This could turn into an interesting case study, FeedLounge has switched databases, from MySQL to PostgreSQL. The two main factors they listed were database size (34Gig for MySQL/InnoDB vs. 9.6Gig for PostgreSQL) and restoration time (24+ hours to restore database in MySQL vs. less than 5 hours for PostgreSQL).
I’m surprised to here about the huge difference in database size between MySQL (w/InnoDB) and PostgreSQL. Perhaps the different indexing techniques are responsible for this, I’m not sure. For large databases the ability to cut your storage requirements by more than 1/3 is huge. My gut feeling though is that this would not be the norm though. More testing is definitely in order to make this comparison more meaningful.
The huge difference in database restoration time also strikes me as a bit high. Certainly another area where further tests would be interesting to see if this pattern holds true in general. Obviously being able to restore your database in less than 1/3 the time is a good thing.
The real kicker in all this though is the chance to make use of the additional features found in PostgreSQL (they point this out in their write up). I hope that Alex continues to write about how this transition progresses.