Hi,
We run Red Hat 5.3. Our production LPARs use FCP attached SCSI for local
file systems, SCSI managed via EDEVs for the OS install (/ and /boot)
and 3390's for VM. It takes a lot of planning up front for the SCSI box
configuration. 50G and 100G are good LUN sizes.  Good I/O performance
requires that there be isolation in the backend real drives for groups
of LUNs. For example, real drives backing up db2data LUNs should be
separate from other file system LUNs. Get lots of real disks.
We use multipathing in failover mode as performance is better than
multibus mode.  We also use the data=writeback option on the mount
command. Also use LVM striping.
Betsie

-----Original Message-----
From: Linux on 390 Port [mailto:[email protected]] On Behalf Of
O'Brien, David W. (NIH/CIT) [C]
Sent: Wednesday, February 18, 2009 7:35 AM
To: [email protected]
Subject: Size of Linux DASD under Z/VM?

Hi,

  We're new to the Linux world. Currently we have 3390 Mod-3 (2.8G) and
Mod-9s (8.4 G) defined to our Linux guests under Z/VM.

  I'm starting to get complaints that these 'partitions' are too small
and do not compare favorably with the dasd sizes available on PC
desktops. (Apples and Oranges, in my opinion.)

  What I'd like to know from the group is - What are your DASD sizes?
Are the complaints about partition sizes valid?

We have a segment of 1.8TB that could be re-configured. I've suggested
50Gig Mod-54s but have been told that's too small.

Thank you,
Dave O'Brien
NIH Contractor
________________________________________
----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions, send
email to [email protected] with the message: INFO LINUX-390 or
visit http://www.marist.edu/htbin/wlvindex?LINUX-390

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to