Networker

Re: [Networker] What does good LTO3 performance look like?

2008-02-19 08:53:36
Subject: Re: [Networker] What does good LTO3 performance look like?
From: Fazil Saiyed <Fazil.Saiyed AT ANIXTER DOT COM>
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Date: Tue, 19 Feb 2008 07:49:48 -0600
Hello,
If you were to eliminate hardware bottlenecks, i wonder if using 
compression directive might improve backup performance ?
Are you using AFTD atall, if not perhaps a VTL or AFTD may also be worth a 
try.
Using Jumbo frames if your GigE supports would yield the best results.
HTH



Ian G Batten <ian.batten AT UK.FUJITSU DOT COM> 
Sent by: EMC NetWorker discussion <NETWORKER AT LISTSERV.TEMPLE DOT EDU>
02/19/2008 06:06 AM
Please respond to
EMC NetWorker discussion <NETWORKER AT LISTSERV.TEMPLE DOT EDU>; Please respond 
to
Ian G Batten <ian.batten AT UK.FUJITSU DOT COM>


To
NETWORKER AT LISTSERV.TEMPLE DOT EDU
cc

Subject
[Networker] What does good LTO3 performance look like?






I'm backing up four streams of small files from a four-way Fujitsu 
PW450 (equivalent to roughly a Sun V440) with a pair of DotHill 
SanNet II arrays running ZFS.

The storage node is a Sun V240 with an ADIC 100 LTO3 library 
attached, the networker node is another V240 with the indexes held on 
internal disks using ZFS.

The storage node and the networker node are at one end of 30 miles of 
GigE, the storage at the other.  The GigE is essentially dedicated.

Everything's Solaris 10.latest, everything's Networker 7.3.3. The 
obvious FSS niceties have been configured and the backup processes 
are all in their own projects with plenty of shares.  The machines 
are all otherwise idle.

I'm seeing about 200GB/hour (ie 55MB/sec, or ~50% of the network 
bandwidth).

I'm not at all unhappy, and this is fine performance for me.

I'm just curious to know what the limiting factor is and if there's 
anything obvious I could do to raise the performance a bit.  This is 
more about learning than about needs.

nsrmmd on the storage node is the obvious candidate, as it's 
consuming ~40% of a processor and making ~5K system calls per 
second.   Is it possible to raise the overall blocking factor to make 
more efficient use of the CPU?

ian

To sign off this list, send email to listserv AT listserv.temple DOT edu and 
type 
"signoff networker" in the body of the email. Please write to 
networker-request AT listserv.temple DOT edu if you have any problems with this 
list. You can access the archives at 
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER



To sign off this list, send email to listserv AT listserv.temple DOT edu and 
type "signoff networker" in the body of the email. Please write to 
networker-request AT listserv.temple DOT edu if you have any problems with this 
list. You can access the archives at 
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER