X-MimeOLE: Produced By Microsoft Exchange V6.5
Received: by onstor-exch02.onstor.net 
	id <01C76FEC.6A134C66@onstor-exch02.onstor.net>; Mon, 26 Mar 2007 14:19:18 -0700
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C76FEC.6A134C66"
Content-class: urn:content-classes:message
Subject: Agenda item
Date: Mon, 26 Mar 2007 14:19:15 -0700
Message-ID: <BB375AF679D4A34E9CA8DFA650E2B04E02FAE66C@onstor-exch02.onstor.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: Agenda item
Thread-Index: Acdv7GhRw26M/QrRQ4eOUlrczvOjvQ==
From: "Jonathan Goldick" <jonathan.goldick@onstor.com>
To: "dl-Cougar" <dl-Cougar@onstor.com>

This is a multi-part message in MIME format.

------_=_NextPart_001_01C76FEC.6A134C66
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

It doesn't have to be tomorrow but at some point I want to discuss the
current polling architecture and what we can expect from cougar in terms
of performance.

Max has been doing some KPI work to measure the time we spend waiting
for the polling loops to recognize that there is work to be done, my
unproven contention has been that this is a large fraction of our
current performance woes.  While I think that this becomes less of a
factor as the load increases, the benchmarks we get killed on today
would seem to be highly vulnerable to polling loop latency.

While the merge of the FC into the FP should eliminate one of our
polling loops we will likely be introducing new ones when handing off
work between the ACPU(s) and NCPU(s) as discussed in previous meetings.

If I'm wrong and this doesn't matter then I'll happily move on to other
irrelevant issues :-)


------_=_NextPart_001_01C76FEC.6A134C66
Content-Type: text/html;
	charset="us-ascii"
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 =
6.5.7652.24">
<TITLE>Agenda item</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/rtf format -->

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
SIZE=3D2 FACE=3D"Arial">It doe</FONT><FONT SIZE=3D2 =
FACE=3D"Arial">sn</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial">&#8217;</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Arial">t have to be tomorrow but at some point I want to discuss =
the current polling architecture and wh</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Arial">at we can expect from</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Arial"> cougar</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial"></FONT> <FONT SIZE=3D2 =
FACE=3D"Arial">in terms of performance.</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial">Max has =
been doing some</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"> <FONT SIZE=3D2 FACE=3D"Arial">KPI</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"> <FONT SIZE=3D2 =
FACE=3D"Arial">work to measure the time we spend waiting for the polling =
loop</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
SIZE=3D2 FACE=3D"Arial">s</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial"> to recognize that there is =
work to be done, my unproven contention has been that this is a large =
fraction</FONT> <FONT SIZE=3D2 FACE=3D"Arial">of our =
current</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"> =
<FONT SIZE=3D2 FACE=3D"Arial">performance</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Arial"></FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"> <FONT SIZE=3D2 FACE=3D"Arial">woes</FONT><FONT SIZE=3D2 =
FACE=3D"Arial">.</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial">&nbsp;</FONT> <FONT =
SIZE=3D2 FACE=3D"Arial">While I think that this becomes less of a factor =
as the load increases,</FONT> <FONT SIZE=3D2 FACE=3D"Arial">the =
benchmarks we get killed on today would seem to be highly vulnerable to =
polling loop latenc</FONT><FONT SIZE=3D2 =
FACE=3D"Arial">y.</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial">While =
the merge of the FC into the FP should</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"> <FONT SIZE=3D2 =
FACE=3D"Arial">eliminate one of our polling loops we will likely be =
introducing new ones when</FONT><FONT SIZE=3D2 FACE=3D"Arial"> handing =
off work between the ACPU(s) and NCPU(s) as discussed in previous =
meetings.</FONT></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial">If =
I</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
SIZE=3D2 FACE=3D"Arial">&#8217;</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Arial">m wrong and this doesn</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Arial">&#8217;</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"><FONT SIZE=3D2 FACE=3D"Arial">t matter then =
I</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT =
SIZE=3D2 FACE=3D"Arial">&#8217;</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"><FONT SIZE=3D2 =
FACE=3D"Arial">ll happily move on to other irrelevant =
issues</FONT></SPAN><SPAN LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"> =
<FONT FACE=3D"Wingdings" SIZE=3D2>J</FONT></SPAN><SPAN =
LANG=3D"en-us"></SPAN><SPAN LANG=3D"en-us"></SPAN></P>

<P DIR=3DLTR><SPAN LANG=3D"en-us"></SPAN><SPAN =
LANG=3D"en-us"></SPAN></P>

</BODY>
</HTML>
------_=_NextPart_001_01C76FEC.6A134C66--
