l i n u x - u s e r s - g r o u p - o f - d a v i s
L U G O D
 
Next Meeting:
January 6: Social gathering
Next Installfest:
TBD
Latest News:
Nov. 18: Club officer elections
Page last updated:
2001 Dec 30 17:03

The following is an archive of a post made to our 'vox-tech mailing list' by one of its subscribers.

Report this post as spam:

(Enter your email address)
Re: [vox-tech] Totally inappropriate question (about HP)
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [vox-tech] Totally inappropriate question (about HP)



Firstly the application is Oracle.  There is a new(ish) paper out about
configuring drives called "Optimal Storage Configuration Made Easy"
(http://technet.oracle.com/deploy/performance)

Machine: K220 2way
scsi interface: I don't know what that means, It has 2 scsi adaptors which
connect to the raid cabinet.

I (the paper suggests) am minimizing head travel.

Jay

----- Original Message -----
From: "Bill Broadley" <bill@math.ucdavis.edu>
To: <vox-tech@franz.mother.com>
Sent: Tuesday, April 17, 2001 5:58 PM
Subject: Re: [vox-tech] Totally inappropriate question (about HP)


> On Tue, Apr 17, 2001 at 05:29:45PM -0500, Jay Strauss wrote:
> > Hoping that some HP sys admins may be part of this list.
>
> Heh, a few, I just got hpux 11.11 today.
>
> > I'm running HP UX v11.0.0, using a HP Model 20 disk array.
>
> What kind of machine?  What kind of scsi interface?
>
> > Question: How do I build my volume groups out of only the outer
cylinders of
> > the disks?  And, do I do it when I'm creating my physical volumes,
physical
> > volume groups, or my logical volumes?
>
> Why?  Bandwidth for arrays is usually limited by the I/O bus or the
> SCSI bus.  SCSI disks can arbitrary remap sectors for numerous reasons
> (hard errors, soft errors, ECC coverage).  Each new disk these days is
> good for 30 MB/sec or so.  Get 8 going and your quickly exhausting the
> PCI bus, an Ultra 160 Channel, or even main memory bandwidth.
>
> I'd just make a volume of the first 1/4th of the block, the 2nd 1/4th, 3rd
> and 4th.  Do some bandwidth tests, if theres a difference then the lower
> perf ones are the inside tracks, and the higher perf are the outside.
>
> As long as the array is configured for parallel use of the raid and
> not sequential (wierd) it should strip the blocks, the stripsize
> is an option during volume creation I believe.
>
> If there is no difference in bandwidth do you care?  I guess you
> could be doing some wierd minimizing of head travel for slightly
> lower seek times.  Of course the usual method is to add more heads...
>
> --
> Bill


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


LinkedIn
LUGOD Group on LinkedIn
Sign up for LUGOD event announcements
Your email address:
facebook
LUGOD Group on Facebook
'Like' LUGOD on Facebook:

Hosting provided by:
Sunset Systems
Sunset Systems offers preconfigured Linux systems, remote system administration and custom software development.

LUGOD: Linux Users' Group of Davis
PO Box 2082, Davis, CA 95617
Contact Us

LUGOD is a 501(c)7 non-profit organization
based in Davis, California
and serving the Sacramento area.
"Linux" is a trademark of Linus Torvalds.

Sponsored in part by:
O'Reilly and Associates
For numerous book donations.