Re: Logical striping over Physical Striping
From: Gary Jahrig (gary_at_X.usnetcastX.com)
Date: 12/30/03
- Next message: Peter Smithson: "Make: Unable to fork. on HPUX 11"
- Previous message: Paul Sure: "Re: Logical striping over Physical Striping"
- In reply to: Paul Sure: "Re: Logical striping over Physical Striping"
- Next in thread: Russ: "Re: Logical striping over Physical Striping"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Tue, 30 Dec 2003 07:34:55 -0800
"Paul Sure" <phv_temp@yahoo.com> wrote in message
news:vv2q848ql0k375@corp.supernews.com...
> The documentation of those high end arrays advise against it. It states
that
> logical striping over a raid lun will confuse the smart caching algorithm
> used to destinguish between random and sequential reads/writes.
>
> In theory it is even possible to reverse the hardware striping with
logical
> striping with as result no striping at all.
>
> Logical striping over separate hardware raids is no problem. Once I
managed
> an IBM Shark with about 20 10 Gb luns on separate raid sets. I made one
big
> logical striped LV with it. sequential reads/writes did not much improve
but
> I never got the luns busy above 2% :).
>
> Paul
Hey thanks Paul. Could you point me to any of that documenatation. Ive
looked and cant find it.
- Next message: Peter Smithson: "Make: Unable to fork. on HPUX 11"
- Previous message: Paul Sure: "Re: Logical striping over Physical Striping"
- In reply to: Paul Sure: "Re: Logical striping over Physical Striping"
- Next in thread: Russ: "Re: Logical striping over Physical Striping"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]