ADSM-L

Re: HELP!... What processes can/can't run simultaneously andwhat throttles throughput on backup sessions ?

2002-09-24 16:28:19
Subject: Re: HELP!... What processes can/can't run simultaneously andwhat throttles throughput on backup sessions ?
From: David Longo <David.Longo AT HEALTH-FIRST DOT ORG>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Tue, 24 Sep 2002 16:26:16 -0400
Another simple test to do to prove to Network guys and others
that TSM is not the problem is to do some FTP's from some of the
clients to the TSM server.  Performance here is strictly
network/CPU/disk
not TSM.  Only exception is if you have heavy backups or restores
going at the time.

David Longo

>>> matt AT IDEXX DOT COM 09/24/02 03:20PM >>>
Check your network settings on your server. Some network cards do not
Auto-negotiate with certain switches. If everything is set to
Auto-negotiate, set the network card and the port of the switch it is
plugged into to 100 MB/s Full duplex. I was getting the same throughput
on
some of my TSM servers until I did that and then I started to get 8-9
Megabyte/sec performance. Your performance may vary.




Thank You,
Matt Martinez
Workgroup Support Analyst
IDEXX Laboratories, Inc
207-856-0656
matt AT idexx DOT com

 -----Original Message-----
From:   Barbara Andrews [mailto:BAndrews AT ERIE1.WNYRIC DOT ORG]
Sent:   Tuesday, September 24, 2002 3:06 PM
To:     ADSM-L AT VM.MARIST DOT EDU
Subject:        HELP!... What processes can/can't run simultaneously
and
what throttles throughput on backup sessions ?

     Up until a few days ago, we were backing up only small amounts of
data
for several nodes each night.  Then an emergency request came in to
start
nightly incremental backups using the TSM Domino TDP for a Domino mail
server that does not do transaction logging. (Note: The Domino support
staff estimates implementation of transaction logging 3-4 weeks from
now.... I wish it were sooner!) All of a sudden we went from a total
backup
of no more than 11GB per night to more than double that, with almost
20-23
GB  per night coming from that one new node alone.  This node's data
is
sent over a 100 MB ethernet connection.  Even with client compression
turned on, the best throughput rate we've seen is 566 Kb/sec., and that
was
when the backup was run on a Sunday when there is very little activity
on
all systems involved (client, server, network).  When other backups
are
running during the night, the best we've seen is 368.72 Kb/sec.  This
results in the backup for this one node taking at least 17 hours to run
to
completion.
     Before the addition of this new node, I had all my processes
(expiration, reclamation, migration, backup storage pool) running
separately and not at the same time as client backup sessions.  Now we
no
longer have enough time in the day to do this.  I need to change the
way my
processes run, but I have several questions I need answered before I
do
this.  If anyone can help me with the following questions, I would
appreciate it:
1) What are the implications of the process generated when I issue
"backup
stg backuppool offsite-pool" running at the same time that a node is
currently in session backing up to the "backuppool" storage pool?
      1a) Will the backup stg process finish during the node's
session?
2) Do we need to be in a state where there are no active processes
and/or
backup/restore sessions running while the TSM database backup is
running?
      2a) What kind of state would the TSM database be in if it was
restored from a backup that had other processes or backup/restore
sessions
running at the time it was taken?
3) Is there any problem with running an "expire inventory" while any of
the
following are running?:
     - a node backup session
     - a node restore session
     - migration
     - reclamation
4) Is there a problem running migration and reclamation at the same
time if
they both access the same pool?  I would have thought the only problem
might be a wait for a tape while one process is using it, but that
this
probably would only occur occasionally.  I have a reclamation process
running that caused a message "Removable volume 30050 is required for
space
reclamation."  A migration process is running which requires writing to
the
same tape pool that 300050 is in, however 300050 displays as last
being
written to the day before.
5) I have very little knowledge of networking.  Could someone comment
on
the throughput values we are seeing?  Our max of 566 Kb/sec is no
where
near the 100 Mb/sec I have been told our ethernet connection is.


"MMS <health-first.org>" made the following
 annotations on 09/24/2002 04:27:35 PM
------------------------------------------------------------------------------
This message is for the named person's use only.  It may contain confidential, 
proprietary, or legally privileged information.  No confidentiality or 
privilege is waived or lost by any mistransmission.  If you receive this 
message in error, please immediately delete it and all copies of it from your 
system, destroy any hard copies of it, and notify the sender.  You must not, 
directly or indirectly, use, disclose, distribute, print, or copy any part of 
this message if you are not the intended recipient.  Health First reserves the 
right to monitor all e-mail communications through its networks.  Any views or 
opinions expressed in this message are solely those of the individual sender, 
except (1) where the message states such views or opinions are on behalf of a 
particular entity;  and (2) the sender is authorized by the entity to give such 
views or opinions.

==============================================================================

<Prev in Thread] Current Thread [Next in Thread>
  • Re: HELP!... What processes can/can't run simultaneously andwhat throttles throughput on backup sessions ?, David Longo <=