Re: Four questions regarding migration, reclamation, utiliza
1998-05-05 13:57:39
Subject: |
Re: Four questions regarding migration, reclamation, utiliza |
From: |
ALLEN BARTH <abarth AT KEMPER DOT COM> |
Date: |
Tue, 5 May 1998 12:57:39 -0500 |
Hi Jennifer,
I'm on AIX but the ?'s you asked don't seem platform related, so here
goes.
1. The default for disk storage pools is cache=yes which means that
even though things are migrated to the next storage pool the data is
left in the dasd (for faster restores) unless the disk space is needed
for new backup data. You can either A) ignore the %used, b) upd
stgpool to cache=no.
2. If you primary tape pool has no next pool then the high & low %mig
can be ignored because there's no place to migrate to. I run reclaim%
at 60, which is where the trial-and-error / bang-for-the-buck/
proccessing-time-vs-storage-vols-uesd works in MY environment. (I
share a 3494 with OS/390).
3. I haven't seen a case where any of my 3490E's don't fill
completely, save colocation and 'filling' tapes. A word of caution
here. You have specified hardware compression for the drives, so the
numbers returned to you by adsm are what it 'thinks' it managed to put
on a tape vol, that is why you see differing amounts for est. capacity
which gets updated when the tape becomes 'full'.
4. In a complete disaster, the adsm utils can recover the server to
the point of the last backup. The VM backups will only be good if you
quiesce adsm before you do them.
Hope this helps,
Al Barth
Scudder Kemper Investments
______________________________ Reply Separator _________________________________
Subject: Four questions regarding migration, reclamation, utilization
Author: "Macek;Jennifer" <Jennifer.Macek AT ABBOTT DOT COM> at ~internet
Date: 5/5/98 11:29 AM
Hello fellow ADSMers!
We are currently running ADSM v2r1 on a VM/ESA 2.1.0 system with
Sterling's VM:Tape product.
Are tapes are being stored in 5 Storage Tek silos that we share with our
MVS system, so we have
a limited amount of space. We seem to be on a constant quest for more
resources, and I wanted
to know if there is anything I can do to make better use of the resources
I'm currently using.
My questions are:
1) The high migration level we are setting for our stgpools doesn't seemed
to be clearing the data off
of the minidisk. We have 6 mod 3 DASD defined to one of our ADSM
servers with the highm=10 & lowm=5 and we're still at
99.2% utilized. This particular server backs up approximately 540
workstations and we are soon to get 400-800 more workstations.
Is there anything else I can do to get better utilization out these
DASD volumes. I can add more DASD for a quicker solution, but I
am interested in long term solution if I can find one. Any input
would be appreciated.
Storage Device Estimated
%Util %Migr High Low Next
Pool Name Class Name Capaci
ty Mig% Mig% Storage
(MB)
Pool
----------- ---------------
--------------- ------ ------- ------ ------ -----------
BACKUPPOOL DISK 14,067.4
BACKUPPOOL DISK 14,067.4
99.2 11.1 10 5 ETAPE
2) Additionally we're currently utilizing 536 etapes for this server with a
reclamation threshold at 50%.
If I lower the highm, the lowm and the reclamation thresholds will this
save me some tape utilization?
Can anyone in ADSM land give me some ideas on what is working for
them?
Storage Device Estimated
%Util %Migr High Low Next
Pool Name Class Name
Capacity Mig% Mig% Storage
(MB)
Pool
----------- ---------------
--------------- ------ ------- ------ ------ -----------
ETAPE ETAPES 436,754.9
ETAPE ETAPES 436,754.9
58.9 82.6 70 40
3) Is there a way I can check these tapes to see if they are utilizing the
entire length of the etapes?
Here are the current settings in our device class for this server:
Device Class Name: ETAPES
Device Access Strategy: Sequential
Storage Pool Count: 2
Last Update by (administrator):
SERVER_CONSOLE
Last Update Date/Time: 01/15/1998 10:34:20
Device Type: Cartridge
Maximum Capacity (MB):
Estimated Capacity (MB): 900.0
Dataset Name Prefix: ADSM
Mount Limit: 8
Mount Retention (min): 0
Label Type: IBMSL
Expiration Date: 99365
Mount Wait (min): 60
Density:
Format: 3490C
Length: ECCST
Display: Alternate
4) Last but not least. We are currently backing up the storage pools &
databases using ADSM's back up utilities, but I've also been backing up these
volumes with Sterling's VM:Backup product to get a backup of the
minidisk itself. Is this really necessary? In a worst case scenario would I
be able
create new minidisks, define them to the failing server and restore the
DB and storage pools just from the ADSM backup?
Any help would be greatly appreciated,
Thanks,
Jennifer Macek
Jennifer.Macek AT Abbott DOT com
|
<Prev in Thread] |
Current Thread |
[Next in Thread> |
- Re: Four questions regarding migration, reclamation, utiliza,
ALLEN BARTH <=
|
|
|