nizky vypocetni vykon na 2CPU stroji (delsi)

[Matus UHLAR - fantomas]

> news.volny.cz napsal(a):
>> mam tu dva stroje, tzv. ,,stary'' s 2 x Intel(R) Pentium(R) III CPU - S 
>> 
>> 1266MHz s 1GB RAM a jadrem 2.4.22-10mdkenterprise, system Mandrake9.2 a
>> pak tzv. ,,novy'' s 2 x Intel(R) Xeon(TM) CPU 2.80GHz s 2GB RAM a jadrem
>> 2.6.11-1.14_FC3smp, system FedoraCore 3.
>> 
>> Zarazi me nizky vypocetni vykon noveho stroje, oba stroje davaji totiz
>> temer totozny vysledek napriklad u primitivniho testu
>> 
>> # time echo 2^1000000 | bc > /dev/null

Michal Dobes <dobes na tesnet.cz> wrote:
> Je otazka, zda bc je zrovna vhodny test. Pro porovnani jsem
> to pustil na par strojich, co mi tu poletuji:
> 
> 2x Opteron 246/1 MiB/2 GHz, 4 GiB RAM, SLES9, 2.6.5-7.151-smp
>  8.92user 0.01system 0:08.94elapsed
> 
> 2x AthlonMP/256 KiB/1,2 GHz, 1 GiB RAM, RH7.3, 2.4.18-18.7.xsmp
>  15.75user 0.01system 0:17.04elapsed
> 
> 2x Pentium III/256 KiB/1 GHz, 1,5 GiB RAM, RH7.3, 2.4.18-18.7.xsmp
>  17.40user 0.02system 0:17.44elapsed

a ako spomenuli "moje kalhotky", dolezite je testovat:

echo 2^1000000 | time bc > /dev/null

aby ste zistili kolko casu trva prevedenie prikazu "bc" a nie prikazu "echo"

-- 
Matus UHLAR - fantomas, uhlar na fantomas.sk ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
Eagles may soar, but weasels don't get sucked into jet engines.