If it's within your required parameters, then it's OK.
It's obviously and trivially true that restoring from mulitplexed
tapes is slower, because the drive has to read data that's not
required to get to the data that is. If this is still fast enough,
then there's no problem.
We multiplex tapes up to the maximum, and typically end up with
restores spread over several tapes. Currently the restore speed has
been within our acceptable limits, however we have only so far been
requested to restore to one machine at a time. If there comes a time
when we have to restore multiple clients 'simultaneously' we may be
in for an unacceptable wait. As far as I know, Netbackup (5.1) only
demuliplexes one stream at a time, so if you have restores to several
machines from the same 'archive', it will read the tape multiple
times.
It is not always practical to use disc staging either. Our full
backup will be about 25TB. That much disc would cost an unacceptably
large amount money. Of course, incrementals are much more amenable to
disc staging because they are smaller.
just my (obvious) 2 penneth worth.
At 8:26 am +0100 31/7/06, WEAVER, Simon wrote:
>Stuart
>I will believe what "I" have seen, and the restores for the 2 SAN Media
>Servers are absolutely well within limits of restoring data. We run an
>extremely fast backbone, so I have no problems with the configuration in
>place.
>
>
>Regards
>
>Simon Weaver
>3rd Line Technical Support
>Windows Domain Administrator
>
>EADS Astrium Limited, B23AA IM (DCS)
>Anchorage Road, Portsmouth, PO3 5PU
>
>Email: Simon.Weaver at Astrium-eads.net
>
>
>
>-----Original Message-----
>From: Liddle, Stuart [mailto:liddles at amgen.com]
>Sent: 31 July 2006 07:59
>To: WEAVER, Simon; 'Mansell, Richard'; veritas-bu at mailman.eng.auburn.edu
>Subject: Re: [Veritas-bu] Multiplexing
>
>
>Simon,
>
>Fine....believe what you want....but the numbers don't lie.
>
>If you ever try any testing of this you will see that restores are
>significantly faster when you do them from non-multiplexed tapes.
>
>--stuart
>
>-----Original Message-----
>From: WEAVER, Simon [mailto:simon.weaver at astrium.eads.net]
>Sent: Sunday, July 30, 2006 10:45 PM
>To: 'Liddle, Stuart'; 'Mansell, Richard'; veritas-bu at mailman.eng.auburn.edu
>Subject: RE: [Veritas-bu] Multiplexing
>
>
>Stuart
>Completely disagree - we get the best of both worlds and restores have never
>been any real issue for us, even large amounts of data across multiple
>tapes.
>
>Regards
>
>Simon Weaver
>3rd Line Technical Support
>Windows Domain Administrator
>
>EADS Astrium Limited, B23AA IM (DCS)
>Anchorage Road, Portsmouth, PO3 5PU
>
>Email: Simon.Weaver at Astrium-eads.net
>
>
>
>-----Original Message-----
>From: Liddle, Stuart [mailto:liddles at amgen.com]
>Sent: 28 July 2006 19:37
>To: WEAVER, Simon; 'Mansell, Richard'; veritas-bu at mailman.eng.auburn.edu
>Subject: RE: [Veritas-bu] Multiplexing
>
>
>That's good for backup speeds, but it sucks for restores. Unless all of
>your clients have really slow network connections, having a high
>multiplexing value is not really a good idea.
>
>You might want to consider dropping the multiplexing down to something like
>6.
>
>Our configuration uses VTL's to our first backup copy and then I use vault
>to put it to physical tape. I multiplex to the VTL and single-stream to the
>physical tapes. Once it is on the physical tape, we have a script that will
>bpexpdate the copy 1 image on the VTL (this is not an automatic option in
>NetBackup until version 6.x).
>
>When you do a restore from a physical tape, you have a single-stream copy
>and not one that is multiplexed....this makes for a faster restore.
