<tt><font size=2>>On 04/04/2012 03:15 AM, Simon Duboue wrote:<br>
>><br>
>> Could the total throughput be considered as the sum of the two
obtained<br>
>> results?<br>
><br>
>Maybe :) There is no explicit synchronization in netperf. There
are a <br>
>few ways to address issues of skew error:<br>
><br>
</font></tt><a href="http://www.netperf.org/svn/netperf2/trunk/doc/netperf.html#Using-Netperf-to-Measure-Aggregate-Performance"><tt><font size=2>>http://www.netperf.org/svn/netperf2/trunk/doc/netperf.html#Using-Netperf-to-Measure-Aggregate-Performance</font></tt></a><tt><font size=2><br>
><br>
>The one I currently use the most is the "demo mode with post-processing"
<br>
>path, though that has an issue with UDP_STREAM - the interim results
<br>
>reported by netperf in a UDP_STREAM test will be sender side, not <br>
>receiver side. In that case the (ab)use of confidence intervals
path <br>
>may be better (though take longer).<br>
><br>
>While the local CPU util in your pair of UDP_STREAM tests was the same,
<br>
>the remote CPU util differed enough that I would not trust the results
<br>
>to have low skew error.<br>
</font></tt>
<br><tt><font size=2>Now I get the same version of netperf on both sides.
The results seem to be in the same range as previously.</font></tt>
<br><tt><font size=2>I come back on what you say before concerning the
use of the -i option. As expected, I have a warning for the confidence
intervals not reached.</font></tt>
<br><tt><font size=2><br>
netperf -i 10 -cC -p 17170 -l 30 -P 0 -H $1 -t TCP_STREAM -- -C &</font></tt>
<br><tt><font size=2>netperf -i 10 -cC -p 17170 -l 30 -P 0 -H $1 -t TCP_STREAM
-- -C &</font></tt>
<br><tt><font size=2>netperf -i 10 -cC -p 17170 -l 30 -P 0 -H $1 -t TCP_STREAM
-- -C &</font></tt>
<br><tt><font size=2>netperf -i 10 -cC -p 17170 -l 30 -P 0 -H $1 -t TCP_STREAM
-- -C </font></tt>
<br>
<br><tt><font size=2>!!! WARNING</font></tt>
<br><tt><font size=2>!!! Desired confidence was not achieved within the
specified iterations.</font></tt>
<br><tt><font size=2>!!! This implies that there was variability in the
test environment that</font></tt>
<br><tt><font size=2>!!! must be investigated before going further.</font></tt>
<br><tt><font size=2>!!! Confidence intervals: Throughput
: 11.9%</font></tt>
<br><tt><font size=2>!!!
Local CPU util : 10.9%</font></tt>
<br><tt><font size=2>!!!
Remote CPU util : 70.5%</font></tt>
<br>
<br><tt><font size=2> 87380 16384 16384 30.06
641.87 6.77 10.06 1.732
10.306 </font></tt>
<br><tt><font size=2>!!! WARNING</font></tt>
<br><tt><font size=2>!!! Desired confidence was not achieved within the
specified iterations.</font></tt>
<br><tt><font size=2>!!! This implies that there was variability in the
test environment that</font></tt>
<br><tt><font size=2>!!! must be investigated before going further.</font></tt>
<br><tt><font size=2>!!! Confidence intervals: Throughput
: 13.4%</font></tt>
<br><tt><font size=2>!!!
Local CPU util : 11.5%</font></tt>
<br><tt><font size=2>!!!
Remote CPU util : 294.0%</font></tt>
<br>
<br><tt><font size=2> 87380 16384 16384 30.08
668.97 6.79 7.66
1.668 7.879 </font></tt>
<br><tt><font size=2>!!! WARNING</font></tt>
<br><tt><font size=2>!!! Desired confidence was not achieved within the
specified iterations.</font></tt>
<br><tt><font size=2>!!! This implies that there was variability in the
test environment that</font></tt>
<br><tt><font size=2>!!! must be investigated before going further.</font></tt>
<br><tt><font size=2>!!! Confidence intervals: Throughput
: 10.4%</font></tt>
<br><tt><font size=2>!!!
Local CPU util : 12.7%</font></tt>
<br><tt><font size=2>!!!
Remote CPU util : 290.8%</font></tt>
<br>
<br><tt><font size=2> 87380 16384 16384 30.10
683.27 6.79 7.68
1.631 7.436 </font></tt>
<br><tt><font size=2>!!! WARNING</font></tt>
<br><tt><font size=2>!!! Desired confidence was not achieved within the
specified iterations.</font></tt>
<br><tt><font size=2>!!! This implies that there was variability in the
test environment that</font></tt>
<br><tt><font size=2>!!! must be investigated before going further.</font></tt>
<br><tt><font size=2>!!! Confidence intervals: Throughput
: 15.0%</font></tt>
<br><tt><font size=2>!!!
