On Dec 8, 2011 1:27 PM, "Chris Travers" <chris.travers@gmail.com> wrote:
>
> Additionally I am not entirely sure what he means by the last point.
> If you look at the work that NTT along with EDB has put into
> Postgres-XC, for example, it looks to me like the Postgres ecosystem
> is growing by leaps and bounds and we are approaching an era where
> Oracle is no longer ahead in any significant use case.
While Pg is impressively capable now, I don't agree that Oracle (if DB2, MS-SQL etc) isn't ahead for any significant use case. Not on a purely technical basis anyway - once cost is considered there may be a stronger argument.
Multi-tenant hosting is a weak pint for Pg for quite a few reasons, done of which appear below. It's not the only role Pg isn't a great fit for, but probably one of the more obvious.
Areas in which Pg seems significantly less capable include:
- multi-tenant hosting and row level security
- admission control, queuing and resource limiting to optimally load a machine. Some limited level is possible with external pooling, but only by limiting concurrent workers.
- performance monitoring and diagnostics. It's way harder to find out what's causing load on a busy Pg server or report on frequent/expensive queries etc. Tooling is limited and fairly primitive. It's find, but nowhere near as powerful and easy as some if the other DBs.
- prioritisation of queries or users. It's hard to say "prefer this query over this one, give it more resources" or "user A's work always preempts user B's" in Pg.
- transparent failover and recovery back to the original master.
- shared-storage clustering. Dunno if anyone still cares about this one though.