You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jmeter-dev@jakarta.apache.org by se...@apache.org on 2009/12/16 01:54:09 UTC
svn commit: r891088 - in /jakarta/jmeter/trunk/xdocs: changes.xml
usermanual/component_reference.xml usermanual/listeners.xml
Author: sebb
Date: Wed Dec 16 00:54:09 2009
New Revision: 891088
URL: http://svn.apache.org/viewvc?rev=891088&view=rev
Log:
Document improved Listener behaviour
Modified:
jakarta/jmeter/trunk/xdocs/changes.xml
jakarta/jmeter/trunk/xdocs/usermanual/component_reference.xml
jakarta/jmeter/trunk/xdocs/usermanual/listeners.xml
Modified: jakarta/jmeter/trunk/xdocs/changes.xml
URL: http://svn.apache.org/viewvc/jakarta/jmeter/trunk/xdocs/changes.xml?rev=891088&r1=891087&r2=891088&view=diff
==============================================================================
--- jakarta/jmeter/trunk/xdocs/changes.xml (original)
+++ jakarta/jmeter/trunk/xdocs/changes.xml Wed Dec 16 00:54:09 2009
@@ -154,7 +154,11 @@
<li>Bug 47952 - Added JSR223 Listener</li>
<li>Bug 47474 - View Results Tree support for plugin renderers</li>
<li>Allow Idle Time to be saved to sample log files</li>
-<li>Bug 48259 - Improve StatCalculator performance by using HashMap</li>
+<li>Bug 48259 - Improve StatCalculator performance by using TreeMap</li>
+<li>Listeners using SamplingStatCalculator have much reduced memory needs
+as the Sample cache has been moved to the new CachingStatCalculator class.
+In particular, Aggregate Report can now handle large numbers of samples.
+</li>
</ul>
<h3>Timers, Assertions, Config, Pre- & Post-Processors</h3>
Modified: jakarta/jmeter/trunk/xdocs/usermanual/component_reference.xml
URL: http://svn.apache.org/viewvc/jakarta/jmeter/trunk/xdocs/usermanual/component_reference.xml?rev=891088&r1=891087&r2=891088&view=diff
==============================================================================
--- jakarta/jmeter/trunk/xdocs/usermanual/component_reference.xml (original)
+++ jakarta/jmeter/trunk/xdocs/usermanual/component_reference.xml Wed Dec 16 00:54:09 2009
@@ -1954,12 +1954,21 @@
</p>
<ul>
<li>Simple Data Writer</li>
-<li>BeanShell Listener</li>
-<li>Assertion Results</li>
+<li>BeanShell/BSF Listener</li>
<li>Mailer Visualizer</li>
<li>Monitor Results</li>
<li>Summary Report</li>
</ul>
+<p>
+The following Listeners no longer need to keep copies of every single sample.
+Instead, samples with the same elapsed time are aggregated.
+Less memory is now needed, especially if most samples only take a second or two at most.
+</p>
+<ul>
+<li>Aggregate Report</li>
+<li>Aggregate Graph</li>
+<li>Distribution Graph</li>
+</ul>
<p>To minimise the amount of memory needed, use the Simple Data Writer, and use the CSV format.</p>
<p>
<note>
@@ -2170,10 +2179,14 @@
It is important to choose the sampler names correctly to get the best results from
the Aggregate Report.
</p>
-<note>
-Calculation of the <a href="glossary.html#Median">Median</a> and 90% Line (90<sup>th</sup> <a href="glossary.html#Percentile">percentile</a>) values requires a lot of memory as details of every Sample have to be saved.
-See the <complink name="Summary Report"/> for a similar Listener that does not need so much memory.
-</note>
+<p>
+Calculation of the <a href="glossary.html#Median">Median</a> and 90% Line (90<sup>th</sup> <a href="glossary.html#Percentile">percentile</a>) values requires additional memory.
+For JMeter 2.3.4 and earlier, details of each sample were saved separately, which meant a lot of memory was needed.
+JMeter now combines samples with the same elapsed time, so far less memory is used.
+However, for samples that take more than a few seconds, the probability is that fewer samples will have identical times,
+in which case more memory will be needed.
+See the <complink name="Summary Report"/> for a similar Listener that does not store individual samples and so needs constant memory.
+</p>
<ul>
<li>Label - The label of the sample.
If "Include group name in label?" is selected, then the name of the thread group is added as a prefix.
Modified: jakarta/jmeter/trunk/xdocs/usermanual/listeners.xml
URL: http://svn.apache.org/viewvc/jakarta/jmeter/trunk/xdocs/usermanual/listeners.xml?rev=891088&r1=891087&r2=891088&view=diff
==============================================================================
--- jakarta/jmeter/trunk/xdocs/usermanual/listeners.xml (original)
+++ jakarta/jmeter/trunk/xdocs/usermanual/listeners.xml Wed Dec 16 00:54:09 2009
@@ -241,12 +241,21 @@
</p>
<ul>
<li>Simple Data Writer</li>
-<li>BeanShell Listener</li>
-<li>Assertion Results</li>
+<li>BeanShell/BSF Listener</li>
<li>Mailer Visualizer</li>
<li>Monitor Results</li>
<li>Summary Report</li>
</ul>
+<p>
+The following Listeners no longer need to keep copies of every single sample.
+Instead, samples with the same elapsed time are aggregated.
+Less memory is now needed, especially if most samples only take a second or two at most.
+</p>
+<ul>
+<li>Aggregate Report</li>
+<li>Aggregate Graph</li>
+<li>Distribution Graph</li>
+</ul>
<p>To minimise the amount of memory needed, use the Simple Data Writer, and use the CSV format.</p>
</section>
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org