Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site ucbvax.BERKELEY.EDU
Path: utzoo!watmath!clyde!cbosgd!ucbvax!info-vax
From: MHJohnson@HI-MULTICS.ARPA (Mark Johnson)
Newsgroups: mod.computers.vax
Subject: Creating volume sets
Message-ID: <860304162431.700172@HI-MULTICS.ARPA>
Date: Tue, 4-Mar-86 11:24:00 EST
Article-I.D.: HI-MULTI.860304162431.700172
Posted: Tue Mar  4 11:24:00 1986
Date-Received: Fri, 7-Mar-86 07:31:35 EST
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The ARPA Internet
Lines: 27
Approved: info-vax@sri-kl.arpa


  Our site has 4-RA81 disks and were having problems with balancing disk
usage across the drives.  We recently created a 3-disk volume set for
all our user files and the following occurred:

  (1) Backed up our disks with
BACKUP/NOCRC/BUFF=5/IMAGE/FAST/BLOCK=16000 etc.  This operation on our
TU78 at 6250 bpi took between 1 and 2 hours per disk drive.  Our main
user disk had about 50k files with heavy fragmentation and took the
longest.

  (2) Created the volume set (INIT times 3) and MOUNT/BIND to make the
volume set.  While we were at it, we rebooted to checkout our system
startup to make sure it was ok.

  (3) Restored the files onto the volume set.  The same sort of backup
command was used as for the save (BACKUP/NOCRC/BUFF=5/FAST...)  except
/IMAGE was omitted.  Well, things got very boring when the system took
about an hour and a half (!!!)  per tape (that is well over 3 hours per
disk) to restore the files

Needless to say, staying up until 1:30am because the restore operation
turned out to be much slower than the save operation is not my idea of
fun.  If someone out there (is Andy listening?)  has an idea why the
restore took so much longer than the save, I'd like to hear about it.
Of course, if you want to do this sometime, you'd like to know too.
  --Mark <MHJohnson @ HI-MULTICS>