This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.
------_=_NextPart_001_01C38D27.3B8BFF9C
Content-Type: text/plain
Steven,
There are many factors that affect this. A good start would be setting
Multiplexing to 3, Number_Data_Buffers at default, and Size_Data_Buffers to
262144. I would start with a benchmark of performance before you change
anything. Also, changing these settings can cause backups to fail, so be
sure to test a backup ALL THE WAY THROUGH before you settle.
Here is a whitepaper that may help:
http://ftp.support.veritas.com/pub/support/products/NetBackup_DataCenter/248
373.pdf
<http://ftp.support.veritas.com/pub/support/products/NetBackup_DataCenter/24
8373.pdf>
Starting at page 18 will help you to truly tune the system.
Steve
-----Original Message-----
From: Green, Steven [mailto:steven.green AT teldta DOT com]
Sent: Tuesday, October 07, 2003 3:25 PM
To: 'veritas-bu AT mailman.eng.auburn DOT edu'
Subject: [Veritas-bu] Windows throughput to LTO2 drives
Hello all...
Does anyone on this list run NBU DataCenter 4.5 on Windows (2K) to LTO-2
drives? If so, what type of throughput are you getting to the drives. Are
you using multiplexing for your backup policies? If so, what MPX level per
drive? What are you using for NBU buffer size and number of buggers? How
about the client communication buffer size? I have unix background so my
familiarity with NBU and Windows is rather light. What I'm looking for are
some ballpark figures as to what I should expect for throughput from Windows
clients (nt/2K/XP servers) to a Windows master/media server with
fibre-attached LTO-2 drives. The clients are all on a gigabit LAN
connection, some with local disk and others with EMC storage. Any
suggestions for tuning a Windows server/client to optimize NBU performance
would be helpful as well.
Thanks in advance...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
steven.green AT teldta DOT com :: TDS Storage Administration
608.664.8210 (desk) :: 608.219.2106 (cell)
It's not that I'm so smart, it's just that I stay with problems longer. A.
Einstein
------_=_NextPart_001_01C38D27.3B8BFF9C
Content-Type: text/html
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii">
<TITLE>Message</TITLE>
<META content=3D"MSHTML 6.00.2800.1226" name=3DGENERATOR></HEAD>
<BODY>
<DIV><SPAN class=3D724170123-07102003><FONT face=3DArial =
color=3D#0000ff=20
size=3D2>Steven,</FONT></SPAN></DIV>
<DIV><SPAN class=3D724170123-07102003><FONT face=3DArial =
color=3D#0000ff size=3D2>There=20
are many factors that affect this. A good start would be setting=20
Multiplexing to 3, Number_Data_Buffers at default, and=20
Size_Data_Buffers to 262144. I would start with a benchmark of =
performance=20
before you change anything. Also, changing these settings can =
cause=20
backups to fail, so be sure to test a backup ALL THE WAY THROUGH before =
you=20
settle.</FONT></SPAN></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2></FONT> </DIV>
<DIV><SPAN class=3D724170123-07102003><FONT face=3DArial =
color=3D#0000ff size=3D2>Here=20
is a whitepaper that may help:</FONT></SPAN></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><A=20
href=3D"http://ftp.support.veritas.com/pub/support/products/NetBackup_Da=
taCenter/248373.pdf">http://ftp.support.veritas.com/pub/support/products=
/NetBackup_DataCenter/248373.pdf</A></FONT></DIV>
<DIV><SPAN style=3D"mso-bidi-font-size: 10.0pt"><?xml:namespace prefix =
=3D o ns =3D=20
"urn:schemas-microsoft-com:office:office" /><o:p><FONT face=3DArial =
color=3D#0000ff=20
size=3D2><SPAN class=3D724170123-07102003>Starting at page 18 will help =
you to truly=20
tune the system. </SPAN></FONT></o:p></SPAN></DIV>
<DIV><SPAN style=3D"mso-bidi-font-size: 10.0pt"><o:p><FONT face=3DArial =
color=3D#0000ff size=3D2><SPAN=20
class=3D724170123-07102003></SPAN></FONT></o:p></SPAN> </DIV>
<DIV><SPAN style=3D"mso-bidi-font-size: 10.0pt"><o:p><FONT face=3DArial =
color=3D#0000ff size=3D2><SPAN=20
class=3D724170123-07102003>Steve</SPAN></FONT></o:p></SPAN></DIV>
<BLOCKQUOTE dir=3Dltr style=3D"MARGIN-RIGHT: 0px">
<DIV></DIV>
<DIV class=3DOutlookMessageHeader lang=3Den-us dir=3Dltr =
align=3Dleft><FONT=20
face=3DTahoma size=3D2>-----Original Message-----<BR><B>From:</B> =
Green, Steven=20
[mailto:steven.green AT teldta DOT com] <BR><B>Sent:</B> Tuesday, October =
07, 2003=20
3:25 PM<BR><B>To:</B> =
'veritas-bu AT mailman.eng.auburn DOT edu'<BR><B>Subject:</B>=20
[Veritas-bu] Windows throughput to LTO2 drives<BR><BR></FONT></DIV>
<P><FONT face=3DArial size=3D2>Hello all...</FONT> </P>
<P><FONT face=3DArial size=3D2>Does anyone on this list run NBU =
DataCenter 4.5 on=20
Windows (2K) to LTO-2 drives? If so, what type of throughput are you =
getting=20
to the drives. Are you using multiplexing for your backup policies? =
If so,=20
what MPX level per drive? What are you using for NBU buffer size and =
number of=20
buggers? How about the client communication buffer size? I have unix=20
background so my familiarity with NBU and Windows is rather light. =
What I'm=20
looking for are some ballpark figures as to what I should expect for=20
throughput from Windows clients (nt/2K/XP servers) to a Windows =
master/media=20
server with fibre-attached LTO-2 drives. The clients are all on a =
gigabit LAN=20
connection, some with local disk and others with EMC storage. Any =
suggestions=20
for tuning a Windows server/client to optimize NBU performance would =
be=20
helpful as well.</FONT></P>
<P><FONT face=3DArial size=3D2>Thanks in advance...</FONT> </P>
<P><FONT face=3DArial=20
=
size=3D1>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~</=
FONT>=20
<BR><FONT face=3DArial size=3D1>steven.green AT teldta DOT com :: TDS =
Storage=20
Administration</FONT> <BR><FONT face=3DArial size=3D1>608.664.8210 =
(desk) ::=20
608.219.2106 (cell)</FONT> <BR><FONT face=3DArial size=3D1>It's not =
that I'm so=20
smart, it's just that I stay with problems longer. A. =
Einstein</FONT>=20
</P></BLOCKQUOTE></BODY></HTML>
------_=_NextPart_001_01C38D27.3B8BFF9C--
|
