Re: Pluggable Indexes (was Re: rmgr hooks (v2)) - Mailing list pgsql-hackers

From Simon Riggs
Subject Re: Pluggable Indexes (was Re: rmgr hooks (v2))
Date
Msg-id 1232582880.2327.696.camel@ebony.2ndQuadrant
Whole thread Raw
In response to Re: Pluggable Indexes (was Re: rmgr hooks (v2))  (Bruce Momjian <bruce@momjian.us>)
Responses Re: Pluggable Indexes (was Re: rmgr hooks (v2))  (Bruce Momjian <bruce@momjian.us>)
Re: Pluggable Indexes (was Re: rmgr hooks (v2))  (Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>)
List pgsql-hackers
On Wed, 2009-01-21 at 18:06 -0500, Bruce Momjian wrote:

> Plugability adds complexity. Heikki's comment is that adding this
> patch make the job of creating pluggable indexes 5% easier, while no
> one is actually working on plugable indexes, and it hard to say that
> making it 5% easier really advances anything, especially since many of
> our existing index types aren't WAL-logged.  Plugability is not a
> zero-cost feature.

Sorry Bruce, but that misses the key point.

Without the patch it is completely *impossible* to write an index plugin
that is *recoverable*. Yes, we have pluggable indexes now, but unless
they are recoverable we certainly can't ever use them in production.

With the patch, you still have to write the index code. I agree it is
hard code to write, but not impossible. I would go so far as to say that
the patch helps you 0% with the task of actually writing the plugin. But
the patch enables you to start and that is all its intended as: an
enabler.

So its not a "slightly easier" thing, its a can/cannot thing.

-- Simon Riggs           www.2ndQuadrant.comPostgreSQL Training, Services and Support



pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: TODO items for window functions
Next
From: Bruce Momjian
Date:
Subject: Re: Pluggable Indexes (was Re: rmgr hooks (v2))