On 11/08/2011 04:23 PM, Kevin Grittner wrote:
> http://wiki.postgresql.org/wiki/Materialized_Views
> That page describes three components: creating MVs, updating MVs, and
> having the planner automatically detect when an MV matches some
> portion of a regular query and using the MV instead of the specified
> tables in such cases.
So far that page is just my initial notes on this subject based on some
research I did, don't read too much into it. Ignoring MV substitution
is certainly the right plan for an initial development target. An
implementation that didn't update automatically at all is also still a
useful step. It's very easy to pick too big of a chunk of this project
to chew on at once.
When I wrote that, it wasn't completely clear to me yet that doing the
updates would involve so many of the same tricky bits that stalled
progress on the MERGE command. Nowadays I think of working out the
concurrency issues in both this and MERGE, in the wake of true
serialization, as another step in this implementation plan. It's not
strictly necessary, but there's some shared issues that might be tackled
easier in the MERGE context instead.
--
Greg Smith 2ndQuadrant US greg@2ndQuadrant.com Baltimore, MD
PostgreSQL Training, Services, and 24x7 Support www.2ndQuadrant.us