X-MimeOLE: Produced By Microsoft Exchange V6.5
Received: by onstor-exch02.onstor.net 
	id <01C74954.7F22C8C2@onstor-exch02.onstor.net>; Mon, 5 Feb 2007 10:36:06 -0800
MIME-Version: 1.0
Content-Type: multipart/report;
	report-type=delivery-status;
	boundary="----_=_NextPart_001_01C74954.7F22C8C2"
X-DSNContext: 335a7efd - 4523 - 00000001 - 80040546
Content-class: urn:content-classes:dsn
Subject: Undeliverable: Re: Test Plan for Small file performance
Date: Mon, 5 Feb 2007 10:36:06 -0800
Message-ID: <yGiNJ4sQE000001fe@onstor-exch02.onstor.net>
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
Thread-Topic: Test Plan for Small file performance
Thread-Index: AcdJVH8gqPEMRitbQhipeC9Gz+46hgAAAABm
From: "System Administrator" <postmaster@onstor.com>
To: "Andy Sharp" <andy.sharp@onstor.com>

This is a multi-part message in MIME format.

------_=_NextPart_001_01C74954.7F22C8C2
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Your message

  To:      Brian DeForest
  Cc:      Raj Kumar; dl-Design Review; dl-QA
  Subject: Re: Test Plan for Small file performance
  Sent:    Mon, 5 Feb 2007 10:36:04 -0800

did not reach the following recipient(s):

dl-Design Review on Mon, 5 Feb 2007 10:36:05 -0800
    You do not have permission to send to this recipient.  For
assistance, contact your system administrator.
    <onstor-exch02.onstor.net #5.7.1>

------_=_NextPart_001_01C74954.7F22C8C2
Content-Type: message/delivery-status
Content-Transfer-Encoding: 7bit

Reporting-MTA: dns; onstor-exch02.onstor.net

Final-Recipient: RFC822; dl-designreview@onstor.com
Action: failed
Status: 5.7.1
X-Supplementary-Info: <onstor-exch02.onstor.net #5.7.1>
X-Display-Name: dl-Design Review

------_=_NextPart_001_01C74954.7F22C8C2
Content-Type: message/rfc822
Content-Transfer-Encoding: 7bit

X-MimeOLE: Produced By Microsoft Exchange V6.5
Received: by onstor-exch02.onstor.net 
	id <01C74954.7E655880@onstor-exch02.onstor.net>; Mon, 5 Feb 2007 10:36:05 -0800
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_002_01C74954.7E655880"
Content-class: urn:content-classes:message
Subject: Re: Test Plan for Small file performance
Date: Mon, 5 Feb 2007 10:36:04 -0800
Message-ID: <20070205103604.378bfcdc@ripper.onstor.net>
In-Reply-To: <BB375AF679D4A34E9CA8DFA650E2B04E0250257E@onstor-exch02.onstor.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: Test Plan for Small file performance
Thread-Index: AcdJVH8gqPEMRitbQhipeC9Gz+46hg==
References: <BB375AF679D4A34E9CA8DFA650E2B04E01D4D471@onstor-exch02.onstor.net><BB375AF679D4A34E9CA8DFA650E2B04E0250257E@onstor-exch02.onstor.net>
From: "Andy Sharp" <andy.sharp@onstor.com>
To: "Brian DeForest" <brian.deforest@onstor.com>
Cc: "Raj Kumar" <raj.kumar@onstor.com>,
	"dl-Design Review" <dl-designreview@onstor.com>,
	"dl-QA" <dl-qa@onstor.com>

This is a multi-part message in MIME format.

------_=_NextPart_002_01C74954.7E655880
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

I wanted to add something that I didn't think of in the review meeting,
and that is that the performance tests should start with an expected
result *before* the tests are run.  For example, the person
implementing the test should determine ahead of time roughly how long
the operation is expected to take if the filer has decent performance.

So for a 'cp -r' test, an expected number could be arrived at by trying
the same test against various Unix NFS servers (Linux, Solaris) with
DAS, and those numbers could be combined with a 'pencil in the ceiling'
calculation, like 'transfering 500MB of files, 25,000 files, should
xfer at roughly 10MB/s, so 50 seconds, give or take'.  The 10MB/s
figure comes from assuming that while some files might be multiple MBs
in size, most are less than 10KB in size.

That way if the results are way off, we'll know right away, and we'll
also know if the [expected] improvement from Lambo to Delo is
improvement enough.

But I'm sure everyone knew that already.  I'm just sayin'.

Cheers,

a=20

On Mon, 5 Feb 2007 10:14:35 -0800 "Brian DeForest"
<brian.deforest@onstor.com> wrote:

