You are viewing a plain text version of this content. The canonical link for it is here.
Posted to java-commits@lucene.apache.org by gs...@apache.org on 2007/01/13 05:08:25 UTC
svn commit: r495834 [2/3] - in /lucene/java/trunk/contrib/benchmark: ./
conf/ src/java/org/apache/lucene/benchmark/byTask/
src/java/org/apache/lucene/benchmark/byTask/feeds/
src/java/org/apache/lucene/benchmark/byTask/programmatic/
src/java/org/apache/...
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/package.html
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/package.html?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/package.html (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/package.html Fri Jan 12 20:08:23 2007
@@ -0,0 +1,480 @@
+<HTML>
+<!--
+* Copyright 2005 The Apache Software Foundation
+*
+* Licensed under the Apache License, Version 2.0 (the "License");
+* you may not use this file except in compliance with the License.
+* You may obtain a copy of the License at
+*
+* http://www.apache.org/licenses/LICENSE-2.0
+*
+* Unless required by applicable law or agreed to in writing, software
+* distributed under the License is distributed on an "AS IS" BASIS,
+* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+* See the License for the specific language governing permissions and
+* limitations under the License.
+*
+-->
+<HEAD>
+ <TITLE>Benchmarking Lucene By Tasks</TITLE>
+</HEAD>
+<BODY>
+<DIV>
+Benchmarking Lucene By Tasks.
+<p>
+This package provides "task based" performance benchmarking of Lucene.
+One can use the predefined benchmarks, or create new ones.
+</p>
+<p>
+Contained packages:
+</p>
+
+<table border=1 cellpadding=4>
+ <tr>
+ <td><b>Package</b></td>
+ <td><b>Description</b></td>
+ </tr>
+ <tr>
+ <td><a href="stats/package-summary.html">stats</a></td>
+ <td>Statistics maintained when running benchmark tasks.</td>
+ </tr>
+ <tr>
+ <td><a href="tasks/package-summary.html">tasks</a></td>
+ <td>Benchmark tasks.</td>
+ </tr>
+ <tr>
+ <td><a href="feeds/package-summary.html">feeds</a></td>
+ <td>Sources foe benchmark inputs: documents and queries.</td>
+ </tr>
+ <tr>
+ <td><a href="utils/package-summary.html">utils</a></td>
+ <td>Utilities used for the benchmark, and for the reports.</td>
+ </tr>
+ <tr>
+ <td><a href="programmatic/package-summary.html">programmatic</a></td>
+ <td>Sample performance test written programatically.</td>
+ </tr>
+</table>
+
+<h2>Table Of Contents</h2>
+<p>
+ <ol>
+ <li><a href="#concept">Benchmarking By Tasks</a></li>
+ <li><a href="#usage">How to use</a></li>
+ <li><a href="#algorithm">Benchmark "algorithm"</a></li>
+ <li><a href="#tasks">Supported tasks/commands</a></li>
+ <li><a href="#properties">Benchmark properties</a></li>
+ <li><a href="#example">Example input algorithm and the result benchmark report.</a></li>
+ </ol>
+</p>
+<a name="concept"></a>
+<h2>Benchmarking By Tasks</h2>
+<p>
+Benchmark Lucene using task primitives.
+</p>
+
+<p>
+A benchmark is composed of some predefined tasks, allowing for creating an index, adding documents,
+optimizing, searching, generating reports, and more. A benchmark run takes an "algorithm" file
+that contains a description of the sequence of tasks making up the run, and some properties defining a few
+additional characteristics of the benchmark run.
+</p>
+
+<a name="usage"></a>
+<h2>How to use</h2>
+<p>
+Easiest way to run a benchmarks is using the predefined ant task:
+<ul>
+ <li>ant run-task
+ <br>- would run the <code>micro-standard.alg</code> "algorithm".
+ </li>
+ <li>ant run-task -Dtask.alg=conf/compound-penalty.alg
+ <br>- would run the <code>compound-penalty.alg</code> "algorithm".
+ </li>
+ <li>ant run-task -Dtask.alg=[full-path-to-your-alg-file]
+ <br>- would run the <code>your perf test</code> "algorithm".
+ </li>
+ <li>java org.apache.lucene.benchmark.byTask.programmatic.Sample
+ <br>- would run a performance test programmatically - without using an alg file.
+ This is less readable, and less convinient, but possible.
+ </li>
+</ul>
+</p>
+
+<p>
+You may find existing tasks sufficient for defining the benchmark <i>you</i> need,
+otherwise, you can extend the framework to meet your needs, as explained herein.
+</p>
+
+<p>
+Each benchmark run has a DocMaker and a QueryMaker. These two should usually match, so
+that "meaningful" queries are used for a certain collection.
+Properties defined at the header of the alg file define which "makers" should be used.
+You can also specify your own makers, implementing the DocMaker and QureyMaker interfaces.
+</p>
+
+<p>
+Benchmark .alg file contains the benchmark "algorithm". The syntax is described below.
+Within the algorithm, you can specify groups of commands, assign them names,
+specify commands that should be repeated,
+do commands in serial or in parallel,
+and also control the speed of "firing" the commands.
+</p>
+
+<p>
+This allows, for instance, to specify
+that an index should be opened for update,
+documents should be added to it one by one but not faster than 20 docs a minute,
+and, in parallel with this,
+some N queries should be searched against that index,
+again, no more than 2 queries a second.
+You can have the searches all share an index searcher,
+or have them each open its own searcher and close it afterwords.
+</p>
+
+<p>
+If the commands available for use in the algorithm do not meet your needs,
+you can add commands by adding a new task under
+org.apache.lucene.benchmark.byTask.tasks -
+you should extend the PerfTask abstract class.
+Make sure that your new task class name is suffixed by Task.
+Assume you added the class "WonderfulTask" - doing so also enables the
+command "Wonderful" to be used in the algorithm.
+</p>
+
+<a name="algorithm"></a>
+<h2>Benchmark "algorithm"</h2>
+
+<p>
+The following is an informal description of the supported syntax.
