Last year, we had to migrate from our 947 to a new 929 for the Year 2000 project.
At that time, I had the luxury of "playing" with a new system, while production was running on the old one.
There were two benefits: 1) I was able to perform a disaster recovery rehearsal (five times!);
2) We were able to split our system into different volume groups.

Now, the five step process is solid, and it will reorganize your data.
I do not know if that will improve your response time.
One of the things to remember is to take the old *log* files out of pub.sys--
you do not need them and they clutter up things.

There has been a great advantage to us to separate the data into volume groups:
System, Production, Test, and we added Tech (for quasi system stuff).

If our system crashes, we may only have to rebuild the system volumes.
If the TEST or TECH volumes crash, we can keep running production.

Also, since the production stuff is the most likely stuff to become fragmented,
you don't have to RELOAD the system to reorganize the production volume groups.

I realize that this may not work for you, and there may be more detail you may want (for example,
using BULDACCT to set up the new volume groups and move production data to them),
but you might find this helpful.

>>> bArtoque <[log in to unmask]> 09/11 7:06 PM >>>
Howdy,

I know it probably has been in the list a couple of times (understatement, I
think...), but (ther's always a but...) I'm gonna ask it anyway. We're
goiung to do a system reload, but have'nt done it EVER. My problem is that a
lot off people have a lot of opinions about what's good/better/best, but
what whould be a good STORE of all files on the sytem and a good
restore/reload procedure to get rid of disk fragmentation? Before the reload
(our client wants us to do it anyway, because they hope for performance
increase instead of bying a new machine) we wanna try "VOLUTIL > CONTIGVOL"
to get rid of a little bit of fragmenation (I now it's no real defrag and we
haven't done this before either ). Should or shouldn't I use buldjob or
other procedures? Has someone maybe put on paper a nice strategy of doing
the store/reload. S'thing like I found in the backissues:

0) Kick everyone off, and shut down networking (so no files will fail to
STORE because they are in use for writing.
 1) As MANAGER.SYS, :RUN BULDACCT;INFO="@"
2)     :FILE T;DEV=TAPE
         :STORE @[log in to unmask]@;*T;DIRECTORY;SHOW
3) VSTORE (or however you like to verify your tapes), on a different tape
drive, if possible.
4) If you have time, make a second backup (I don't trust tapes)
5) Go into SYSGEN, and make a new SLT

When you're ready to reload:
1) Boot from the SLT
2) Use VOLUTIL to add the rest of your disc drives
            :VOLUTIL
            volutil: NEWVOL volumesetname:membername LDEV=nn
3) :RESTORE *T;@[log in to unmask];CREATE;KEEP;OLDDATE;SHOW=OFFLINE
            :FILE T;DEV=TAPE
            :RESTORE *T;@[log in to unmask]@;KEEP;CREATE;OLDDATE;DIRECTORY
4) :STREAM BULDJOB1.PUB.SYS
5) :RESTORE *T;/;CREATE;KEEP;OLDDATE;SHOW=OFFLINE

What bothers me, is to actually make a reload, you have to purge EVERYTHING
from the system (maybe only not @.pub.sys or so, cause you probably won't
win that much defrag and still have the operating working system?!?).

Please try to spare a couple of moments of your valueble time to help me
(and probably others), cause I'm now at the point of calling HP for help,
for s'thing which should be a simple (but timeconsuming) system task.

Th@nx in advance,

bArtoque