> Raj, a few suggestions were mentioned in the design review that would
> provide additional data points.
>=20
> In addition to 30,000 files per test, you could add another dimension
> to the testing to vary the number of files in the directory, e.g. 100,
> 1000, 2000, ...32000 files.   This would verify performance scales
> linearly (as expected) as well as cover the range of likely/typical
> customer tests..   Cutting/pasting 30,000 files from explorer may
> take a looong time...   (Not sure why this is a magic number). =20
>=20
> The other area mentioned was to include tar/cpio testing to determine
> if any additional (future) work is needed and also determine if any
> performance improvements are achieved in Delorean (though not a goal
> or requirement).  Rsync and robocopy testing were mentioned for
> similar reasons.
>=20
> _____________________________________________
> From: Raj Kumar=20
> Sent: Wednesday, December 27, 2006 11:26 AM
> To: dl-Design Review; dl-QA
> Subject: Test Plan for Small file performance
>=20
> Hi,
>=20
> This is the test plan based on the Delorean PRD & the draft functional
> specification for "small file performance". Please review and send the
> comments by 01/03/2007.
>=20
> Test Plan:
> \\mightydog\Program
> =
Management\Delorean\QE\TestPlans\SmallFilePerformance_TestPlan_Ver0_1.do
> c
>=20
> Related Documents:
> PRD: \\mightydog\Program Management\Delorean\Marketing\Delorean
> MRD-PRD-REV1-6.xls
> Draft functional spec: \\mightydog\Program
> Management\Delorean\QE\DRAFT-Smal fileperformance-Spec.doc
>=20
> Thanks.
>=20
> --kumar :-)


------_=_NextPart_002_01C74954.7E655880
Content-Type: text/html;
	charset="iso-8859-1"
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=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
6.5.7650.28">
<TITLE>Re: Test Plan for Small file performance</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->

<P><FONT SIZE=3D2>I wanted to add something that I didn't think of in =
the review meeting,<BR>
and that is that the performance tests should start with an expected<BR>
result *before* the tests are run.&nbsp; For example, the person<BR>
implementing the test should determine ahead of time roughly how =
long<BR>
the operation is expected to take if the filer has decent =
performance.<BR>
<BR>
So for a 'cp -r' test, an expected number could be arrived at by =
trying<BR>
the same test against various Unix NFS servers (Linux, Solaris) with<BR>
DAS, and those numbers could be combined with a 'pencil in the =
ceiling'<BR>
calculation, like 'transfering 500MB of files, 25,000 files, should<BR>
xfer at roughly 10MB/s, so 50 seconds, give or take'.&nbsp; The =
10MB/s<BR>
figure comes from assuming that while some files might be multiple =
MBs<BR>
in size, most are less than 10KB in size.<BR>
<BR>
That way if the results are way off, we'll know right away, and =
we'll<BR>
also know if the [expected] improvement from Lambo to Delo is<BR>
improvement enough.<BR>
<BR>
But I'm sure everyone knew that already.&nbsp; I'm just sayin'.<BR>
<BR>
Cheers,<BR>
<BR>
a<BR>
<BR>
On Mon, 5 Feb 2007 10:14:35 -0800 &quot;Brian DeForest&quot;<BR>
&lt;brian.deforest@onstor.com&gt; wrote:<BR>
<BR>
&gt; Raj, a few suggestions were mentioned in the design review that =
would<BR>
&gt; provide additional data points.<BR>
&gt;<BR>
&gt; In addition to 30,000 files per test, you could add another =
dimension<BR>
&gt; to the testing to vary the number of files in the directory, e.g. =
100,<BR>
&gt; 1000, 2000, ...32000 files.&nbsp;&nbsp; This would verify =
performance scales<BR>
&gt; linearly (as expected) as well as cover the range of =
likely/typical<BR>
&gt; customer tests..&nbsp;&nbsp; Cutting/pasting 30,000 files from =
explorer may<BR>
&gt; take a looong time...&nbsp;&nbsp; (Not sure why this is a magic =
number).&nbsp;<BR>
&gt;<BR>
&gt; The other area mentioned was to include tar/cpio testing to =
determine<BR>
&gt; if any additional (future) work is needed and also determine if =
any<BR>
&gt; performance improvements are achieved in Delorean (though not a =
goal<BR>
&gt; or requirement).&nbsp; Rsync and robocopy testing were mentioned =
for<BR>
&gt; similar reasons.<BR>
&gt;<BR>
&gt; _____________________________________________<BR>
&gt; From: Raj Kumar<BR>
&gt; Sent: Wednesday, December 27, 2006 11:26 AM<BR>
&gt; To: dl-Design Review; dl-QA<BR>
&gt; Subject: Test Plan for Small file performance<BR>
&gt;<BR>
&gt; Hi,<BR>
&gt;<BR>
&gt; This is the test plan based on the Delorean PRD &amp; the draft =
functional<BR>
&gt; specification for &quot;small file performance&quot;. Please review =
and send the<BR>
&gt; comments by 01/03/2007.<BR>
&gt;<BR>
&gt; Test Plan:<BR>
&gt; \\mightydog\Program<BR>
&gt; =
Management\Delorean\QE\TestPlans\SmallFilePerformance_TestPlan_Ver0_1.do<=
BR>
&gt; c<BR>
&gt;<BR>
&gt; Related Documents:<BR>
&gt; PRD: \\mightydog\Program Management\Delorean\Marketing\Delorean<BR>
&gt; MRD-PRD-REV1-6.xls<BR>
&gt; Draft functional spec: \\mightydog\Program<BR>
&gt; Management\Delorean\QE\DRAFT-Smal fileperformance-Spec.doc<BR>
&gt;<BR>
&gt; Thanks.<BR>
&gt;<BR>
&gt; --kumar :-)<BR>
<BR>
</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_002_01C74954.7E655880--

------_=_NextPart_001_01C74954.7F22C8C2--