+</p>
+
+<ol>
+ <li>
+ <b>Measuring</b>: When a command is executed, statistics for the elapsed execution time and memory consumption are collected.
+ At any time, those statistics can be printed, using one of the available ReportTasks.
+ </li>
+ <li>
+ <b>Comments</b> start with '<font color="#FF0066">#</font>'.
+ </li>
+ <li>
+ <b>Serial</b> sequences are enclosed within '<font color="#FF0066">{ }</font>'.
+ </li>
+ <li>
+ <b>Parallel</b> sequences are enclosed within '<font color="#FF0066">[ ]</font>'
+ </li>
+ <li>
+ <b>Sequence naming:</b> To name a sequence, put '<font color="#FF0066">"name"</font>' just after '<font color="#FF0066">{</font>' or '<font color="#FF0066">[</font>'.
+ <br>Example - <font color="#FF0066">{ "ManyAdds" AddDoc } : 1000000</font> - would
+ name the sequence of 1M add docs "ManyAdds", and this name would later appear in statistic reports.
+ If you don't specify a name for a sequence, it is given one: you can see it as the
+ algorithm is printed just before benchmark execution starts.
+ </li>
+ <li>
+ <b>Repeating</b>:
+ To repeat sequence tasks N times, add '<font color="#FF0066">: N</font>' just after the
+ sequence closing tag - '<font color="#FF0066">}</font>' or '<font color="#FF0066">]</font>' or '<font color="#FF0066">></font>'.
+ <br>Example - <font color="#FF0066">[ AddDoc ] : 4</font> - would do 4 addDoc in parallel, spawning 4 threads at once.
+ <br>Example - <font color="#FF0066">[ AddDoc AddDoc ] : 4</font> - would do 8 addDoc in parallel, spawning 8 threads at once.
+ <br>Example - <font color="#FF0066">{ AddDoc } : 30</font> - would do addDoc 30 times in a row.
+ <br>Example - <font color="#FF0066">{ AddDoc AddDoc } : 30</font> - would do addDoc 60 times in a row.
+ </li>
+ <li>
+ <b>Command parameter</b>: a command can take a single parameter.
+ If the certain command does not support a parameter, or if the parameter is of the wrong type,
+ reading the algorithm will fail with an exception and the test would not start.
+ Currently only AddDoc supports a (numeric) parameter, which indicates the required size of added document.
+ If the DocMaker implementation used in the test does not support makeDoc(size), an exception would be thrown and the test would fail.
+ <br>Example - <font color="#FF0066">AddDoc(2000)</font> - would add a document of size 2000 (~bytes).
+ <br>See conf/task-sample.alg for how this can be used, for instance, to check which is faster, adding
+ many smaller documents, or few larger documents.
+ Next candidates for supporting a parameter may be the Search tasks, for controlling the qurey size.
+ </li>
+ <li>
+ <b>Statistic recording elimination</b>: - a sequence can also end with '<font color="#FF0066">></font>',
+ in which case child tasks would not store their statistics.
+ This can be useful to avoid exploding stats data, for adding say 1M docs.
+ <br>Example - <font color="#FF0066">{ "ManyAdds" AddDoc > : 1000000</font> -
+ would add million docs, measure that total, but not save stats for each addDoc.
+ <br>Notice that the granularity of System.currentTimeMillis() (which is used here) is system dependant,
+ and in some systems an operation that takes 5 ms to complete may show 0 ms latency time in performance measurements.
+ Therefore it is sometimes more accurate to look at the elapsed time of a larger sequence, as demonstrated here.
+ </li>
+ <li>
+ <b>Rate</b>:
+ To set a rate (ops/sec or ops/min) for a sequence, add '<font color="#FF0066">: N : R</font>' just after sequence closing tag.
+ This would specify repetition of N with rate of R operations/sec.
+ Use '<font color="#FF0066">R/sec</font>' or '<font color="#FF0066">R/min</font>'
+ to explicitely specify that the rate is per second or per minute.
+ The default is per second,
+ <br>Example - <font color="#FF0066">[ AddDoc ] : 400 : 3</font> - would do 400 addDoc in parallel, starting up to 3 threads per second.
+ <br>Example - <font color="#FF0066">{ AddDoc } : 100 : 200/min</font> - would do 100 addDoc serially,
+ waiting before starting next add, if otherwise rate would exceed 200 adds/min.
+ </li>
+ <li>
+ <b>Command names</b>: Each class "AnyNameTask" in the package org.apache.lucene.benchmark.byTask.tasks,
+ that extends PerfTask, is supported as command "AnyName" that can be
+ used in the benchmark "algorithm" description.
+ This allows to add new commands by just adding such classes.
+ </li>
+</ol>
+
+
+<a name="tasks"></a>
+<h2>Supported tasks/commands</h2>
+
+<p>
+Existing tasks can be divided into a few groups:
+regular index/search work tasks, report tasks, and control tasks.
+</p>
+
+<ol>
+
+ <li>
+ <b>Report tasks</b>: There are a few Report commands for generating reports.
+ Only task runs that were completed are reported.
+ (The 'Report tasks' themselves are not measured and not reported.)
+ <ul>
+ <li>
+ <font color="#FF0066">RepAll</font> - all (completed) task runs.
+ </li>
+ <li>
+ <font color="#FF0066">RepSumByName</font> - all statistics, aggregated by name. So, if AddDoc was executed 2000 times,
+ only 1 report line would be created for it, aggregating all those 2000 statistic records.
+ </li>
+ <li>
+ <font color="#FF0066">RepSelectByPref prefixWord</font> - all records for tasks whose name start with <font color="#FF0066">prefixWord</font>.
+ </li>
+ <li>
+ <font color="#FF0066">RepSumByPref prefixWord</font> - all records for tasks whose name start with <font color="#FF0066">prefixWord</font>,
+ aggregated by their full task name.
