X-MimeOLE: Produced By Microsoft Exchange V6.5
Received: by onstor-exch02.onstor.net 
	id <01C862EC.E6CB34CF@onstor-exch02.onstor.net>; Tue, 29 Jan 2008 20:05:00 -0700
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C862EC.E6CB34CF"
Content-class: urn:content-classes:message
Subject: RE: Cleary Gottlieb - NDMP Performance & Data Mirroring
Date: Tue, 29 Jan 2008 20:04:59 -0700
Message-ID: <BB375AF679D4A34E9CA8DFA650E2B04E07FB5B35@onstor-exch02.onstor.net>
In-Reply-To: <BB375AF679D4A34E9CA8DFA650E2B04E07FB56EC@onstor-exch02.onstor.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: Cleary Gottlieb - NDMP Performance & Data Mirroring
Thread-Index: AchiqNFwTwK7aipRTm6ds9uLBSlKWwAA+pJQAA9NOIA=
From: "Jobi Ariyamannil" <jobi.ariyamannil@onstor.com>
To: "Tim Gardner" <tim.gardner@onstor.com>,
	"dl-Engineering" <dl-engineering@onstor.com>,
	"dl-Customer-Engineering" <dl-Customer-Engineering@onstor.com>,
	"dl-se" <dl-se@onstor.com>,
	"Bob Miller" <bob.miller@onstor.com>

This is a multi-part message in MIME format.

------_=_NextPart_001_01C862EC.E6CB34CF
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Good news, indeed!

R3.2 made a lot of significant performance improvements.

=20

Before this becoming a standard to use Raid 1/0 for the initial region
of the filesystems for performance reasons, I would like to point out
some risks involved.

These customers use Raid-5 for most of the filesystem storage to recover
from disk failures.

Now we are asking to use lower RAID protection for filesystem log and
some of the critical metadata for performance reasons.

=20

1.	The filesystem uses the first 64MB for its journal.  If we use
Raid-0 for that, we may not recover from disk failures.  And a failure
during log replay means immediate full filesystem consistency check (eek
repair) and associated downtime.  Customers who prefer to use this
configuration must be warned about this.
2.	Our filesystem places some of the critical metadata (block free
bitmaps, inodes, snapshots etc) in the initial portion of the filesystem
immediately after the filesystem log.  So if this portion is not RAID
protected, we may not recover from disk failures.  Also note that
usually these portions of the filesystem is heavily read and written.
It is preferred to use the lun just for the size of the filesystem log
for the first Raid-1/0 lun instead of everybody using random sized LUNs
for this.

=20

Regards,

Jobi

=20

________________________________

From: Tim Gardner=20
Sent: Tuesday, January 29, 2008 11:37 AM
To: dl-Engineering; dl-Customer-Engineering; dl-se; Bob Miller
Subject: FW: Cleary Gottlieb - NDMP Performance & Data Mirroring

=20

Folks,

=20

It is not often that we get an email like this from a customer telling
us how satisfied they are with our product.

Usually we only hear from customers when the product is not working so I
wanted to share this with you.

=20

This also very nicely validates the information that we discussed in the
last tech talk about storage configuration.

=20

Tim

=20

=20

________________________________

From: David Flewitt [mailto:DFlewitt@cgsh.com]=20
Sent: Tuesday, January 29, 2008 10:57 AM
To: Tim Gardner
Cc: Bill Duffy; Robert Campana; Scott Moyer; DC-Network@cgsh.com;
William Tedrow; John R Daggan; Daniel Stotz
Subject: Cleary Gottlieb - NDMP Performance & Data Mirroring

=20


Good afternoon Tim.=20

I just wanted to give you some preliminary feedback on our resent tests.


NDMP=20

Bill Duffy upgraded all of our Bobcats to V3.2.  Following your
recommendation we re-structured our RAID groups on our storage array as
follows.=20

1st LUN of each volume                40GB                RAID 1/0 (a
total of ten 146GB Fibre Channel disks)=20
2nd LUN of each volume                100GB                RAID 5 (a
total of five 500GB ATA disks)=20
3rd LUN of each volume                100GB                RAID 5 (a
total of five 500GB ATA disks)=20
4th LUN of each volume                100GB                RAID 5 (a
total of five 500GB ATA disks)=20
5th LUN of each volume                100GB                RAID 5 (a
total of five 500GB ATA disks)=20

I created a volume using the above structure and copied over the same
sample data we used previously namely, 270GB and approx 5 million files.
When we ran a full NDMP backup job on this data previously, it took
approx 14 hours to complete (19GB/hour).=20

I am pleased (and thankful) to report that the same full backup of the
same data set using the above structure completed in approx 2.5 hours
(108GB/hour)!!!!!!!!!!!!!  That's a five fold improvement in
performance.  We consider that "spectacular"!=20

So although I was a little sceptical about your RAID group approach
initially, I am now convinced it is best practice.=20

Data Mirroring=20

Since the OS upgrade to Ver 3.2 we have initiated and successfully
completed two baseline data mirrors.  One mirror was 270GB (5 million
files) and the other was 470GB (7.5 million files).  Although they
predictable took several days to finish, they did finish!  No aborting
or Bobcat re-booting.=20

We are in the process of upgrading our Riverbed IP optimizer to a model
that will handle a 45Mbps WAN link.  The above test was run with the
existing 10Mbps link.=20

Thanks again for your assistance.=20

Regards,

Dave Flewitt
____________________
Network Analyst
Cleary Gottlieb Steen & Hamilton LLP
2000 Pennsylvania Avenue N.W.
Washington, DC 20006
Tel: 202-974-1726
dflewitt@cgsh.com
http://www.clearygottlieb.com=20

This message is being sent from a law firm and may contain
confidential or privileged information.  If you are not
the intended recipient, please advise the sender
immediately by reply e-mail and delete this message and
any attachments without retaining a copy.

------_=_NextPart_001_01C862EC.E6CB34CF
Content-Type: text/html;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" =
xmlns:o=3D"urn:schemas-microsoft-com:office:office" =
xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:st1=3D"urn:schemas-microsoft-com:office:smarttags" =
xmlns=3D"http://www.w3.org/TR/REC-html40">

<head>
<meta http-equiv=3DContent-Type content=3D"text/html; =
charset=3Dus-ascii">
<meta name=3DGenerator content=3D"Microsoft Word 11 (filtered medium)">
<!--[if !mso]>
<style>
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style>
<![endif]--><o:SmartTagType
 namespaceuri=3D"urn:schemas-microsoft-com:office:smarttags" =
name=3D"PostalCode"/>
<o:SmartTagType =
namespaceuri=3D"urn:schemas-microsoft-com:office:smarttags"
 name=3D"State"/>
<o:SmartTagType =
namespaceuri=3D"urn:schemas-microsoft-com:office:smarttags"
 name=3D"Street"/>
<o:SmartTagType =
namespaceuri=3D"urn:schemas-microsoft-com:office:smarttags"
 name=3D"City"/>
<o:SmartTagType =
namespaceuri=3D"urn:schemas-microsoft-com:office:smarttags"
 name=3D"address"/>
<o:SmartTagType =
namespaceuri=3D"urn:schemas-microsoft-com:office:smarttags"
 name=3D"place"/>
<o:SmartTagType =
namespaceuri=3D"urn:schemas-microsoft-com:office:smarttags"
 name=3D"PersonName"/>
<!--[if !mso]>
<style>
st1\:*{behavior:url(#default#ieooui) }
</style>
<![endif]-->
<style>
<!--
 /* Font Definitions */
 @font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
 /* Style Definitions */
 p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman";}
a:link, span.MsoHyperlink
	{color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{color:purple;
	text-decoration:underline;}
p
	{mso-margin-top-alt:auto;
	margin-right:0in;
	mso-margin-bottom-alt:auto;
	margin-left:0in;
	font-size:12.0pt;
	font-family:"Times New Roman";}
pre
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:10.0pt;
	font-family:"Courier New";}
span.EmailStyle19
	{mso-style-type:personal;
	font-family:Arial;
	color:navy;}
span.EmailStyle20
	{mso-style-type:personal-reply;
	font-family:Arial;
	color:navy;}
@page Section1
	{size:8.5in 11.0in;
	margin:1.0in 1.25in 1.0in 1.25in;}
div.Section1
	{page:Section1;}
 /* List Definitions */
 @list l0
	{mso-list-id:1258906188;
	mso-list-type:hybrid;
	mso-list-template-ids:-2135004888 67698703 67698713 67698715 67698703 =
67698713 67698715 67698703 67698713 67698715;}
@list l0:level1
	{mso-level-tab-stop:.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
ol
	{margin-bottom:0in;}
ul
	{margin-bottom:0in;}
-->
</style>
<!--[if gte mso 9]><xml>
 <o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
 <o:shapelayout v:ext=3D"edit">
  <o:idmap v:ext=3D"edit" data=3D"1" />
 </o:shapelayout></xml><![endif]-->
</head>

<body lang=3DEN-US link=3Dblue vlink=3Dpurple>

<div class=3DSection1>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>Good news, =
indeed!<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>R3.2 made a lot of significant =
performance
improvements.<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>Before this becoming a standard to =
use
Raid 1/0 for the initial region of the filesystems for performance =
reasons, I
would like to point out some risks =
involved.<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>These customers use Raid-5 for most =
of the
filesystem storage to recover from disk =
failures.<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>Now we are asking to use lower RAID
protection for filesystem log and some of the critical metadata for =
performance
reasons.<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'><o:p>&nbsp;</o:p></span></font></p>

<ol style=3D'margin-top:0in' start=3D1 type=3D1>
 <li class=3DMsoNormal style=3D'color:navy;mso-list:l0 level1 =
lfo1'><font size=3D2
     color=3Dnavy face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>The
     filesystem uses the first 64MB for its journal.&nbsp; If we use =
Raid-0 for
     that, we may not recover from disk failures.&nbsp; And a failure =
during
     log replay means immediate full filesystem consistency check (eek =
repair)
     and associated downtime.&nbsp; Customers who prefer to use this
     configuration must be warned about =
this.<o:p></o:p></span></font></li>
 <li class=3DMsoNormal style=3D'color:navy;mso-list:l0 level1 =
lfo1'><font size=3D2
     color=3Dnavy face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>Our
     filesystem places some of the critical metadata (block free =
bitmaps,
     inodes, snapshots etc) in the initial portion of the filesystem
     immediately after the filesystem log.&nbsp; So if this portion is =
not RAID
     protected, we may not recover from disk failures.&nbsp; Also note =
that
     usually these portions of the filesystem is heavily read and =
written.&nbsp;
     It is preferred to use the lun just for the size of the filesystem =
log for
     the first Raid-1/0 lun instead of everybody using random sized LUNs =
for
     this.<o:p></o:p></span></font></li>
</ol>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>Regards,<o:p></o:p></span></font></p=
>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>Jobi<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'><o:p>&nbsp;</o:p></span></font></p>

<div>

<div class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><font =
size=3D3
face=3D"Times New Roman"><span style=3D'font-size:12.0pt'>

<hr size=3D2 width=3D"100%" align=3Dcenter tabindex=3D-1>

</span></font></div>

<p class=3DMsoNormal><b><font size=3D2 face=3DTahoma><span =
style=3D'font-size:10.0pt;
font-family:Tahoma;font-weight:bold'>From:</span></font></b><font =
size=3D2
face=3DTahoma><span style=3D'font-size:10.0pt;font-family:Tahoma'> Tim =
Gardner <br>
<b><span style=3D'font-weight:bold'>Sent:</span></b> Tuesday, January =
29, 2008
11:37 AM<br>
<b><span style=3D'font-weight:bold'>To:</span></b> dl-Engineering;
dl-Customer-Engineering; dl-se; Bob Miller<br>
<b><span style=3D'font-weight:bold'>Subject:</span></b> FW: Cleary =
Gottlieb -
NDMP Performance &amp; Data Mirroring</span></font><o:p></o:p></p>

</div>

<p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span =
style=3D'font-size:
12.0pt'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>Folks,<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>It is not often that we get an =
email like
this from a customer telling us how satisfied they are with our =
product.<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>Usually we only hear from customers =
when
the product is not working so I wanted to share this with =
you.<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>This also very nicely validates the
information that we discussed in the last tech talk about storage
configuration.<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'>Tim<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span =
style=3D'font-size:
10.0pt;font-family:Arial;color:navy'><o:p>&nbsp;</o:p></span></font></p>

<div>

<div class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><font =
size=3D3
face=3D"Times New Roman"><span style=3D'font-size:12.0pt'>

<hr size=3D2 width=3D"100%" align=3Dcenter tabindex=3D-1>

</span></font></div>

<p class=3DMsoNormal><b><font size=3D2 face=3DTahoma><span =
style=3D'font-size:10.0pt;
font-family:Tahoma;font-weight:bold'>From:</span></font></b><font =
size=3D2
face=3DTahoma><span style=3D'font-size:10.0pt;font-family:Tahoma'> David =
Flewitt
[mailto:DFlewitt@cgsh.com] <br>
<b><span style=3D'font-weight:bold'>Sent:</span></b> Tuesday, January =
29, 2008
10:57 AM<br>
<b><span style=3D'font-weight:bold'>To:</span></b> <st1:PersonName =
w:st=3D"on">Tim
 Gardner</st1:PersonName><br>
<b><span style=3D'font-weight:bold'>Cc:</span></b> Bill Duffy; =
<st1:PersonName
w:st=3D"on">Robert Campana</st1:PersonName>; Scott Moyer; =
DC-Network@cgsh.com;
William Tedrow; John R Daggan; Daniel Stotz<br>
<b><span style=3D'font-weight:bold'>Subject:</span></b> Cleary Gottlieb =
- NDMP
Performance &amp; Data Mirroring</span></font><o:p></o:p></p>

