Обсуждение: BDR-project

On 12/25/13, 7:01 AM, Andreas Joseph Krogh wrote:
> http://wiki.postgresql.org/wiki/BDR_Project
>  
> Is implementing main BDR features into core Postgres a probable
> objective to version 9.4?

No.

No BDR feature has been proposed for core PostgreSQL so far.  Also, this
feature builds on top of the logical replication features, which are
themselves questionable for 9.4.

Hi,

On 2013-12-26 14:26:15 -0500, Peter Eisentraut wrote:
> On 12/25/13, 7:01 AM, Andreas Joseph Krogh wrote:
> > http://wiki.postgresql.org/wiki/BDR_Project
> > Is implementing main BDR features into core Postgres a probable
> > objective to version 9.4?

Not fully. We hope to integrate some of the building blocks into
postgres soon, so we can reduce the size of the diff. Then the plan is
to iteratively (continue) to submit features required.

> No BDR feature has been proposed for core PostgreSQL so far.  Also, this
> feature builds on top of the logical replication features, which are
> themselves questionable for 9.4.

Well, that's not really true. For one, changeset extraction actually is
part of the effort, which surely has been submitted. There also have
been some further submissions which help build robust replication
solutions. One or two of those might be possible to get into 9.4 (they
have been in the last CF...)

Greetings,

Andres Freund

-- Andres Freund                       http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training &
Services

On 25 December 2013 12:01, Andreas Joseph Krogh <andreak@officenet.no> wrote:

> Ref:
> http://wiki.postgresql.org/wiki/BDR_Project
>
> Is implementing main BDR features into core Postgres a probable objective to
> version 9.4?

I've not updated hackers recently on this point, so thanks for asking
the question. This was discussed in the ending keynote of the PG
Europe conference, but I appreciate that's not the same thing as
saying it here.

The plan is

* submit the core logical replication technology for 9.4
* submit online upgrade as a feature for 9.5, allowing upgrades from 9.4+
* submit full BDR features for 9.6

BDR code will be released as a separate open source project until/if
core accepts/modifies that. There's lots of work and discussion to be
had yet, so the above plan is a reasonable schedule for achieving
change allowing input from all. Design to full feature submission
would be 4.5 years, plus we expect the features to mature/extend after
that, so there's no rush, just steady movement.

No attempts to publicise that as yet, but if all goes well we expect
to do that once 9.4 is released.

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

On 25 December 2013 12:01, Andreas Joseph Krogh <andreak@officenet.no> wrote:

> Ref:
> http://wiki.postgresql.org/wiki/BDR_Project
>
> Is implementing main BDR features into core Postgres a probable objective to
> version 9.4?

I've not updated hackers recently on this point, so thanks for asking
the question. This was discussed in the ending keynote of the PG
Europe conference, but I appreciate that's not the same thing as
saying it here.

The plan is

* submit the core logical replication technology for 9.4
* submit online upgrade as a feature for 9.5, allowing upgrades from 9.4+
* submit full BDR features for 9.6

BDR code will be released as a separate open source project until/if
core accepts/modifies that. There's lots of work and discussion to be
had yet, so the above plan is a reasonable schedule for achieving
change allowing input from all. Design to full feature submission
would be 4.5 years, plus we expect the features to mature/extend after
that, so there's no rush, just steady movement.

No attempts to publicise that as yet, but if all goes well we expect
to do that once 9.4 is released.