+ </li>
+ <li>
+ <font color="#FF0066">RepSumByNameRound</font> - all statistics, aggregated by name and by <font color="#FF0066">Round</font>.
+ So, if AddDoc was executed 2000 times in each of 3 <font color="#FF0066">rounds</font>, 3 report lines would be created for it,
+ aggregating all those 2000 statistic records in each round. See more about rounds in the <font color="#FF0066">NewRound</font> command description below.
+ </li>
+ <li>
+ <font color="#FF0066">RepSumByPrefRound prefixWord</font> - similar to <font color="#FF0066">RepSumByNameRound</font>,
+ just that only tasks whose name starts with <font color="#FF0066">prefixWord</font> are included.
+ </li>
+ </ul>
+ If needed, additional reports can be added by extending the abstract class ReportTask, and by
+ manipulating the statistics data in Points and TaskStats.
+ </li>
+
+ <li><b>Control tasks</b>: Few of the tasks control the benchmark algorithm all over:
+ <ul>
+ <li>
+ <font color="#FF0066">ClearStats</font> - clears the entire statistics.
+ Further reports would only include task runs that would start after this call.
+ </li>
+ <li>
+ <font color="#FF0066">NewRound</font> - virtually start a new round of performance test.
+ Although this command can be placed anywhere, it mostly makes sense at the end of an outermost sequence.
+ <br>This increments a global "round counter". All task runs that would start now would
+ record the new, updated round counter as their round number. This would appear in reports.
+ In particular, see <font color="#FF0066">RepSumByNameRound</font> above.
+ <br>An additional effect of NewRound, is that numeric and boolean properties defined in the
+ .properties file as a sequence of values, e.g. <font color="#FF0066">merge.factor=mrg:10:100:10:100</font> would
+ increment (cyclic) to the next value.
+ Note: this would also be reflected in the reports, in this case under a column that would be named "mrg".
+ </li>
+ <li>
+ <font color="#FF0066">ResetInputs</font> - DocMaker and the various QueryMakers
+ would reset their counters to start.
+ The way these Maker interfaces work, each call for makeDocument()
+ or makeQuery() creates the next document or query
+ that it "knows" to create.
+ If that pool is "exhausted", the "maker" start over again. The resetInpus command
+ therefore allows to make the rounds comparable.
+ It is therefore useful to invoke ResetInputs together with NewRound.
+ </li>
+ <li>
+ <font color="#FF0066">ResetSystemErase</font> - reset all index and input data and call gc.
+ Does NOT reset statistics. This contains ResetInputs.
+ All writers/readers are nullified, deleted, closed.
+ Index is erased.
+ Directory is erased.
+ You would have to call CreateIndex once this was called...
+ </li>
+ <li>
+ <font color="#FF0066">ResetSystemSoft</font> - reset all index and input data and call gc.
+ Does NOT reset statistics. This contains ResetInputs.
+ All writers/readers are nullified, closed.
+ Index is NOT erased.
+ Directory is NOT erased.
+ This is useful for testing performance on an existing index, for instance if the construction of a large index
+ took a very long time and now you would to test its search or update performance.
+ </li>
+ </ul>
+ </li>
+
+ <li>
+ Other existing tasks are quite straightforward and would just be briefly described here.
+ <ul>
+ <li>
+ <font color="#FF0066">CreateIndex</font> and <font color="#FF0066">OpenIndex</font> both leave the index open for later update operations.
+ <font color="#FF0066">CloseIndex</font> would close it.
+ </li>
+ <li>
+ <font color="#FF0066">OpenReader</font>, similarly, would leave an index reader open for later search operations.
+ But this have further semantics.
+ If a Read operation is performed, and an open reader exists, it would be used.
+ Otherwise, the read operation would open its own reader and close it when the read operation is done.
+ This allows testing various scenarios - sharing a reader, searching with "cold" reader, with "warmed" reader, etc.
+ The read operations affected by this are: <font color="#FF0066">Warm</font>,
+ <font color="#FF0066">Search</font>, <font color="#FF0066">SearchTrav</font> (search and traverse),
+ and <font color="#FF0066">SearchTravRet</font> (search and traverse and retrieve).
+ Notice that each of the 3 search task types maintains its own queryMaker instance.
+ </li>
+ </ul
+ </li>
+ </ol>
+
+<a name="properties"></a>
+<h2>Benchmark properties</h2>
+
+<p>
+Properties are read from the header of the .alg file, and
+define several parameters of the performance test.
+As mentioned above for the <font color="#FF0066">NewRound</font> task,
+numeric and boolean properties that are defined as a sequence
+of values, e.g. <font color="#FF0066">merge.factor=mrg:10:100:10:100</font>
+would increment (cyclic) to the next value, when NewRound is called, and would also
+appear as a named column in the reports (column name would be "mrg" in this example).
+</p>
+
+<p>
+Some of the currently defined properties are:
+</p>
+
+<ol>
+ <li>
+ <font color="#FF0066">analyzer</font> - full class name for the analyzer to use.
+ Same analyzer would be used in the entire test.
+ </li>
+
+ <li>
+ <font color="#FF0066">directory</font> - valid values are FSDirectory and RAMDirectory.
+ This tells which directory to use for the performance test.
+ </li>
+
+ <li>
+ <b>Index work parameters</b>:
+ Multi int/boolean values would be iterated with calls to NewRound.
+ There would be also added as columns in the reports, first string in the
+ sequence is the column name.
+ (Make sure it is no shorter than any value in the sequence).
+ <ul>
+ <li><font color="#FF0066">max.buffered</font>
+ <br>Example: buffered=buf.10.10.100.100 -
+ this would define using maxBufferedDocs of 10 in iterations 0 and 1,
+ and 100 in iterations 2 and 3.
+ </li>
+ <li>
+ <font color="#FF0066">merge.factor</font> - which
+ merge factor to use.
+ </li>
+ <li>
+ <font color="#FF0066">compound</font> - whether the index is
+ using the compound format or not. Valid values are "true" and "false".
