Re: Comparison of PGSQL and DB2

On Tue, 2004-03-09 at 13:22, Josh Berkus wrote:
> Shashank:
>
> What follows is my reply to IBM's 2-year-old feature comparison of PostgreSQL
> and DB2.   Each of the entries is an item that DB2 has and claims that
> PostgreSQL does not.    I've put my replies to these claims.
>
<snip>
>
> > Isolation Levels
> > (Support all four ANSI isolation levels (UR, CS, RS, RR).)
>
> I'm not sure about this one; I suspect that we do, however, since MVCC,
> invented for the Postgres Project, has become a standard for transaction
> isolation in the database industry.
>

Not exactly sure which acronyms above correspond to which levels (as I
know them by different names apparently) but we support Read Committed
and Serializable levels, which are by far the most common of the two
AFAIK. The others as I know them are Read Uncommitted aka phantom reads,
and Repeatable Read, which we do not support. Of course we also supply a
extensive locking methods for further control.

More info on our isolation levels can be found at
http://www.postgresql.org/docs/7.4/static/transaction-iso.html


<snip>
> > Federated Database Support
> > (Ability to allow applications to access & perform JOIN operations on
> multiple disparate databases.)
>
> This is a feature which we do not have because it violates the ANSI SQL
> Specification.
>

Perhaps worth a mention that using dblink, you can actually make
functionality similar to this.  I have recently been experimenting with
using views calling dblink functions to create "local tables" that
really live in a separate database on a completely separate machine. I
don't know if I would recommend the technique but I think the
capabilities are there.

Robert Treat
--
Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL