Veritas-bu

[Veritas-bu] backing up LARGE oracle databases (looking for s ugge stions)

2005-01-31 13:26:40
Subject: [Veritas-bu] backing up LARGE oracle databases (looking for s ugge stions)
From: SJACOBSO AT novell DOT com (Scott Jacobson)
Date: Mon, 31 Jan 2005 11:26:40 -0700
This is a MIME message. If you are reading this text, you may want to 
consider changing to a mail reader or gateway that understands how to 
properly handle MIME multipart messages.

--=__PartDFFF6D70.2__=
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

For additional validation, you may want to consider looking at the BPTM
logs and see if there any "Waiting for [empty/full] buffer" entries. 
You would then be able to determine from which side (network/drives) the
latency, if any, is occurring.
 
Scott

>>> "Dobbertien, Matthew" <MDobbertien AT tribune DOT com> 1/31/2005 10:38 AM
>>>

We have checked bus bandwidth.  We believe this was a problem on the
old
system, but now the bus is 64 Bit/66 MhZ PCI, which should provide 512
MB/s of bandwidth--more than enough for both disk and network.  Doing
network only tests (no disk, no real data), we were able to achieve
over
100 MB/s on a link directly between two hosts.  Adding in our Cisco
switches reduced bandwidth to about 70 MB/s max (this is still
theoretical, of course).  We have seen some information on spanning
tree
reducing the overall speed of the network, but our management of the
switches is limited.  

Matthew Dobbertien
Sr. Systems Analyst
Chicago Tribune
mdobbertien AT tribune DOT com

-----Original Message-----
From: Charles Ballowe [mailto:cballowe AT gmail DOT com] 
Sent: Monday, January 31, 2005 11:29 AM
To: Dobbertien, Matthew
Cc: blaine_robison AT netzero DOT com; Dwayne.Brzozowski AT mail.va DOT gov;
veritas-bu AT mailman.eng.auburn DOT edu
Subject: Re: [Veritas-bu] backing up LARGE oracle databases (looking
for
s ugge stions)

Your bottleneck could also be bus bandwidth. If your drives and
network card are both on the same 33 Mhz 32 bit pci bus, the max
bandwidth is around 132000000 bytes/sec - overhead. That would be 60
read from disk + 60 write to network. Since I don't know enough about
your server config, I can't say whether that's the case, but it could
be. You could try piping /dev/zero into a network stream just to prove
the point.

-Charlie


On Mon, 31 Jan 2005 09:46:30 -0600, Dobbertien, Matthew
<MDobbertien AT tribune DOT com> wrote:
> We have run multiple tests on the system.  Writing the backup to
> /dev/null results in over 100 MB/s throughput, but running the same
jobs
> over the network maxes out at 60 MB/s regardless of number of
streams
or
> tape drives or disk on the back end (we generally write to virtual
tape,
> which has a throughput of over 150 MB/s).  We do see a maximum
> throughput of 40 MB/s per stream--I believe this is just the
overhead
> from bpbkar.  The disks we are reading from are capable of pushing
data
> much faster than that (backend disks are on a clarion cx600).  Even
> running three backup streams from the same disk results in over 100
MB/s
> throughput.
> 
> Matthew Dobbertien
> Sr. Systems Analyst
> Chicago Tribune
> mdobbertien AT tribune DOT com
> 312-222-2203 (Work)
> 312-656-3019 (Cell)
> 
> -----Original Message-----
> From: blaine_robison AT netzero DOT com [mailto:blaine_robison AT netzero DOT 
> com]
> Sent: Monday, January 31, 2005 9:09 AM
> To: Dwayne.Brzozowski AT mail.va DOT gov
> Cc: Dobbertien, Matthew; veritas-bu AT mailman.eng.auburn DOT edu
> Subject: RE: [Veritas-bu] backing up LARGE oracle databases (looking
for
> s ugge stions)
> 
> Hey,
> thanks Dwayne,
> 
> Matt how do you know the GigE connection is the choke point?
> 
> Blaine Robison
> Solaris Certified System Administrator
> Solaris Certified Network Administrator
> Sun Certified Backup and Recovery Engineer
> Veritas Certified Professional Data Recovery
> 
> _______________________________________________
> Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
>


