On 13 April 2011 16:52, Dami Laurent (PJ) <[email protected]>wrote:
> I have an important database migration project, aiming at replacing an > old DBMS by something more up to date, for mission-critical applications. We > are still at a very early phase of the selection process, but some of the > requirements will be : high-availability , support for multi-values, > fulltext search, data domains, inheritance, CHECK constraints and triggers. > The short list of candidate DBMS is likely to be Oracle, PostgreSQL and > MySQL. > > > > Question : do you know of some good sources of factual information that > could help us in the comparison process ? > For starters this is a good coverage of two of those http://www.wikivs.com/wiki/MySQL_vs_PostgreSQL Oracle will do what you want but cost you lots and every time I use it I hate it more (just a personal view). I've used MySQL a lot over the years but have bitten by things like quirks on their fulltext search and until recently it was slow at allocating autoincrement primary keys in a cluster. The fallout with the takeover by Sun and then Oracle and Monty leaving was fairly disastrous. http://monty-says.blogspot.com/2008/11/oops-we-did-it-again-mysql-51-released.html http://monty-says.blogspot.com/2010/12/quick-look-at-mysql-55-ga.html and then Oracle's poor behaviour with the open source community http://arstechnica.com/open-source/news/2010/12/apache-resigns-from-jcp-in-protest-of-oracle-governance-failures.ars http://www.ibtimes.com/articles/69096/20101006/google-oracle-lawsuit-java-sun-microsystems-dalvik-abstraction-virtual-machine-open-source-apache-ha.htm would make me cautious about committing to it long term. I think PostgreSQL will do everything you want - have a close read of the manual and try out your exact requirements. At BBC WS we moved recently from Pg8.3 to a PostgreSQL 9.0 database cluster that uses streaming replication to a hot standby off-site for disaster recovery purposes. Performance is great and it works fine for us. Check out the options in http://www.postgresql.org/docs/9.0/interactive/high-availability.html and http://wiki.postgresql.org/wiki/Replication,_Clustering,_and_Connection_Pooling also the discussions in http://www.sraoss.co.jp/event_seminar/2010/20100702-03char10.pdf For HA today my guess is you'd use middleware in front of replication http://pgpool.projects.postgresql.org/ Bear in mind also the PG guys have been very active adding new features the past couple of years and are adding in multi-master MVCC for release 9.1 so the situation will keep improving: http://wiki.postgresql.org/wiki/Postgres-XC Regards, Peter
_______________________________________________ List: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/dbix-class IRC: irc.perl.org#dbix-class SVN: http://dev.catalyst.perl.org/repos/bast/DBIx-Class/ Searchable Archive: http://www.grokbase.com/group/[email protected]
