On 06/16/2011 10:06 AM, Achilleas Mantzios wrote:
> Till the end of July i must have finished all the migration to the new versions.
> So i am asking what would be better from your perspective to do?
> Go for 9.1? or stick to 9.0 and try to deploy it and take the most out of it?
> When is a stable (release) version of 9.1 be available?
> Has any one faced any issues migrating from 9.0 to 9.1
>
I would place odds at about 1/3 that 9.1 will be available by the end of
July. But you will still need to do testing of your application first
before deploying onto that version. Realistically, even the earliest of
9.1 adopters is unlikely to launch before August. As such, there's not
very much experience about the migration available yet, either.
A large number of the new features in 9.1 aim at making certain types of
development easier. The must-have features I am hearing demand for from
my customers (who admittedly care more about replication and performance
features than most), such that they are postponing some deployments
until 9.1 ships because 9.0 just doesn't do what they want, are:
-Synchronous replication
-Support for MIN/MAX queries against partitioned tables
-Feedback mechanism to reduce query conflict resolution when using Hot
Standby
-Much improved monitoring for replication and Hot Standby queries
I'd suggest you take a look at the 9.1 release notes and beta
announcement: http://www.postgresql.org/about/news.1313 ,
http://www.postgresql.org/docs/9.1/static/release-9-1.html
And if you don't see a major compelling reason to wait for 9.1, some
feature in that list that makes your life a lot easier, you really
should just deploy 9.0 and move on. The most critical thing fixed in
9.1 development that may apply to what you're doing--some bug fixes to
pg_upgrade--have all been backported to 9.0 now.
--
Greg Smith 2ndQuadrant US greg@2ndQuadrant.com Baltimore, MD
PostgreSQL Training, Services, and 24x7 Support www.2ndQuadrant.us
"PostgreSQL 9.0 High Performance": http://www.2ndQuadrant.com/books
