ADSM-L

Re: Question about ADSM Server Capacity

1997-05-29 14:02:10
Subject: Re: Question about ADSM Server Capacity
From: dan thompson <thompsod AT USAA DOT COM>
Date: Thu, 29 May 1997 13:02:10 -0500
TammyKat,

  We have broken our ADSM MVS servers up into separate started tasks.  The
main reason was our concern over the size of the database and the need for
a large number of concurrent sessions.  As one of our existing databases is
growing we are attempting to come up with a rule-of-thumb as to when we
need to separate again. Due to the somewhat extended amount of time that a
recovery of a database can take we are probably going to let this drive the
decision.  I hope someone can give you some more specific hints.

Our database sizes are:
8784mb - 69.4%
6488 mb - 64.4%
17484mb - 57.6%

Dan T.
----------
> From: Tammy Kilgore <usbbktbk AT IBMMAIL DOT COM>
> From: Tammy Kilgore <usbbktbk AT IBMMAIL DOT COM>
> To: ADSM-L AT VM.MARIST DOT EDU
> Subject: Question about ADSM Server Capacity
> Date: Thursday, May 29, 1997 8:55 AM
>
> Tammy R. Kilgore
> Barnett Bank, NA
> 904-464-6405
> SUBJECT: Question about ADSM Server Capacity
>
> I am currently running one ADSM for MVS server as follows:
>
> ADSM Database:                2,340 MB  71% utilized
> ADSM Recovery Log:              800 MB  (Roll Forward Recovery)
> ADSM Nodes Registered:          308
> Average Nightly Backup:          20 GB
> Number of files managed:  3,192,964
> Space managed:                  400 GB
> Max hour CPU util during backup 50% (of one ES/9021-982 60 MIP engine)
> Expiration runs an average of 2 hours once a week
> Maximum number of simultaneous ADSM sessions averages around 30
> ADSM Region Size (under MVS) 128 MB
> Most all of our clients use TCP/IP for connectivity
>
> I will be adding an additional 300 plus nodes over the next year.  I
> am trying to determine at what point I should bring up an additional
> ADSM for MVS server to help distribute the workload.  I am more
> concerned about capacity issues specific to MVS such as virtual storage,
> CPU utilization, ADSM database size, etc. as opposed to network
> capacity issues.
>
> My questions are:
>
> 1. Has anyone had to bring additional ADSM for MVS server(s) to
>    handle capacity?
>
> 2. If so, what factors caused you to decide to bring up additional
>    ADSM server(s)?
>
> 3. Which resources under MVS should I monitor to help me determine
>    when I should bring up an additional ADSM server to handle the
>    workload?
>
> I am trying to be proactive.  I don't want to wait until I have nodes
> missing scheduled backups to know I have a capacity problem.
>
> By the way,  I have looked at the ADSM Capacity Planning Tool (Release
> 1 Level 0.0) and it did not provide me with any useful information.
>
> Thanks,
> Tammy Kilgore
>
> ****************************************
> * IBMLink: SBTTCEU  at ELINK           *
> ****************************************
<Prev in Thread] Current Thread [Next in Thread>