Local CPU util : 11.4%</font></tt>
<br><tt><font size=2>!!!
Remote CPU util : 70.2%</font></tt>
<br>
<br><tt><font size=2> 87380 16384 16384 30.12
643.66 6.79 10.14 1.735
10.239</font></tt>
<br>
<br><tt><font size=2>Considering the cpu utilization, can I consider that
the test are running in parallel?</font></tt>
<br>
<br><tt><font size=2>I increase the length of the test and the number of
iterations and I still not achieve this interval but they decrease.</font></tt>
<br><font size=2 face="Courier New">netperf -i 100 </font><tt><font size=2>-cC
</font></tt><font size=2 face="Courier New">-p 17170 -l 60 -P 0 -H $1 -t
TCP_STREAM -- -C &</font>
<br><font size=2 face="Courier New">netperf -i 100 </font><tt><font size=2>-cC
</font></tt><font size=2 face="Courier New">-p 17170 -l 60 -P 0 -H $1 -t
TCP_STREAM -- -C &</font>
<br><font size=2 face="Courier New">netperf -i 100 </font><tt><font size=2>-cC
</font></tt><font size=2 face="Courier New">-p 17170 -l 60 -P 0 -H $1 -t
TCP_STREAM -- -C &</font>
<br><font size=2 face="Courier New">netperf -i 100 </font><tt><font size=2>-cC
</font></tt><font size=2 face="Courier New">-p 17170 -l 60 -P 0 -H $1 -t
TCP_STREAM -- -C </font>
<br><tt><font size=2>Second results:</font></tt>
<br><font size=2 face="Courier New">!!! WARNING</font>
<br><font size=2 face="Courier New">!!! Desired confidence was not achieved
within the specified iterations.</font>
<br><font size=2 face="Courier New">!!! This implies that there was variability
in the test environment that</font>
<br><font size=2 face="Courier New">!!! must be investigated before going
further.</font>
<br><font size=2 face="Courier New">!!! Confidence intervals: Throughput
: 4.2%</font>
<br><font size=2 face="Courier New">!!!
Local CPU util : 4.1%</font>
<br><font size=2 face="Courier New">!!!
Remote CPU util : 9.1%</font>
<br>
<br><font size=2 face="Courier New"> 87380 16384 16384
60.06 642.17 6.58
10.74 1.680 10.985 </font>
<br><font size=2 face="Courier New">!!! WARNING</font>
<br><font size=2 face="Courier New">!!! Desired confidence was not achieved
within the specified iterations.</font>
<br><font size=2 face="Courier New">!!! This implies that there was variability
in the test environment that</font>
<br><font size=2 face="Courier New">!!! must be investigated before going
further.</font>
<br><font size=2 face="Courier New">!!! Confidence intervals: Throughput
: 6.1%</font>
<br><font size=2 face="Courier New">!!!
Local CPU util : 4.0%</font>
<br><font size=2 face="Courier New">!!!
Remote CPU util : 10.4%</font>
<br>
<br><font size=2 face="Courier New"> 87380 16384 16384
60.08 643.44 6.57
10.73 1.679 10.960 </font>
<br><font size=2 face="Courier New">!!! WARNING</font>
<br><font size=2 face="Courier New">!!! Desired confidence was not achieved
within the specified iterations.</font>
<br><font size=2 face="Courier New">!!! This implies that there was variability
in the test environment that</font>
<br><font size=2 face="Courier New">!!! must be investigated before going
further.</font>
<br><font size=2 face="Courier New">!!! Confidence intervals: Throughput
: 5.1%</font>
<br><font size=2 face="Courier New">!!!
Local CPU util : 4.1%</font>
<br><font size=2 face="Courier New">!!!
Remote CPU util : 9.0%</font>
<br>
<br><font size=2 face="Courier New"> 87380 16384 16384
60.09 648.31 6.58
10.76 1.665 10.896 </font>
<br><font size=2 face="Courier New">!!! WARNING</font>
<br><font size=2 face="Courier New">!!! Desired confidence was not achieved
within the specified iterations.</font>
<br><font size=2 face="Courier New">!!! This implies that there was variability
in the test environment that</font>
<br><font size=2 face="Courier New">!!! must be investigated before going
further.</font>
<br><font size=2 face="Courier New">!!! Confidence intervals: Throughput
: 5.5%</font>
<br><font size=2 face="Courier New">!!!
Local CPU util : 4.0%</font>
<br><font size=2 face="Courier New">!!!
Remote CPU util : 9.0%</font>
<br>
<br><font size=2 face="Courier New"> 87380 16384 16384
60.09 651.78 6.57
10.73 1.657 10.817 </font>
<br>
<br>
<br><font size=2 face="Courier New">For the -s -S and -m -M options, what
scenario could be a good start in order to test their influence?</font>
<br>
<br>
<br><font size=2 face="Courier New">Have a nice day.</font>
<br>
<br><font size=2 face="Courier New">Simon Duboué</font>
<br>
<br>