</div>

<p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span =
style=3D'font-size:
12.0pt'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span =
style=3D'font-size:
12.0pt'><br>
</span></font><font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:
Arial'>Good afternoon Tim.</span></font> <br>
<br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>I just
wanted to give you some preliminary feedback on our resent =
tests.</span></font>
<br>
<br>
<u><font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>NDMP</span></font></u>
<br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'><br>
Bill Duffy upgraded all of our Bobcats to V3.2. &nbsp;Following your
recommendation we re-structured our RAID groups on our storage array as
follows.</span></font> <br>
<br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>1st
LUN of each volume &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp;40GB
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;RAID 1/0 (a total =
of ten
146GB Fibre Channel disks)</span></font> <br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>2nd
LUN of each volume &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp;100GB
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;RAID 5 (a total =
of five
500GB ATA disks)</span></font> <br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>3rd
LUN of each volume &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp;100GB
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;RAID 5 (a total =
of five
500GB ATA disks)</span></font> <br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>4th
LUN of each volume &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp;100GB
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;RAID 5 (a total =
of five
500GB ATA disks)</span></font> <br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>5th
LUN of each volume &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp;100GB
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;RAID 5 (a total =
of five
500GB ATA disks)</span></font> <br>
<br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>I
created a volume using the above structure and copied over the same =
sample data
we used previously namely, 270GB and approx 5 million files. &nbsp;When =
we ran
a full NDMP backup job on this data previously, it took approx 14 hours =
to
complete (19GB/hour).</span></font> <br>
<br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>I am
pleased (and thankful) to report that the same full backup of the same =
data set
using the above structure completed in approx 2.5 hours
(108GB/hour)!!!!!!!!!!!!! &nbsp;That's a five fold improvement in =
performance.
&nbsp;We consider that &quot;spectacular&quot;!</span></font> <br>
<br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>So
although I was a little sceptical about your RAID group approach =
initially, I
am now convinced it is best practice.</span></font> <br>
<br>
<u><font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>Data
Mirroring</span></font></u> <br>
<br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>Since
the OS upgrade to Ver 3.2 we have initiated and successfully completed =
two
baseline data mirrors. &nbsp;One mirror was 270GB (5 million files) and =
the
other was 470GB (7.5 million files). &nbsp;Although they predictable =
took
several days to finish, they did finish! &nbsp;No aborting or Bobcat
re-booting.</span></font> <br>
<br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>We are
in the process of upgrading our Riverbed IP optimizer to a model that =
will
handle a 45Mbps WAN link. &nbsp;The above test was run with the existing =
10Mbps
link.</span></font> <br>
<br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>Thanks
again for your assistance.</span></font> <br>
<br>
<font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;font-family:Arial'>Regards,<br>
<br>
Dave Flewitt<br>
____________________<br>
Network Analyst<br>
Cleary Gottlieb Steen &amp; Hamilton LLP<br>
<st1:Street w:st=3D"on"><st1:address w:st=3D"on">2000 Pennsylvania =
Avenue N.W.</st1:address></st1:Street><br>
<st1:place w:st=3D"on"><st1:City w:st=3D"on">Washington</st1:City>, =
<st1:State
 w:st=3D"on">DC</st1:State> <st1:PostalCode =
w:st=3D"on">20006</st1:PostalCode></st1:place><br>
Tel: 202-974-1726<br>
dflewitt@cgsh.com<br>
http://www.clearygottlieb.com</span></font> <o:p></o:p></p>

<pre wrap><font size=3D2 face=3D"Courier New"><span =
style=3D'font-size:10.0pt'>This message is being sent from a law firm =
and may contain<o:p></o:p></span></font></pre><pre><font
size=3D2 face=3D"Courier New"><span =
style=3D'font-size:10.0pt'>confidential or privileged information.&nbsp; =
If you are not<o:p></o:p></span></font></pre><pre><font
size=3D2 face=3D"Courier New"><span style=3D'font-size:10.0pt'>the =
intended recipient, please advise the =
sender<o:p></o:p></span></font></pre><pre><font
size=3D2 face=3D"Courier New"><span =
style=3D'font-size:10.0pt'>immediately by reply e-mail and delete this =
message and<o:p></o:p></span></font></pre><pre><font
size=3D2 face=3D"Courier New"><span style=3D'font-size:10.0pt'>any =
attachments without retaining a =
copy.<o:p></o:p></span></font></pre></div>

</body>

</html>

------_=_NextPart_001_01C862EC.E6CB34CF--
