Skip to content

9.6, or 10.0?

Even though PostgreSQL 9.5 was released not too long ago, we are almost getting closer towards the feature freeze for 9.6.

When Windows support and PITR capabilities were added to 7.5, the community decided to change the major version to 8.0, as it represents a significant changes at those times. Similar thing happened at 8.5 development cycle: We now had in core replication, and it was released as 9.0.

Now, as 9.6 is coming closer, I seriously think that we should release 10.0, not 9.6 (I've been ranting about this on my Twitter account for a while). Following the commit of Parallel sequential scan , subsequent commits for:

- Parallel Joins
- Parallel Aggregation

increased parallelism features.

Another big infrastructure change is Tom's patch on "Making the upper part of the planner work by generating and comparing Paths."

Apart from these, following the commit that changes the VM file format, another commit titled
Don't vacuum all-frozen pages will help us to deploy PostgreSQL in some environments that we could not do before, I think.

To keep it short, I think that these infrastructural changes should result in a .0 release, as we did before.

What do you think?

Trackbacks

No Trackbacks

Comments

Display comments as Linear | Threaded

Victor on :

I would prefer to keep it 9.6 for now and wait till native partitioning will be added, hopefully in 9.7.

For along with Partitioning, we have Multivariate Stats, Covering and Including Indexes, Access Methods extendability patches, that are also in the queue.

Devrim Gündüz on :

I think we don't change the version based on the feature list -- it is more like a backwards compatibility.

aRkadeFR on :

That would make sense to me (user of postgresql).

Ron Dunn on :

My view is that a version change should represent a DBMS structure change that requires conversion, migration, or some similar process. Minor versions should work with existing database files. In line with that view, we should keep the data directory at the version level, not the dot point.

Matt on :

Parallelism is great for bulk data, but there seems to be little else in this release. 9.5 had a greater quantity of interesting features. Personally, I look forward to BDR in core or as an extension, compiled stored procedures, and something that lets stored procedures crunch data in parallel (as opposed to just SQL primitives).

Add Comment

Enclosing asterisks marks text as bold (*word*), underscore are made via _word_.
Standard emoticons like :-) and ;-) are converted to images.

To prevent automated Bots from commentspamming, please enter the string you see in the image below in the appropriate input box. Your comment will only be submitted if the strings match. Please ensure that your browser supports and accepts cookies, or your comment cannot be verified correctly.
CAPTCHA

Form options