[Beowulf] [gorelsky at stanford.edu: CCL:dual-core Opteron 275performance]

Mikhail Kuzminsky kus at free.net
Tue Jul 5 09:33:56 PDT 2005


In message from Alan Louis Scheinine <scheinin at crs4.it> (Tue, 05 Jul 
2005 14:07:20 +0200):
>With regard to performance of dual-core Opteron, your milage may 
>vary.
>I just finished a benchmark using a program that needs a high
>bandwidth to memory. 
Yes, it's good idea.

> A quad-CPU board with single-core Opteron was
>nearly twice as fast as a dual-CPU board with dual-core Opteron,
But this result means, that 4 cores of Opteron are "equal by 
performance" to 2 "single core" Opterons. If it'll be *exactly*,
your program looks as working "only" w/RAM (I suppose that
memory throughput don't scale from single core Opteron to 2-cores 
chip, what is, generally speaking, incorrect), and there is
practically no "memory-independed" computations !

But I understand that some typical finite elements theory loops  
may contain a lot of memeory references.

I plan to receive Opteron275 dual cpu server in pair of days,
and also to perform a set of tests.

Yours
Mikhail

>in each case four MPI processes; compiled with MPICH with PGI;
>in every case 2.2 GHz CPU frequency.  The program is meteorology,
>short-term weather prediction.
>
>-- Alan
>
>  Centro di Ricerca, Sviluppo e Studi Superiori in Sardegna
>  Center for Advanced Studies, Research, and Development in Sardinia
>
>  Postal Address:               |  Physical Address for FedEx, UPS, 
>DHL:
>  ---------------               | 
> -------------------------------------
>  Alan Scheinine                |  Alan Scheinine
>  c/o CRS4                      |  c/o CRS4
>  C.P. n. 25                    |  Loc. Pixina Manna Edificio 1
>  09010 Pula (Cagliari), Italy  |  09010 Pula (Cagliari), Italy
>
>  Email: scheinin at crs4.it
>
>  Phone: 070 9250 238  [+39 070 9250 238]
>  Fax:   070 9250 216 or 220  [+39 070 9250 216 or +39 070 9250 220]
>  Operator at reception: 070 9250 1  [+39 070 9250 1]
>  Mobile phone: 347 7990472  [+39 347 7990472]
>
>_______________________________________________
>Beowulf mailing list, Beowulf at beowulf.org
>To change your subscription (digest mode or unsubscribe) visit 
>http://www.beowulf.org/mailman/listinfo/beowulf




More information about the Beowulf mailing list