>
>--stuart
>
>-----Original Message-----
>From: veritas-bu-bounces at mailman.eng.auburn.edu
>[mailto:veritas-bu-bounces at mailman.eng.auburn.edu] On Behalf Of WEAVER,
>Simon
>Sent: Thursday, July 27, 2006 11:22 PM
>To: 'Mansell, Richard'; veritas-bu at mailman.eng.auburn.edu
>Subject: Re: [Veritas-bu] Multiplexing
>
>
>Richard
>Yes I use multiplexing a lot
>
>NBU 5.1 Master MP2 + 2 SAN Media Servers
>
>4 drives in a robot - using SSO
>
>1 Storage Unit set to multiplex 25 jobs per drive (Master)
>2 SAN Storage Units set to use 2 drives, multiplex 8 per drive
>
>Seems to work - get a lot of jobs done - around 300 per night if that helps.
>
>Regards
>
>Simon Weaver
>3rd Line Technical Support
>Windows Domain Administrator
>
>EADS Astrium Limited, B23AA IM (DCS)
>Anchorage Road, Portsmouth, PO3 5PU
>
>Email: Simon.Weaver at Astrium-eads.net
>
>
>
>-----Original Message-----
>From: Mansell, Richard [mailto:Richard.Mansell at ccc.govt.nz]
>Sent: 27 July 2006 22:31
>To: veritas-bu at mailman.eng.auburn.edu
>Subject: [Veritas-bu] Multiplexing
>
>
>
>We are struggling to get all of the backups run during the backup window and
>one of the obvious solutions is to start using multiplexing as the tape
>drives are not being driven at any where near their maximum throughput.
>There has been a rather nasty multiplexing problem fixed in 6.0 MP3 but I
>would be interested to do a straw poll of how many people are using
>multiplexing and if possible any issues encountered.
>
>I would appreciate it if you would respond (off list) as to whether you use
>multiplexing or not (any version of NetBackup). I will summarise the results
>and post back to the list.
>
>Regards
>
>Richard
>
>
>**********************************************************************
>This electronic email and any files transmitted with it are intended
>solely for the use of the individual or entity to whom they are addressed.
>
>The views expressed in this message are those of the individual
>sender and may not necessarily reflect the views of the Christchurch City
>Council.
>
>If you are not the correct recipient of this email please advise the sender
>and delete.
>
>Christchurch City Council
>http://www.ccc.govt.nz
>**********************************************************************
>
>
>_______________________________________________
>Veritas-bu maillist - Veritas-bu at mailman.eng.auburn.edu
>http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
>
>This email is for the intended addressee only.
>If you have received it in error then you must not use, retain, disseminate
>or otherwise deal with it. Please notify the sender by return email. The
>views of the author may not necessarily constitute the views of Astrium
>Limited. Nothing in this email shall bind Astrium Limited in any contract or
>obligation.
>
>Astrium Limited, Registered in England and Wales No. 2449259 Registered
>Office: Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2AS, England
>_______________________________________________
>Veritas-bu maillist - Veritas-bu at mailman.eng.auburn.edu
>http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
>
>This email is for the intended addressee only.
>If you have received it in error then you must not use, retain, disseminate
>or otherwise deal with it. Please notify the sender by return email. The
>views of the author may not necessarily constitute the views of Astrium
>Limited. Nothing in this email shall bind Astrium Limited in any contract or
>obligation.
>
>Astrium Limited, Registered in England and Wales No. 2449259 Registered
>Office: Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2AS, England
>_______________________________________________
>Veritas-bu maillist - Veritas-bu at mailman.eng.auburn.edu
>http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
>
>This email is for the intended addressee only.
>If you have received it in error then you must not use, retain,
>disseminate or otherwise deal with it.
>Please notify the sender by return email.
>The views of the author may not necessarily constitute the views of
>Astrium Limited.
>Nothing in this email shall bind Astrium Limited in any contract or
>obligation.
>
>Astrium Limited, Registered in England and Wales No. 2449259
>Registered Office: Gunnels Wood Road, Stevenage, Hertfordshire, SG1
>2AS, England
--
---
REMEMBER - the safety of your data is your responsibility
If it's important to you - ** MAKE A COPY ** or 2 or 3 or 4 or ........
There is only one certainty about storage systems - they *will* fail !
Chris Freemantle
Data Manager
Wellcome Department of Imaging Neuroscience
Tel. +44 (0)20 7833 7472 reception
Tel. +44 (0)20 7833 7496 direct
Fax +44 (0)20 7813 1420
Home Page: http://www.fil.ion.ucl.ac.uk
|
