> RAID-1-nel az olvasasi sebesesseg elvileg duplaja lehetne a fizikai
> diszk-nel (ha ket fizikai diszk alkotja a RAID-1-et), hiszen a 2
> fizikai
Elvileg. Gyakorlatilag nem lattam meg ilyen implementaciot se szoftver
(linux), se hardver RAID megoldasokban.
> Solaris DiskSuit, HP-UX LVM, es sok mas esetben ez mukodik is
> kivaloan.
ezekkel meg nem talalkoztam kozelebbrol :) Kivancsisagbol, mi tud
meg ilyet?
> Linux-nal hogy all az ilyen?
aszongya, drivers/md/raid1.c, read_balance(), 2.6.15.6:
* This routine returns the disk from which the requested read should
* be done. There is a per-array 'next expected sequential IO' sector
* number - if this matches on the next IO then we use the last disk.
* There is also a per-disk 'last know head position' sector that is
* maintained from IRQ contexts, both the normal and the resync IO
* completion handlers update this position correctly. If there is no
* perfect sequential match then we pick the disk whose head is closest.
Igyekszik intelligens lenni, de ugy tunik, pont hogy ugyanazt a diszket
preferalja. Ugyanez csucsul a raid10.c-ben, plusz:
/*
* FIXME: possibly should rethink readbalancing and do it differently
* depending on near_copies / far_copies geometry.
*/
a dm-raid1.c-ben meg ez van:
static struct mirror *choose_mirror(struct mirror_set *ms, sector_t
sector)
{
/* FIXME: add read balancing */
return ms->mirror + DEFAULT_MIRROR;
}
ami egy eleg szuk gyerekciponek tunik. :-)
raas
ps. multicount-ot kapcsolgasd be a diszkeken.
--
Those who say it cannot be done should not interrupt the person doing it.
-- Chinese proverb
_________________________________________________
linux lista - [email protected]
http://mlf2.linux.rulez.org/mailman/listinfo/linux
