<html><body>
<p>Thanks Jonathan for the good tips. :-)<br>
<br>
- Harris<br>
<br>
<img width="16" height="16" src="cid:1__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt="Inactive hide details for Jonathan Lassoff ---10/02/2010 AM 12:14:30---While having an increased MTU across your WAN can improv"><font color="#424282">Jonathan Lassoff ---10/02/2010 AM 12:14:30---While having an increased MTU across your WAN can improve throughout greatly, I would suggest tuning your TCP stack for a "Long</font><br>
<br>
<table width="100%" border="0" cellspacing="0" cellpadding="0">
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">From:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt=""><br>
<font size="2">Jonathan Lassoff <jof@thejof.com></font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">To:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt=""><br>
<font size="2">Harris Hui/Hong Kong/IBM@IBMHK</font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">Cc:</font></td><td width="100%" valign="middle"><img width="1" height="1" src="cid:2__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt=""><br>
<font size="2">juniper-nsp@puck.nether.net</font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">Date:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt=""><br>
<font size="2">10/02/2010 AM 12:14</font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">Subject:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=C7BBFD23DF84369B8f9e8a93df93@hk1.ibm.com" border="0" alt=""><br>
<font size="2">Re: [j-nsp] J6350 Jumbo frame MTU and OSPF setting</font></td></tr>
</table>
<hr width="100%" size="2" align="left" noshade style="color:#8091A5; "><br>
<br>
<br>
<font size="4">While having an increased MTU across your WAN can improve throughout greatly, I would suggest tuning your TCP stack for a "Long Fat Pipe", as many operating systems are not designed to work well with high-throughput, high-latency links.</font>
<p><font size="4">There are some good tips here: </font><a href="http://fasterdata.es.net/"><u><font size="4" color="#0000FF">http://fasterdata.es.net/</font></u></a>
<p><font size="4">Cheers,<br>
jof</font>
<ul>
<ul><font size="4">On Sep 30, 2010 10:38 PM, "Harris Hui" <</font><a href="mailto:harris.hui@hk1.ibm.com"><u><font size="4" color="#0000FF">harris.hui@hk1.ibm.com</font></u></a><font size="4">> wrote:<br>
<br>
<br>
<br>
Dear all,<br>
<br>
We had subscribed a private line circuit between 2 different data center<br>
for Data Backup and replication. The bandwidth of the private line is<br>
100Mbps.<br>
<br>
According to the provider, The Circuit is Built across their<br>
Network as 2 STS1's or High Speed DS3's which equals 100meg.<br>
<br>
Their GE port setting as follows.<br>
<br>
MTU Size - 9600<br>
Auto Negotiation - OFF<br>
Remote Client Fail - Disabled.<br>
<br>
The private circuit is connected directly to the fiber module of our J6350<br>
Services router at each Data Center. The Circuit is up and running but when<br>
we perform some TCP throughput test, we only get ~3Mbps for a Single TCP<br>
session with iPerf and the latency between two data center across the<br>
private circuit is ~80ms.<br>
<br>
I am trying to configure our J6350 fiber interface to MTU 9192 to get a<br>
better TCP throughput. However, I can only able to configure the MTU size<br>
below 1500, when I configure the MTU to 9192 and commit the changes, it<br>
still shows MTU 1500 on the physical interface.<br>
<br>
Do you have any experience on using Jumbo frame MTU size on the J6350? We<br>
are also running OSPF across the private circuit, is JUNOS support "OSPF<br>
ignore-mtu" like cisco?<br>
<br>
Please advise.<br>
<br>
Fiber module<br>
================================================<br>
FPC 3 REV 18 750-013599 AAAH7361 FPC<br>
PIC 0 1x GE SFP<br>
Xcvr 0 REV 02 740-011614 PG336CS SFP-LX10<br>
<br>
show interfaces ge-3/0/0<br>
speed 1g;<br>
mtu 1400;<br>
link-mode full-duplex;<br>
gigether-options {<br>
no-auto-negotiation;<br>
}<br>
unit 0 {<br>
family inet {<br>
address xxx.xxx.xxx.253/30;<br>
}<br>
}<br>
<br>
harris@J6350# run show interfaces ge-3/0/0<br>
Physical interface: ge-3/0/0, Enabled, Physical link is Up<br>
Interface index: 152, SNMP ifIndex: 184<br>
Link-level type: Ethernet, MTU: 1400, Speed: 1000mbps, BPDU Error: None,<br>
MAC-REWRITE Error: None, Loopback: Disabled,<br>
Source filtering: Disabled, Flow control: Enabled, Auto-negotiation:<br>
Disabled, Remote fault: Online<br>
Device flags : Present Running<br>
Interface flags: SNMP-Traps Internal: 0x4000<br>
Link flags : None<br>
CoS queues : 8 supported, 8 maximum usable queues<br>
Current address: b0:c6:9a:87:35:36, Hardware address: b0:c6:9a:87:35:36<br>
Last flapped : 2010-09-27 02:32:24 UTC (4d 02:29 ago)<br>
Input rate : 0 bps (0 pps)<br>
Output rate : 0 bps (0 pps)<br>
Active alarms : None<br>
Active defects : None<br>
<br>
show interfaces ge-3/0/0<br>
speed 1g;<br>
mtu 9192;<br>
link-mode full-duplex;<br>
gigether-options {<br>
no-auto-negotiation;<br>
}<br>
unit 0 {<br>
family inet {<br>
address xxx.xxx.xxx.253/30;<br>
}<br>
}<br>
<br>
run show interfaces ge-3/0/0<br>
Physical interface: ge-3/0/0, Enabled, Physical link is Up<br>
Interface index: 152, SNMP ifIndex: 184<br>
Link-level type: Ethernet, MTU: 1514, Speed: 1000mbps, BPDU Error: None,<br>
MAC-REWRITE Error: None, Loopback: Disabled,<br>
Source filtering: Disabled, Flow control: Enabled, Auto-negotiation:<br>
Disabled, Remote fault: Online<br>
Device flags : Present Running<br>
Interface flags: SNMP-Traps Internal: 0x4000<br>
Link flags : None<br>
CoS queues : 8 supported, 8 maximum usable queues<br>
Current address: b0:c6:9a:87:35:36, Hardware address: b0:c6:9a:87:35:36<br>
Last flapped : 2010-09-27 02:32:24 UTC (4d 02:35 ago)<br>
Input rate : 0 bps (0 pps)<br>
Output rate : 1192 bps (2 pps)<br>
Active alarms : None<br>
Active defects : None<br>
<br>
<br>
Thanks<br>
Harris<br>
_______________________________________________<br>
juniper-nsp mailing list </font><a href="mailto:juniper-nsp@puck.nether.net"><u><font size="4" color="#0000FF">juniper-nsp@puck.nether.net</font></u></a><u><font size="4" color="#0000FF"><br>
</font></u><a href="https://puck.nether.net/mailman/listinfo/juniper-nsp" target="_blank"><u><font size="4" color="#0000FF">https://puck.nether.net/mailman/listinfo/juniper-nsp</font></u></a>
<p>
<p></ul>
</ul>
</body></html>