"Fabrizio Magni" <fabrizio.magni@(protected)
a.m.:

> Hello Herman,
>
>
> > Furthermore, if one of these processor hangs you get an automatic
reboot
> > and until these processor is replaced you cannot startup the box.
>
>
>
> This is interesting.
> Can you point me to some published bug or something in the AMD support
about
> this issue?
Is not a bug, is the Opteron architecture, all processors have a path with
the other for snoops requests, IBM enginners told me they help to AMD with
redefine this situation trying to cross paths to have redundancy between
processors.
>
> We test IBM xSeries servers with X3 chipset and there are more performant
> > in OLTP real situations because they have a memory controller with the
> > pointer of all data in processor and avoid snoop request.
> > In summary AMD Opteron is faster than Intel with 8500 chipset (HP, Dell
> > and
> > many others) and slower than Intel with X3 chipset in OLTP
transactions.
>
>
>
> Have you got any benchmark on this?
http://www.spec.org/jbb2000/results/res2005q2/jbb2000-20050324-00322.html
publish 167515 ops/s for IBM x366 4-WAY (we test this server)
http://www.spec.org/jbb2000/results/res2005q1/jbb2000-20050315-00310.html
publish 116142 ops/s for Sun Fire V40 Opteron.
> Can you describe the tests you performed?
We install SAP R/3 Enterprise 4.70 with Oracle 9.2.0.6 for 64bits in an
x366 4-way EM64T with X3 chipset and performs many typical transactions
from SD module and compare with the same database with the same data in an
HP DL585 with AMD Opteron, and with the same transactions we saw 20%
greather performance in x366 box.
In both cases connected to an FC external storage Hitachi AMS500
Tagmastore, and Suse SLES9 SP2.
> Really glad
> Fabrizio
Regards
Herman


--
To unsubscribe, email: suse-oracle-unsubscribe@(protected)
For additional commands, email: suse-oracle-help@(protected)
Please see http://www.suse.com/oracle/ before posting