> Mark A wrote:
>
> > So are you or Daniel claiming that Oracle RAC would have beaten the
> > DB2 benchmark with the slightly faster chips now available from Intel?
>>
> "Mark Townsend" <markbtownsend_at_comcast.net> wrote in message
> I'm certainly not, not because I know, but because I don't know. Early
> performance testing of Oracle on similar IBM hardware as used in the DB2
> benchmark, however, is very encouraging. And that should not be
> construed as an indication of anything to come, because it is not.
>
As I mentioned, I think that would be interesting. But would that include
using RAC on a single node with 64 processors?
I personally doubt that there is much difference in performance in the 2
databases on the same hardware (and I have state this many times before),
but I think that there are performance differences with RAC (which DB2 does
not have and therefore cannot be compared). Clearly, RAC does have some
other non-performance advantages, but if the difference in performance is
2.5-1 than other high availability strategies may be interesting.
Received on Sat Jan 29 2005 - 15:45:55 CST
