Re: Call for 7.5 feature completion

From: Hannu Krosing <hannu(at)skype(dot)net>
To: Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org>
Cc: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: Call for 7.5 feature completion
Date: 2005-08-26 14:53:06
Message-ID: 1125067987.4823.24.camel@fuji.krosing.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On N, 2005-08-25 at 19:13 -0400, Alvaro Herrera wrote:

> We have gone a long way now, even though it was only a year ago. My
> question for everyone on this list is: What are the "few remaining big
> features" that you see missing for PostgreSQL?
>
> Or, slightly different, what are people's most wanted features?

my pet wishes are :

24/7 OLTP related things

* vacuums that ignore other vacuums when deciding what tuples to free
(should be mostly done, my patch wasleft to 8.2 due to some doubts by
Tom)
* non-blocking CREATE INDEX / REINDEX (so indexes can be added to huge
tables on busy databases without downtime)
* related to last one - command to promote UNIQUE INDEX to PRIMARY KEY.
* multiple WAL's, assignable to objects (similaŕ to tablespaces).
* better 64-bit support inside db engine.
* real background vacuuming, using something like FSM, may be integrated
with background writer.
* VACUUM FULL/CLUSTER added behaviour of leaving pages half-empty (or
any.other-percentage-empty) for good update behaviour.

OLAP stuff

* table partitioning to move forward.
* archive tables (append (==insert) only, only one writer at a time,
vacuum needed after rollbacked insert, visibility determined by "last
valid ctid" marker, so will not need most of header fields either).
* index-only scans over archive tables (possible without altering
current index structure, as visibility can be determined by ctid which
is already present in index leaf).

> Has PostgreSQL started slowing down in getting new features, and
> concentrating mostly on performance issues?

I can't think of this as new features vs. performance thing as many of
the new features *are* largely about performance, both on database
engine and on user side.

--
Hannu Krosing <hannu(at)skype(dot)net>

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message David Fetter 2005-08-26 15:00:36 Re: Call for 7.5 feature completion
Previous Message Greg Sabino Mullane 2005-08-26 14:52:44 Re: Stuff running slooow