<html><body>
<p>Hello Eli,<br>
<br>
What's the CPU utilization? I do see much better performance (40-100%) regarding ehca. I wonder whether CPU is the limitation there. Have you tried multiple streams, duplex?<br>
<br>
Thanks<br>
Shirley <br>
<br>
<br>
<img width="16" height="16" src="cid:1__=08BBF976DFCF1C348f9e8a93df938@us.ibm.com" border="0" alt="Inactive hide details for Eli Cohen <eli@mellanox.co.il>">Eli Cohen <eli@mellanox.co.il><br>
<br>
<br>
<table width="100%" border="0" cellspacing="0" cellpadding="0">
<tr valign="top"><td style="background-image:url(cid:2__=08BBF976DFCF1C348f9e8a93df938@us.ibm.com); background-repeat: no-repeat; " width="29%">
<ul>
<ul>
<ul>
<ul><b><font size="2">Eli Cohen <eli@mellanox.co.il></font></b><font size="2"> </font><br>
<font size="2">Sent by: general-bounces@lists.openfabrics.org</font>
<p><font size="2">02/04/08 08:31 AM</font></ul>
</ul>
</ul>
</ul>
</td><td width="71%">
<table width="100%" border="0" cellspacing="0" cellpadding="0">
<tr valign="top"><td width="1%"><img width="58" height="1" src="cid:3__=08BBF976DFCF1C348f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<div align="right"><font size="2">To</font></div></td><td width="100%"><img width="1" height="1" src="cid:3__=08BBF976DFCF1C348f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">mashirle@linux.vnet.ibm.com, Tziporet Koren <tziporet@mellanox.co.il>, openfabrics <general@lists.openfabrics.org></font></td></tr>
<tr valign="top"><td width="1%"><img width="58" height="1" src="cid:3__=08BBF976DFCF1C348f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<div align="right"><font size="2">cc</font></div></td><td width="100%"><img width="1" height="1" src="cid:3__=08BBF976DFCF1C348f9e8a93df938@us.ibm.com" border="0" alt=""><br>
</td></tr>
<tr valign="top"><td width="1%"><img width="58" height="1" src="cid:3__=08BBF976DFCF1C348f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<div align="right"><font size="2">Subject</font></div></td><td width="100%"><img width="1" height="1" src="cid:3__=08BBF976DFCF1C348f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">[ofa-general] comparison between 4K mtu and none 4K mtu</font></td></tr>
</table>
<table border="0" cellspacing="0" cellpadding="0">
<tr valign="top"><td width="58"><img width="1" height="1" src="cid:3__=08BBF976DFCF1C348f9e8a93df938@us.ibm.com" border="0" alt=""></td><td width="336"><img width="1" height="1" src="cid:3__=08BBF976DFCF1C348f9e8a93df938@us.ibm.com" border="0" alt=""></td></tr>
</table>
</td></tr>
</table>
<br>
<tt>FYI<br>
<br>
I have made a few experiments with the 4K mtu patches based on Roland's<br>
tree. The results are in the attached execl file. I used netperf to take<br>
these results.<br>
[attachment "mtu_4k.xls" deleted by Shirley Ma/Beaverton/IBM] _______________________________________________<br>
general mailing list<br>
general@lists.openfabrics.org<br>
</tt><tt><a href="http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general">http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general</a></tt><tt><br>
<br>
To unsubscribe, please visit </tt><tt><a href="http://openib.org/mailman/listinfo/openib-general">http://openib.org/mailman/listinfo/openib-general</a></tt><br>
</body></html>