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


--=__PartDFFF6D70.2__=
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; charset=3Diso-8859-1"=
>
<META content=3D"MSHTML 6.00.2800.1479" name=3DGENERATOR></HEAD>
<BODY style=3D"MARGIN: 4px 4px 1px; FONT: 10pt Tahoma">
<DIV>For additional validation, you may want to consider looking at the =
BPTM logs and see&nbsp;if there any "Waiting for [empty/full] buffer" =
entries.&nbsp; You would then be able to determine from which side =
(network/drives) the latency, if any, is occurring.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Scott<BR><BR>&gt;&gt;&gt; "Dobbertien, Matthew" &lt;MDobbertien@tribun=
e.com&gt; 1/31/2005 10:38 AM &gt;&gt;&gt;<BR></DIV>
<DIV style=3D"COLOR: #000000">We have checked bus bandwidth.&nbsp; We =
believe this was a problem on the old<BR>system, but now the bus is 64 =
Bit/66 MhZ PCI, which should provide 512<BR>MB/s of bandwidth--more than =
enough for both disk and network.&nbsp; Doing<BR>network only tests (no =
disk, no real data), we were able to achieve over<BR>100 MB/s on a link =
directly between two hosts.&nbsp; Adding in our Cisco<BR>switches reduced =
bandwidth to about 70 MB/s max (this is still<BR>theoretical, of course).&n=
bsp; We have seen some information on spanning tree<BR>reducing the =
overall speed of the network, but our management of the<BR>switches is =
limited.&nbsp; <BR><BR>Matthew Dobbertien<BR>Sr. Systems Analyst<BR>Chicago=
 Tribune<BR>mdobbertien AT tribune DOT com<BR><BR>-----Original 
Message-----<BR>Fr=
om: Charles Ballowe [<A href=3D"mailto:cballowe AT gmail DOT 
com]">mailto:cballowe=
@gmail.com]</A> <BR>Sent: Monday, January 31, 2005 11:29 AM<BR>To: =
Dobbertien, Matthew<BR>Cc: blaine_robison AT netzero DOT com; 
Dwayne.Brzozowski@ma=
il.va.gov;<BR>veritas-bu AT mailman.eng.auburn DOT edu<BR>Subject: Re: 
[Veritas-bu=
] backing up LARGE oracle databases (looking for<BR>s ugge stions)<BR><BR>Y=
our bottleneck could also be bus bandwidth. If your drives and<BR>network =
card are both on the same 33 Mhz 32 bit pci bus, the max<BR>bandwidth is =
around 132000000 bytes/sec - overhead. That would be 60<BR>read from disk =
+ 60 write to network. Since I don't know enough about<BR>your server =
config, I can't say whether that's the case, but it could<BR>be. You could =
try piping /dev/zero into a network stream just to prove<BR>the point.<BR><=
BR>-Charlie<BR><BR><BR>On Mon, 31 Jan 2005 09:46:30 -0600, Dobbertien, =
Matthew<BR>&lt;MDobbertien AT tribune DOT com&gt; wrote:<BR>&gt; We have run =
multiple tests on the system.&nbsp; Writing the backup to<BR>&gt; =
/dev/null results in over 100 MB/s throughput, but running the same<BR>jobs=
<BR>&gt; over the network maxes out at 60 MB/s regardless of number of =
streams<BR>or<BR>&gt; tape drives or disk on the back end (we generally =
write to virtual<BR>tape,<BR>&gt; which has a throughput of over 150 =
MB/s).&nbsp; We do see a maximum<BR>&gt; throughput of 40 MB/s per =
stream--I believe this is just the overhead<BR>&gt; from bpbkar.&nbsp; The =
disks we are reading from are capable of pushing<BR>data<BR>&gt; much =
faster than that (backend disks are on a clarion cx600).&nbsp; Even<BR>&gt;=
 running three backup streams from the same disk results in over 100<BR>MB/=
s<BR>&gt; throughput.<BR>&gt; <BR>&gt; Matthew Dobbertien<BR>&gt; Sr. =
Systems Analyst<BR>&gt; Chicago Tribune<BR>&gt; mdobbertien AT tribune DOT 
com<BR>=
&gt; 312-222-2203 (Work)<BR>&gt; 312-656-3019 (Cell)<BR>&gt; <BR>&gt; =
-----Original Message-----<BR>&gt; From: blaine_robison AT netzero DOT com [<A =
href=3D"mailto:blaine_robison AT netzero DOT 
com]">mailto:[email protected]=
om]</A><BR>&gt; Sent: Monday, January 31, 2005 9:09 AM<BR>&gt; To: =
Dwayne.Brzozowski AT mail.va DOT gov<BR>&gt; Cc: Dobbertien, Matthew; 
veritas-bu@m=
ailman.eng.auburn.edu<BR>&gt; Subject: RE: [Veritas-bu] backing up LARGE =
oracle databases (looking<BR>for<BR>&gt; s ugge stions)<BR>&gt; <BR>&gt; =
Hey,<BR>&gt; thanks Dwayne,<BR>&gt; <BR>&gt; Matt how do you know the GigE =
connection is the choke point?<BR>&gt; <BR>&gt; Blaine Robison<BR>&gt; =
Solaris Certified System Administrator<BR>&gt; Solaris Certified Network =
Administrator<BR>&gt; Sun Certified Backup and Recovery Engineer<BR>&gt; =
Veritas Certified Professional Data Recovery<BR>&gt; <BR>&gt; _____________=
__________________________________<BR>&gt; Veritas-bu maillist&nbsp; =
-&nbsp; Veritas-bu AT mailman.eng.auburn DOT edu<BR>&gt; <A 
href=3D"http://mailman=
.eng.auburn.edu/mailman/listinfo/veritas-bu">http://mailman.eng.auburn.edu/=
mailman/listinfo/veritas-bu</A><BR>&gt;<BR><BR><BR>________________________=
_______________________<BR>Veritas-bu maillist&nbsp; -&nbsp; Veritas-bu@mai=
lman.eng.auburn.edu<BR><A href=3D"http://mailman.eng.auburn.edu/mailman/lis=
tinfo/veritas-bu">http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu=
</A><BR></DIV></BODY></HTML>

--=__PartDFFF6D70.2__=--

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