> So, yes, clock-for-clock (and for my usage) Xeon 51xxs are > faster than > Opterons. But, if your code hits memory *really hard* (which > that heart > model does), then the multiple paths to memory available to > the Opterons > allow them to scale better.
Yes, and this is consistent with stream numbers we were discussing a while ago. Enabling cache snoop filter can pump up stream numbers considerably but unfortunately I find its effect on application performance to be erratic: from -10% to +4%. If the original poster goes for Xeon I'd suggest considering to pay a little premium (20 pounds) and upgrade to 667 MHz FB DIMMs. _______________________________________________ Beowulf mailing list, Beowulf@beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