+ </li>
+ </ul>
+</ol>
+
+<p>
+For additional defined properties see the *.alg files under conf.
+</p>
+
+<a name="example"></a>
+<h2>Example input algorithm and the result benchmark report</h2>
+<p>
+The following example is in conf/sample.alg:
+<pre>
+<font color="#003333"># --------------------------------------------------------
+#
+# Sample: what is the effect of doc size on indexing time?
+#
+# There are two parts in this test:
+# - PopulateShort adds 2N documents of length L
+# - PopulateLong adds N documents of length 2L
+# Which one would be faster?
+# The comparison is done twice.
+#
+# --------------------------------------------------------
+
+<font color="#990066"># -------------------------------------------------------------------------------------
+# multi val params are iterated by NewRound's, added to reports, start with column name.
+merge.factor=mrg:10:20
+max.buffered=buf:100:1000
+compound=true
+
+analyzer=org.apache.lucene.analysis.standard.StandardAnalyzer
+directory=FSDirectory
+
+doc.stored=true
+doc.tokenized=true
+doc.term.vector=false
+doc.add.log.step=500
+
+docs.dir=reuters-out
+
+doc.maker=org.apache.lucene.benchmark.byTask.feeds.SimpleDocMaker
+
+query.maker=org.apache.lucene.benchmark.byTask.feeds.SimpleQueryMaker
+
+# task at this depth or less would print when they start
+task.max.depth.log=2
+
+log.queries=false
+# -------------------------------------------------------------------------------------</font>
+<font color="#3300FF">{
+
+ { "PopulateShort"
+ CreateIndex
+ { AddDoc(4000) > : 20000
+ Optimize
+ CloseIndex
+ >
+
+ ResetSystemErase
+
+ { "PopulateLong"
+ CreateIndex
+ { AddDoc(8000) > : 10000
+ Optimize
+ CloseIndex
+ >
+
+ ResetSystemErase
+
+ NewRound
+
+} : 2
+
+RepSumByName
+RepSelectByPref Populate
+</font>
+</pre>
+</p>
+
+<p>
+The command line for running this sample:
+<br><code>ant run-task -Dtask.alg=conf/sample.alg</code>
+</p>
+
+<p>
+The output report from running this test contains the following:
+<pre>
+Operation round mrg buf runCnt recsPerRun rec/s elapsedSec avgUsedMem avgTotalMem
+PopulateShort 0 10 100 1 20003 119.6 167.26 12,959,120 14,241,792
+PopulateLong - - 0 10 100 - - 1 - - 10003 - - - 74.3 - - 134.57 - 17,085,208 - 20,635,648
+PopulateShort 1 20 1000 1 20003 143.5 139.39 63,982,040 94,756,864
+PopulateLong - - 1 20 1000 - - 1 - - 10003 - - - 77.0 - - 129.92 - 87,309,608 - 100,831,232
+</pre>
+</p>
+</DIV>
+<DIV> </DIV>
+</BODY>
+</HTML>
\ No newline at end of file
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/programmatic/Sample.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/programmatic/Sample.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/programmatic/Sample.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/programmatic/Sample.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,81 @@
+package org.apache.lucene.benchmark.byTask.programmatic;
+
+import java.io.IOException;
+import java.util.Properties;
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.benchmark.byTask.tasks.AddDocTask;
+import org.apache.lucene.benchmark.byTask.tasks.CloseIndexTask;
+import org.apache.lucene.benchmark.byTask.tasks.CreateIndexTask;
+import org.apache.lucene.benchmark.byTask.tasks.RepSumByNameTask;
+import org.apache.lucene.benchmark.byTask.tasks.TaskSequence;
+import org.apache.lucene.benchmark.byTask.utils.Config;
+
+/**
+ * Sample performance test written programatically - no algorithm file is needed here.
+ */
+public class Sample {
+
+ /**
+ * @param args
+ * @throws Exception
+ * @throws IOException
+ */
+ public static void main(String[] args) throws Exception {
+ Properties p = initProps();
+ Config conf = new Config(p);
+ PerfRunData runData = new PerfRunData(conf);
+
+ // 1. top sequence
+ TaskSequence top = new TaskSequence(runData,null,null,false); // top level, not parralel
+
+ // 2. task to create the index
+ CreateIndexTask create = new CreateIndexTask(runData);
+ top.addTask(create);
+
+ // 3. task seq to add 500 docs (order matters - top to bottom - add seq to top, only then add to seq)
+ TaskSequence seq1 = new TaskSequence(runData,"AddDocs",top,false);
+ seq1.setRepetitions(500);
+ seq1.setNoChildReport();
+ top.addTask(seq1);
+
+ // 4. task to add the doc
+ AddDocTask addDoc = new AddDocTask(runData);
+ //addDoc.setParams("1200"); // doc size limit if supported
+ seq1.addTask(addDoc); // order matters 9see comment above)
+
+ // 5. task to close the index
+ CloseIndexTask close = new CloseIndexTask(runData);
+ top.addTask(close);
+
+ // task to report
+ RepSumByNameTask rep = new RepSumByNameTask(runData);
+ top.addTask(rep);
+ // execute
+ top.doLogic();
+ }
+
+ // Sample programmatic settings. Could also read from file.
