Veritas-bu

[Veritas-bu] Backup dilemma

2003-08-19 08:39:57
Subject: [Veritas-bu] Backup dilemma
From: Cecil.Hypolite AT frx DOT com (Cecil.Hypolite AT frx DOT com)
Date: Tue, 19 Aug 2003 08:39:57 -0400
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_01C3664E.FF7B2A90
Content-Type: text/plain

Hey Wayne,
Thanks for you response
Cecil

-----Original Message-----
From: Wayne Smith [mailto:WTSmith AT Maine DOT edu] 
Sent: Saturday, August 16, 2003 11:42 PM
To: veritas-bu AT mailman.eng.auburn DOT edu
Subject: Re: [Veritas-bu] Backup dilemma

Cecil Hypolite wrote, in part:

>I have to do a weekly full backup of a system that has 17 partitions and
>about 500 GB of data. The backup will be done on the weekend, when there
are
>few other jobs running. Here are some specs:
>network: 100 MB/sec (12.5 MB/sec)
>SCSI-2 connection : (40 MB/sec)80 MB/sec (two drives)
>Drives: 2 SDLT220's 79.2 GB/hr
>Produce: Netbackup Datacenter 4.5
>
>It seems that the network is a bottle neck as it can only transfer 45 Gb/hr
>(3600 *(12.5 MB/sec) = 45 GB/hr)
>With that network transfer speed of 45 GB/hr I should be able to back this
>500 GB of data within a 12 hour window (11.2 hours to be exact), are my
>calculations correct?
>

The calculations may be correct, but I wouldn't think it wise to plan on 
that throughput.  My best 100mbps clients only get 20-25 MB/hr ... most 
much less.

>-Also since the network transfer speed is only 45 Gb/hr I should not
>implement multiplexing, as one of the drives will start "shoe shining"
>because the network cannot sustain a transfer rate of 79.2 GB.hr. Is this
>correct?
>

I don't understand the reasoning. IMHO, this may be a case where you can 
stand some multiplexing. Already, this client can't give a tape drive 
fast enough.  If you have other data coming in (presumably via another 
network),  the sum of the data would do better at keeping a drive busy.  
Beyond that, if you are restoring, the client's data will still come off 
the drive fast enough to keep the network busy (assuming enough buffers 
and an appropriate level of multiplexing.

Home this helps, wayne

_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu

------_=_NextPart_001_01C3664E.FF7B2A90
Content-Type: text/html
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3DUS-ASCII">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2653.12">
<TITLE>RE: [Veritas-bu] Backup dilemma</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2>Hey Wayne,</FONT>
<BR><FONT SIZE=3D2>Thanks for you response</FONT>
<BR><FONT SIZE=3D2>Cecil</FONT>
</P>

<P><FONT SIZE=3D2>-----Original Message-----</FONT>
<BR><FONT SIZE=3D2>From: Wayne Smith [<A =
HREF=3D"mailto:WTSmith AT Maine DOT edu">mailto:WTSmith AT Maine DOT edu</A>] 
</FONT>
<BR><FONT SIZE=3D2>Sent: Saturday, August 16, 2003 11:42 PM</FONT>
<BR><FONT SIZE=3D2>To: veritas-bu AT mailman.eng.auburn DOT edu</FONT>
<BR><FONT SIZE=3D2>Subject: Re: [Veritas-bu] Backup dilemma</FONT>
</P>

<P><FONT SIZE=3D2>Cecil Hypolite wrote, in part:</FONT>
</P>

<P><FONT SIZE=3D2>&gt;I have to do a weekly full backup of a system =
that has 17 partitions and</FONT>
<BR><FONT SIZE=3D2>&gt;about 500 GB of data. The backup will be done on =
the weekend, when there are</FONT>
<BR><FONT SIZE=3D2>&gt;few other jobs running. Here are some =
specs:</FONT>
<BR><FONT SIZE=3D2>&gt;network: 100 MB/sec (12.5 MB/sec)</FONT>
<BR><FONT SIZE=3D2>&gt;SCSI-2 connection : (40 MB/sec)80 MB/sec (two =
drives)</FONT>
<BR><FONT SIZE=3D2>&gt;Drives: 2 SDLT220's 79.2 GB/hr</FONT>
<BR><FONT SIZE=3D2>&gt;Produce: Netbackup Datacenter 4.5</FONT>
<BR><FONT SIZE=3D2>&gt;</FONT>
<BR><FONT SIZE=3D2>&gt;It seems that the network is a bottle neck as it =
can only transfer 45 Gb/hr</FONT>
<BR><FONT SIZE=3D2>&gt;(3600 *(12.5 MB/sec) =3D 45 GB/hr)</FONT>
<BR><FONT SIZE=3D2>&gt;With that network transfer speed of 45 GB/hr I =
should be able to back this</FONT>
<BR><FONT SIZE=3D2>&gt;500 GB of data within a 12 hour window (11.2 =
hours to be exact), are my</FONT>
<BR><FONT SIZE=3D2>&gt;calculations correct?</FONT>
<BR><FONT SIZE=3D2>&gt;</FONT>
</P>

<P><FONT SIZE=3D2>The calculations may be correct, but I wouldn't think =
it wise to plan on </FONT>
<BR><FONT SIZE=3D2>that throughput.&nbsp; My best 100mbps clients only =
get 20-25 MB/hr ... most </FONT>
<BR><FONT SIZE=3D2>much less.</FONT>
</P>

<P><FONT SIZE=3D2>&gt;-Also since the network transfer speed is only 45 =
Gb/hr I should not</FONT>
<BR><FONT SIZE=3D2>&gt;implement multiplexing, as one of the drives =
will start &quot;shoe shining&quot;</FONT>
<BR><FONT SIZE=3D2>&gt;because the network cannot sustain a transfer =
rate of 79.2 GB.hr. Is this</FONT>
<BR><FONT SIZE=3D2>&gt;correct?</FONT>
<BR><FONT SIZE=3D2>&gt;</FONT>
</P>

<P><FONT SIZE=3D2>I don't understand the reasoning. IMHO, this may be a =
case where you can </FONT>
<BR><FONT SIZE=3D2>stand some multiplexing. Already, this client can't =
give a tape drive </FONT>
<BR><FONT SIZE=3D2>fast enough.&nbsp; If you have other data coming in =
(presumably via another </FONT>
<BR><FONT SIZE=3D2>network),&nbsp; the sum of the data would do better =
at keeping a drive busy.&nbsp; </FONT>
<BR><FONT SIZE=3D2>Beyond that, if you are restoring, the client's data =
will still come off </FONT>
<BR><FONT SIZE=3D2>the drive fast enough to keep the network busy =
(assuming enough buffers </FONT>
<BR><FONT SIZE=3D2>and an appropriate level of multiplexing.</FONT>
</P>

<P><FONT SIZE=3D2>Home this helps, wayne</FONT>
</P>

<P><FONT =
SIZE=3D2>_______________________________________________</FONT>
<BR><FONT SIZE=3D2>Veritas-bu maillist&nbsp; -&nbsp; =
Veritas-bu AT mailman.eng.auburn DOT edu</FONT>
<BR><FONT SIZE=3D2><A =
HREF=3D"http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu"; =
TARGET=3D"_blank">http://mailman.eng.auburn.edu/mailman/listinfo/veritas=
-bu</A></FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01C3664E.FF7B2A90--

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