ADSM-L

Re: MVS ADSM Server and HSM

1994-12-19 15:16:01
Subject: Re: MVS ADSM Server and HSM
From: Bill Colwell <BColwell AT CCLINK.DRAPER DOT COM>
Date: Mon, 19 Dec 1994 20:16:01 GMT
In <1994Dec19.194421.1211 AT draper DOT com>, KL2%NIHCU.BITNET AT vm.marist DOT 
edu (Bob Klein) writes:
>On November 25th I submitted a request to the adsm-l list
>regarding the use of hsm to manage adsm disk storage pools on an
>MVS ADSM server.  The reason for the request was so that we would
>not need to have two series of tapes, one for hsm and one for
>adsm.  I got a reply from IBM, which mentioned the new EXPIRE
>INVENTORY command and the fact that automatic file expiration can
>now be turned off.  The result of using this facility is that
>storage volumes are not accessed except by an adsm client and,
>therefore, hsm would eventually move the storage volumes to a
>lower level in the storage hierarchy and eventually to tape.
>
>We have found, however, that whenever ADSM is restarted (which
>happens whenever the system it runs on is IPLd) all the storage
>volumes are opened.  This obviously defeats the goal of letting
>hsm manage the disk storage volumes.
>
>I would be very interested to hear from other sites running the
>ADSM server on an MVS platform and are using HSM whether they
>share our interest in using HSM to manage ADSM storage volumes
>rather than setting up a separate tape storage pool for ADSM.  In
>the apparent absence, currently, of the ability to use HSM to
>manage ADSM storage volumes, what has been the experience of MVS
>shops in the use of tape storage volumes for ADSM (e.g. what sort
>of operator load is there when inventory expiration is carried
>out).  Thanks for any input.
>
>------------------------------------------------------------------
>Robert P. Klein                          KL2 AT CU.NIH DOT GOV
>Phone: 301-496-7400                      Fax: 301-496-6905
>Mail:  DCRT/CFB/ETS, 12A/1033, 12 SOUTH DR MSC 5607,
>       BETHESDA, MD 20892-5607
>------------------------------------------------------------------
>
We have both HSM and ADSM.  2000 tapes are assigned to the HSM ML2
pool and 1000 tapes are assigned to the ADSM pool.  There are 2 ADSM
tape pools -- 1 is for 3480 tapes (18 track) and the other is for 3490
tapes (36 track).  We are in the process of doing MOVE DATA commands
on the 18 track tapes to convert the pool to all 36 track format.

There is NO interaction between the tape pool and the client inventory
delete (CID). For the disk pool, CID immediately updates the disk
pool space maps, but for tape resident file versions, CID just updates
database resident informatiaon about the tape.  Therefore, there is no
'operator load' for tapes during CID.

Here are some statistics about my ADSM system (disk to tape migration
is going on at the moment, forced by the hi and low mig values for
BACKUPPOOL)  -

adsm> q db

Available Assigned   Maximum   Maximum    Page     Total      Used %Util  Max.
    Space Capacity Extension Reduction    Size    Usable     Pages       %Util
     (MB)     (MB)      (MB)      (MB) (bytes)     Pages
--------- -------- --------- --------- ------- --------- --------- ----- -----
    4,132    4,132         0       684   4,096 1,057,792   881,152  83.3  83.4
    4,132    4,132         0       684   4,096 1,057,792   881,152  83.3  83.4

adsm> q stg

Storage      Device       Estimated  %Util  %Migr  High   Low  Next
Pool Name    Class Name    Capacity                Mig%  Mig%  Storage
                               (MB)                            Pool
-----------  ----------  ----------  -----  -----  ----  ----  -----------
ARCARTPOOL   CARTRIDGE     22,056.4   33.6   40.0    90    70
ARCARTPOOL   CARTRIDGE     22,056.4   33.6   40.0    90    70
ARCHIVEPOOL  DISK             512.7   94.8    0.1    90     0  ARCARTPOOL
BACKUPPOOL   DISK          17,635.0   99.1   54.7    50    40  BKCT36POOL
BKCARTPOOL   CARTRIDGE    144,580.2   58.0   91.3    90    70
BKCT36POOL   CARTRIDGE-   167,944.3   64.4   84.3    90    70
              36
ZZZ          DISK               0.0    0.0    0.0    90    70

Bill Colwell
C. S. Draper Lab
Email: BColwell AT draper DOT com
Voice: 617-258-1550
<Prev in Thread] Current Thread [Next in Thread>