mean, "it can be done, but we haven't done
>> it".  But there is no link to any documentation that explains how to do
>> this.  The commit that appears to have added this feature touches
>> postgres_fdw, but neither the documentation nor the tests show anything
>> about this feature, and I haven't been able to reproduce anything that
>> would seem to indicate anything like this is at work.
>>
>> So what is the actual extent of this feature?
>>
> It says these enhancement on interface allows extensions to implement
> join operation in their own way (including remote join in case of FDW),
> however, enhancement of postgres_fdw is not yet upstreamed.

Thank you for this clarification.  I suppose this item is correct as
written.




Re: What is the extent of FDW join pushdown support in 9.5?

From
Bruce Momjian
Date:
On Wed, Sep 16, 2015 at 04:32:51PM -0400, Peter Eisentraut wrote:
> On 9/15/15 10:02 PM, Kouhei Kaigai wrote:
> >> The 9.5 release notes contain this promising but cryptic item:
> >>
> >> - Allow foreign data wrappers and custom scans to implement join
> >> pushdown (KaiGai Kohei)
> >>
> >> As phrased, this seems to mean, "it can be done, but we haven't done
> >> it".  But there is no link to any documentation that explains how to do
> >> this.  The commit that appears to have added this feature touches
> >> postgres_fdw, but neither the documentation nor the tests show anything
> >> about this feature, and I haven't been able to reproduce anything that
> >> would seem to indicate anything like this is at work.
> >>
> >> So what is the actual extent of this feature?
> >>
> > It says these enhancement on interface allows extensions to implement
> > join operation in their own way (including remote join in case of FDW),
> > however, enhancement of postgres_fdw is not yet upstreamed.
> 
> Thank you for this clarification.  I suppose this item is correct as
> written.

Yes, it was cryptically written on purpose.  ;-)

--  Bruce Momjian  <bruce@momjian.us>        http://momjian.us EnterpriseDB
http://enterprisedb.com

+ As you are, so once was I. As I am, so you will be. +
+ Roman grave inscription                             +