+ private static Properties initProps() {
+ Properties p = new Properties();
+ p.setProperty ( "task.max.depth.log" , "3" );
+ p.setProperty ( "max.buffered" , "buf:10:10:100:100:10:10:100:100" );
+ p.setProperty ( "doc.maker" , "org.apache.lucene.benchmark.byTask.feeds.ReutersDocMaker" );
+ p.setProperty ( "doc.add.log.step" , "2000" );
+ p.setProperty ( "doc.delete.log.step" , "2000" );
+ p.setProperty ( "doc.delete.step" , "8" );
+ p.setProperty ( "analyzer" , "org.apache.lucene.analysis.standard.StandardAnalyzer" );
+ p.setProperty ( "doc.term.vector" , "false" );
+ p.setProperty ( "directory" , "FSDirectory" );
+ p.setProperty ( "query.maker" , "org.apache.lucene.benchmark.byTask.feeds.ReutersQueryMaker" );
+ p.setProperty ( "doc.stored" , "true" );
+ p.setProperty ( "docs.dir" , "reuters-out" );
+ p.setProperty ( "compound" , "cmpnd:true:true:true:true:false:false:false:false" );
+ p.setProperty ( "doc.tokenized" , "true" );
+ p.setProperty ( "merge.factor" , "mrg:10:100:10:100:10:100:10:100" );
+ return p;
+ }
+
+
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/programmatic/package.html
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/programmatic/package.html?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/programmatic/package.html (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/programmatic/package.html Fri Jan 12 20:08:23 2007
@@ -0,0 +1,5 @@
+<html>
+<body>
+Sample performance test written programatically - no algorithm file is needed here.
+</body>
+</html>
\ No newline at end of file
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/Points.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/Points.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/Points.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/Points.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,343 @@
+package org.apache.lucene.benchmark.byTask.stats;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.util.ArrayList;
+import java.util.Collection;
+import java.util.Iterator;
+import java.util.LinkedHashMap;
+
+import org.apache.lucene.benchmark.byTask.tasks.PerfTask;
+import org.apache.lucene.benchmark.byTask.utils.Config;
+import org.apache.lucene.benchmark.byTask.utils.Format;
+
+
+/**
+ * Test run data points collected as the test proceeds.
+ */
+public class Points {
+
+ private Config config;
+
+ private static final String newline = System.getProperty("line.separator");
+
+ // stat points ordered by their start time.
+ // for now we collect points as TaskStats objects.
+ // later might optimize to collect only native data.
+ private ArrayList points = new ArrayList();
+
+ private int nextTaskRunNum = 0;
+
+ /**
+ * Get a textual summary of the benchmark results, average from all test runs.
+ */
+ static final String OP = "Operation ";
+ static final String ROUND = " round";
+ static final String RUNCNT = " runCnt";
+ static final String RECCNT = " recsPerRun";
+ static final String RECSEC = " rec/s";
+ static final String ELAPSED = " elapsedSec";
+ static final String USEDMEM = " avgUsedMem";
+ static final String TOTMEM = " avgTotalMem";
+ static final String COLS[] = {
+ RUNCNT,
+ RECCNT,
+ RECSEC,
+ ELAPSED,
+ USEDMEM,
+ TOTMEM
+ };
+
+ /**
+ * Create a Points statistics object.
+ */
+ public Points (Config config) {
+ this.config = config;
+ }
+
+ private String tableTitle (String longestOp) {
+ StringBuffer sb = new StringBuffer();
+ sb.append(Format.format(OP,longestOp));
+ sb.append(ROUND);
+ sb.append(config.getColsNamesForValsByRound());
+ for (int i = 0; i < COLS.length; i++) {
+ sb.append(COLS[i]);
+ }
+ return sb.toString();
+ }
+
+ /**
+ * Report detailed statistics as a string
+ * @return the report
+ */
+ public Report reportAll() {
+ String longestOp = longestOp(points);
+ boolean first = true;
+ StringBuffer sb = new StringBuffer();
+ sb.append(tableTitle(longestOp));
+ sb.append(newline);
+ int reported = 0;
+ for (Iterator it = points.iterator(); it.hasNext();) {
+ TaskStats stat = (TaskStats) it.next();
+ if (stat.getElapsed()>=0) { // consider only tasks that ended
+ if (!first) {
+ sb.append(newline);
+ }
+ first = false;
+ String line = taskReportLine(longestOp, stat);
+ reported++;
+ if (points.size()>2&& reported%2==0) {
+ line = line.replaceAll(" "," - ");
+ }
+ sb.append(line);
+ }
+ }
+ String reptxt = (reported==0 ? "No Matching Entries Were Found!" : sb.toString());
+ return new Report(reptxt,reported,reported,points.size());
+ }
+
+ /**
+ * Report statistics as a string, aggregate for tasks named the same.
+ * @return the report
+ */
+ public Report reportSumByName() {
+ // aggregate by task name
+ int reported = 0;
+ LinkedHashMap p2 = new LinkedHashMap();
+ for (Iterator it = points.iterator(); it.hasNext();) {
+ TaskStats stat1 = (TaskStats) it.next();
+ if (stat1.getElapsed()>=0) { // consider only tasks that ended
+ reported++;
+ String name = stat1.getTask().getName();
+ TaskStats stat2 = (TaskStats) p2.get(name);
+ if (stat2 == null) {
+ try {
+ stat2 = (TaskStats) stat1.clone();
+ } catch (CloneNotSupportedException e) {
+ throw new RuntimeException(e);
+ }
+ p2.put(name,stat2);
+ } else {
+ stat2.add(stat1);
+ }
+ }
+ }
+ // now generate report from secondary list p2
+ return genReportFromList(reported, p2);
+ }
+
+ /**
+ * Report statistics as a string, aggregate for tasks named the same, and from the same round.
