Networker

Re: [Networker] Cloning performance

2004-10-19 12:01:50
Subject: Re: [Networker] Cloning performance
From: "Krishnan, Ramamurthy" <Ramamurthy.Krishnan AT KPMG.CO DOT UK>
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Date: Tue, 19 Oct 2004 17:00:17 +0100
Far as I know, NetWorker does'nt do parallel clonings.  Say for example,
your backups go to 4 different tapes a night.  NetWorker waits for the
4th tape to finish the backups before it can start cloning the first
tape.  I guess, it becomes essential for NetWorker to finish index
writes for the backups to know what to clone.  I remember someone in
this forum (or in tek-tips) quoting this as "not a good go" with
NetWorker.  I also have read most of the "clone" users use a post backup
script to clone.. I'd vote for a simpler option.. Take weekly full
backup tapes offsite..

- Ram.

-----Original Message-----
From: Legato NetWorker discussion [mailto:NETWORKER AT LISTMAIL.TEMPLE DOT EDU]
On Behalf Of Ulrich Bernhard
Sent: 19 October 2004 16:50
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Subject: Re: [Networker] Cloning performance

Darren,
thanks for your hint.

- We never have made a restore at least not in the time where I have to
be responsible for these systems....

- Yes we are using parallelisme for the backup. The values are 8 for
every client and 24 as server setting.

- I have no idea how the cloning is working. But in the sessions window
I see messages like:
cloning session: 2 of 26 save set(s) reading from ......
zuvrobot:cloning session saving to pool ....

So I think we are cloning one saveset at a time.

Sorry, but I am new to the Legato Networker. Is it possible to clone
also in a parallele mode like the backup?

Thanks

Ulrich

Darren Dunham wrote:
>>We run Legato Networker 6.1.4 on a tru64 5.1 platform with a ESL9198 
>>tape library with 6 dlt8000 tape drives.
>>We make every week one full backup of all connected systems (7 tru64 
>>5.1b server). This backup is defined with 'clones = yes'.
>>The backup needs about 12 hours but the clone needs more than 48
hours.
>>
>>The write speed for the clone sessions is mostly below 1MB per second.
>>
>>What can we do to increase this write performance for the cloning?
>
>
> My guess is that it's the read performance and not the write 
> performance.  Do you test restores periodically?  Do you get good 
> performance on them?
>
> Are you using muliplexing on the the backup?  If so, how are you doing

> the cloning?  If you clone one saveset at a time, you're going to end 
> up demultiplexing the savesets.  This will increase the speed of 
> restores from the clones, but it'll take a lot of time to do.  You 
> might be able to increase the overall time (at a backup penalty) by 
> reducing the multiplexing if you're doing it.
>
> Some places are starting to go with disk staging arenas for items like

> this.
>
> If you're not doing multiplexing, then you'll need to do some 
> investigation and describe your cloning process in more detail.
>
>
> --
> Darren Dunham
ddunham AT taos DOT com
> Senior Technical Consultant         TAOS
http://www.taos.com/
> Got some Dr Pepper?                           San Francisco, CA bay
area
>          < This line left intentionally blank to confuse you. >
>
> --
> Note: To sign off this list, send a "signoff networker" command via 
> email to listserv AT listmail.temple DOT edu or visit the list's Web site at 
> http://listmail.temple.edu/archives/networker.html where you can also 
> view and post messages to the list. Questions regarding this list 
> should be sent to stan AT temple DOT edu 
> =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
  +-----------------------------------------------------------+
  | Ulrich Bernhard                                           |
  | Informatikdienste      Tel: (+41-44) 635 67 73            |
  | Universitaet Zuerich   Fax: (+41-44) 635 45 05            |
  | Winterthurerstr. 190                                      |
  | CH-8057 Zuerich        Email: ulrich.bernhard AT id.unizh DOT ch |
  | Switzerland                                               |
  +-----------------------------------------------------------+

--
Note: To sign off this list, send a "signoff networker" command via
email to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can also
view and post messages to the list. Questions regarding this list should
be sent to stan AT temple DOT edu
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=




                Email Disclaimer

This email has been sent from KPMG LLP, a UK limited
liability partnership, or from one of the companies within
its control (which include KPMG Audit Plc , KPMG United
Kingdom Plc and KPMG UK Limited). The information in
this email is confidential and may be legally privileged.
It is intended solely for the addressee.  Access to this
email by anyone else is unauthorised. If you are not the
intended recipient, any disclosure, copying, distribution
or any action taken or omitted to be taken in reliance on it,
 is prohibited and may be unlawful.  When addressed to
our clients any opinions or advice contained in this email
are subject to the terms and conditions expressed in the
governing KPMG client engagement letter.

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list. Questions regarding this list
should be sent to stan AT temple DOT edu
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

<Prev in Thread] Current Thread [Next in Thread>