+ * @return the report
+ */
+ public Report reportSumByNameRound() {
+ // aggregate by task name and round
+ LinkedHashMap p2 = new LinkedHashMap();
+ int reported = 0;
+ for (Iterator it = points.iterator(); it.hasNext();) {
+ TaskStats stat1 = (TaskStats) it.next();
+ if (stat1.getElapsed()>=0) { // consider only tasks that ended
+ reported++;
+ String name = stat1.getTask().getName();
+ String rname = stat1.getRound()+"."+name; // group by round
+ TaskStats stat2 = (TaskStats) p2.get(rname);
+ if (stat2 == null) {
+ try {
+ stat2 = (TaskStats) stat1.clone();
+ } catch (CloneNotSupportedException e) {
+ throw new RuntimeException(e);
+ }
+ p2.put(rname,stat2);
+ } else {
+ stat2.add(stat1);
+ }
+ }
+ }
+ // now generate report from secondary list p2
+ return genReportFromList(reported, p2);
+ }
+
+ private String longestOp(Collection c) {
+ String longest = OP;
+ for (Iterator it = c.iterator(); it.hasNext();) {
+ TaskStats stat = (TaskStats) it.next();
+ if (stat.getElapsed()>=0) { // consider only tasks that ended
+ String name = stat.getTask().getName();
+ if (name.length() > longest.length()) {
+ longest = name;
+ }
+ }
+ }
+ return longest;
+ }
+
+ private String taskReportLine(String longestOp, TaskStats stat) {
+ PerfTask task = stat.getTask();
+ StringBuffer sb = new StringBuffer();
+ sb.append(Format.format(task.getName(), longestOp));
+ String round = (stat.getRound()>=0 ? ""+stat.getRound() : "-");
+ sb.append(Format.formatPaddLeft(round, ROUND));
+ sb.append(config.getColsValuesForValsByRound(stat.getRound()));
+ sb.append(Format.format(stat.getNumRuns(), RUNCNT));
+ sb.append(Format.format(stat.getCount() / stat.getNumRuns(), RECCNT));
+ long elapsed = (stat.getElapsed()>0 ? stat.getElapsed() : 1); // assume at least 1ms
+ sb.append(Format.format(1,(float) (stat.getCount() * 1000.0 / elapsed), RECSEC));
+ sb.append(Format.format(2, (float) stat.getElapsed() / 1000, ELAPSED));
+ sb.append(Format.format(0, (float) stat.getMaxUsedMem() / stat.getNumRuns(), USEDMEM));
+ sb.append(Format.format(0, (float) stat.getMaxTotMem() / stat.getNumRuns(), TOTMEM));
+ return sb.toString();
+ }
+
+ public Report reportSumByPrefix(String prefix) {
+ // aggregate by task name
+ int reported = 0;
+ LinkedHashMap p2 = new LinkedHashMap();
+ for (Iterator it = points.iterator(); it.hasNext();) {
+ TaskStats stat1 = (TaskStats) it.next();
+ if (stat1.getElapsed()>=0 && stat1.getTask().getName().startsWith(prefix)) { // only ended tasks with proper name
+ reported++;
+ String name = stat1.getTask().getName();
+ TaskStats stat2 = (TaskStats) p2.get(name);
+ if (stat2 == null) {
+ try {
+ stat2 = (TaskStats) stat1.clone();
+ } catch (CloneNotSupportedException e) {
+ throw new RuntimeException(e);
+ }
+ p2.put(name,stat2);
+ } else {
+ stat2.add(stat1);
+ }
+ }
+ }
+ // now generate report from secondary list p2
+ return genReportFromList(reported, p2);
+ }
+
+ public Report reportSumByPrefixRound(String prefix) {
+ // aggregate by task name and by round
+ int reported = 0;
+ LinkedHashMap p2 = new LinkedHashMap();
+ for (Iterator it = points.iterator(); it.hasNext();) {
+ TaskStats stat1 = (TaskStats) it.next();
+ if (stat1.getElapsed()>=0 && stat1.getTask().getName().startsWith(prefix)) { // only ended tasks with proper name
+ reported++;
+ String name = stat1.getTask().getName();
+ String rname = stat1.getRound()+"."+name; // group by round
+ TaskStats stat2 = (TaskStats) p2.get(rname);
+ if (stat2 == null) {
+ try {
+ stat2 = (TaskStats) stat1.clone();
+ } catch (CloneNotSupportedException e) {
+ throw new RuntimeException(e);
+ }
+ p2.put(rname,stat2);
+ } else {
+ stat2.add(stat1);
+ }
+ }
+ }
+ // now generate report from secondary list p2
+ return genReportFromList(reported, p2);
+ }
+
+ private Report genReportFromList(int reported, LinkedHashMap p2) {
+ String longetOp = longestOp(p2.values());
+ boolean first = true;
+ StringBuffer sb = new StringBuffer();
+ sb.append(tableTitle(longetOp));
+ sb.append(newline);
+ int lineNum = 0;
+ for (Iterator it = p2.values().iterator(); it.hasNext();) {
+ TaskStats stat = (TaskStats) it.next();
+ if (!first) {
+ sb.append(newline);
+ }
+ first = false;
+ String line = taskReportLine(longetOp,stat);
+ lineNum++;
+ if (p2.size()>2&& lineNum%2==0) {
+ line = line.replaceAll(" "," - ");
+ }
+ sb.append(line);
+ }
+ String reptxt = (reported==0 ? "No Matching Entries Were Found!" : sb.toString());
+ return new Report(reptxt,p2.size(),reported,points.size());
+ }
+
+ public Report reportSelectByPrefix(String prefix) {
+ String longestOp = longestOp(points);
+ boolean first = true;
+ StringBuffer sb = new StringBuffer();
+ sb.append(tableTitle(longestOp));
+ sb.append(newline);
+ int reported = 0;
+ for (Iterator it = points.iterator(); it.hasNext();) {
+ TaskStats stat = (TaskStats) it.next();
+ if (stat.getElapsed()>=0 && stat.getTask().getName().startsWith(prefix)) { // only ended tasks with proper name
+ reported++;
+ if (!first) {
+ sb.append(newline);
+ }
+ first = false;
+ String line = taskReportLine(longestOp,stat);
+ if (points.size()>2&& reported%2==0) {
+ line = line.replaceAll(" "," - ");
+ }
+ sb.append(line);
+ }
+ }
+ String reptxt = (reported==0 ? "No Matching Entries Were Found!" : sb.toString());
+ return new Report(reptxt,reported,reported, points.size());
+ }
+
+ /**
+ * Mark that a task is starting.
+ * Create a task stats for it and store it as a point.
+ * @param task the starting task.
+ * @return the new task stats created for the starting task.
+ */
+ public synchronized TaskStats markTaskStart (PerfTask task, int round) {
+ TaskStats stats = new TaskStats(task, nextTaskRunNum(), round);
+ points.add(stats);
+ return stats;
+ }
+
+ // return next task num
+ private synchronized int nextTaskRunNum() {
+ return nextTaskRunNum++;
+ }
+
+ /**
+ * mark the end of a task
+ */
+ public synchronized void markTaskEnd (TaskStats stats, int count) {
+ int numParallelTasks = nextTaskRunNum - 1 - stats.getTaskRunNum();
+ // note: if the stats were cleared, might be that this stats object is
+ // no longer in points, but this is just ok.
+ stats.markEnd(numParallelTasks, count);
+ }
+
+ /**
+ * Clear all data, prepare for more tests.
+ */
+ public void clearData() {
+ points.clear();
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/Report.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/Report.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/Report.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/Report.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,64 @@
+package org.apache.lucene.benchmark.byTask.stats;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+/**
+ * Textual report of current statistics.
+ */
+public class Report {
+
+ private String text;
+ private int size;
+ private int outOf;
+ private int reported;
+
+ Report (String text, int size, int reported, int outOf) {
+ this.text = text;
+ this.size = size;
+ this.reported = reported;
+ this.outOf = outOf;
+ }
+
+ /**
+ * Returns total number of stats points when this report was created.
+ */
+ public int getOutOf() {
+ return outOf;
+ }
+
+ /**
+ * Returns number of lines in the reoprt.
+ */
+ public int getSize() {
+ return size;
+ }
+
+ /**
+ * Returns the report text.
+ */
+ public String getText() {
+ return text;
+ }
+
+ /**
+ * Returns number of stats points represented in this report.
+ */
+ public int getReported() {
+ return reported;
+ }
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/TaskStats.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/TaskStats.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/TaskStats.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/TaskStats.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,192 @@
+package org.apache.lucene.benchmark.byTask.stats;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import org.apache.lucene.benchmark.byTask.tasks.PerfTask;
+
+/**
+ * Statistics for a task run.
+ * <br>The same task can run more than once, but, if that task records statistics,
+ * each run would create its own TaskStats.
+ */
+public class TaskStats implements Cloneable {
+
+ /** task for which data was collected */
+ private PerfTask task;
+
+ /** round in which task run started */
+ private int round;
+
+ /** task start time */
+ private long start;
+
+ /** task elapsed time. elapsed >= 0 indicates run completion! */
+ private long elapsed = -1;
+
+ /** max tot mem during task */
+ private long maxTotMem;
+
+ /** max used mem during task */
+ private long maxUsedMem;
+
+ /** serial run number of this task run in the perf run */
+ private int taskRunNum;
+
+ /** number of other tasks that started to run while this task was still running */
+ private int numParallelTasks;
+
+ /** number of work items done by this task.
+ * For indexing that can be number of docs added.
+ * For warming that can be number of scanned items, etc.
+ * For repeating tasks, this is a sum over repetitions.
+ */
+ private int count;
+
+ /** Number of similar tasks aggregated into this record.
+ * Used when summing up on few runs/instances of similar tasks.
+ */
+ private int numRuns = 1;
+
+ /**
+ * Create a run data for a task that is starting now.
+ * To be called from Points.
+ */
+ TaskStats (PerfTask task, int taskRunNum, int round) {
+ this.task = task;
+ this.taskRunNum = taskRunNum;
+ this.round = round;
+ maxTotMem = Runtime.getRuntime().totalMemory();
+ maxUsedMem = maxTotMem - Runtime.getRuntime().freeMemory();
+ start = System.currentTimeMillis();
+ }
+
+ /**
+ * mark the end of a task
+ */
+ void markEnd (int numParallelTasks, int count) {
+ elapsed = System.currentTimeMillis() - start;
+ long totMem = Runtime.getRuntime().totalMemory();
+ if (totMem > maxTotMem) {
+ maxTotMem = totMem;
+ }
+ long usedMem = totMem - Runtime.getRuntime().freeMemory();
+ if (usedMem > maxUsedMem) {
+ maxUsedMem = usedMem;
+ }
+ this.numParallelTasks = numParallelTasks;
+ this.count = count;
+ }
+
+ /**
+ * @return the taskRunNum.
+ */
+ public int getTaskRunNum() {
+ return taskRunNum;
+ }
+
+ /* (non-Javadoc)
+ * @see java.lang.Object#toString()
+ */
+ public String toString() {
+ StringBuffer res = new StringBuffer(task.getName());
+ res.append(" ");
+ res.append(count);
+ res.append(" ");
+ res.append(elapsed);
+ return res.toString();
+ }
+
+ /**
+ * @return Returns the count.
+ */
+ public int getCount() {
+ return count;
+ }
+
+ /**
+ * @return elapsed time.
+ */
+ public long getElapsed() {
+ return elapsed;
+ }
+
+ /**
+ * @return Returns the maxTotMem.
+ */
+ public long getMaxTotMem() {
+ return maxTotMem;
+ }
+
+ /**
+ * @return Returns the maxUsedMem.
+ */
+ public long getMaxUsedMem() {
+ return maxUsedMem;
+ }
+
+ /**
+ * @return Returns the numParallelTasks.
+ */
+ public int getNumParallelTasks() {
+ return numParallelTasks;
+ }
+
+ /**
+ * @return Returns the task.
+ */
+ public PerfTask getTask() {
+ return task;
+ }
+
+ /**
+ * @return Returns the numRuns.
+ */
+ public int getNumRuns() {
+ return numRuns;
+ }
+
+ /**
+ * Add data from another stat, for aggregation
+ * @param stat2 the added stat data.
+ */
+ public void add(TaskStats stat2) {
+ numRuns += stat2.getNumRuns();
+ elapsed += stat2.getElapsed();
+ maxTotMem += stat2.getMaxTotMem();
+ maxUsedMem += stat2.getMaxUsedMem();
+ count += stat2.getCount();
+ if (round != stat2.round) {
+ round = -1; // no meaning if agregating tasks of different ruond.
+ }
+ }
+
+ /* (non-Javadoc)
+ * @see java.lang.Object#clone()
+ */
+ protected Object clone() throws CloneNotSupportedException {
+ return super.clone();
+ }
+
+ /**
+ * @return the round number.
+ */
+ int getRound() {
+ return round;
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/package.html
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/package.html?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/package.html (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/stats/package.html Fri Jan 12 20:08:23 2007
@@ -0,0 +1,5 @@
+<html>
+<body>
+ Statistics maintained when running benchmark tasks.
+</body>
+</html>
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/AddDocTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/AddDocTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/AddDocTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/AddDocTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,88 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.benchmark.byTask.feeds.DocMaker;
+import org.apache.lucene.document.Document;
+
+
+/**
+ * Add a document, optionally with of a cetrain size.
+ * Other side effects: none.
+ */
+public class AddDocTask extends PerfTask {
+
+ public AddDocTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ private static int logStep = -1;
+ private int docSize = 0;
+
+ // volatile data passed between setup(), doLogic(), tearDown().
+ private Document doc = null;
+
+ /*
+ * (non-Javadoc)
+ * @see PerfTask#setup()
+ */
+ public void setup() throws Exception {
+ super.setup();
+ DocMaker docMaker = getRunData().getDocMaker();
+ if (docSize > 0) {
+ doc = docMaker.makeDocument(docSize);
+ } else {
+ doc = docMaker.makeDocument();
+ }
+ }
+
+ /* (non-Javadoc)
+ * @see PerfTask#tearDown()
+ */
+ public void tearDown() throws Exception {
+ DocMaker docMaker = getRunData().getDocMaker();
+ log(docMaker.getCount());
+ doc = null;
+ super.tearDown();
+ }
+
+ public int doLogic() throws Exception {
+ getRunData().getIndexWriter().addDocument(doc);
+ return 1;
+ }
+
+ private void log (int count) {
+ if (logStep<0) {
+ // avoid sync although race possible here
+ logStep = getRunData().getConfig().get("doc.add.log.step",500);
+ }
+ if (logStep>0 && (count%logStep)==0) {
+ System.out.println("--> processed "+count+" docs");
+ }
+ }
+
+ /**
+ * Set the params (docSize only)
+ * @param params docSize, or 0 for no limit.
+ */
+ public void setParams(String params) {
+ super.setParams(params);
+ docSize = (int) Float.parseFloat(params);
+ }
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/ClearStatsTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/ClearStatsTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/ClearStatsTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/ClearStatsTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,44 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+/**
+ * Clear statistics data.
+ * Other side effects: None.
+ */
+public class ClearStatsTask extends PerfTask {
+
+ public ClearStatsTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws Exception {
+ getRunData().getPoints().clearData();
+ return 0;
+ }
+
+ /* (non-Javadoc)
+ * @see PerfTask#shouldNotRecordStats()
+ */
+ protected boolean shouldNotRecordStats() {
+ return true;
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CloseIndexTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CloseIndexTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CloseIndexTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CloseIndexTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,44 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.io.IOException;
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.index.IndexWriter;
+
+/**
+ * Close index writer.
+ * Other side effects: index writer object in perfRunData is nullified.
+ */
+public class CloseIndexTask extends PerfTask {
+
+ public CloseIndexTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws IOException {
+ IndexWriter iw = getRunData().getIndexWriter();
+ if (iw!=null) {
+ iw.close();
+ }
+ getRunData().setIndexWriter(null);
+ return 1;
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CloseReaderTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CloseReaderTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CloseReaderTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CloseReaderTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,45 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.io.IOException;
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.index.IndexReader;
+
+/**
+ * Close index reader.
+ * Other side effects: index reader in perfRunData is nullified.
+ * This would cause read related tasks to reopen their own reader.
+ */
+public class CloseReaderTask extends PerfTask {
+
+ public CloseReaderTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws IOException {
+ IndexReader reader= getRunData().getIndexReader();
+ if (reader!=null) {
+ reader.close();
+ }
+ getRunData().setIndexReader(null);
+ return 1;
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CreateIndexTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CreateIndexTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CreateIndexTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/CreateIndexTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,59 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.io.IOException;
+
+import org.apache.lucene.analysis.Analyzer;
+import org.apache.lucene.index.IndexWriter;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.benchmark.byTask.utils.Config;
+
+
+/**
+ * Create an index.
+ * Other side effects: index writer object in perfRunData is set.
+ */
+public class CreateIndexTask extends PerfTask {
+
+ public CreateIndexTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws IOException {
+ Directory dir = getRunData().getDirectory();
+ Analyzer analyzer = getRunData().getAnalyzer();
+
+ IndexWriter iw = new IndexWriter(dir, analyzer, true);
+
+ Config config = getRunData().getConfig();
+
+ boolean cmpnd = config.get("compound",true);
+ int mrgf = config.get("merge.factor",10);
+ int mxbf = config.get("max.buffered",10);
+
+ iw.setUseCompoundFile(cmpnd);
+ iw.setMergeFactor(mrgf);
+ iw.setMaxBufferedDocs(mxbf);
+
+ getRunData().setIndexWriter(iw);
+ return 1;
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/DeleteDocTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/DeleteDocTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/DeleteDocTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/DeleteDocTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,86 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+/**
+ * Delete a document by docid.
+ * Other side effects: none.
+ */
+public class DeleteDocTask extends PerfTask {
+
+ public DeleteDocTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ private static int logStep = -1;
+ private static int deleteStep = -1;
+ private static int numDeleted = 0;
+ private static int lastDeleted = -1;
+
+ private int docid = -1;
+ private boolean byStep = true;
+
+ public int doLogic() throws Exception {
+ getRunData().getIndexReader().deleteDocument(docid);
+ lastDeleted = docid;
+ return 1; // one work item done here
+ }
+
+ /* (non-Javadoc)
+ * @see org.apache.lucene.benchmark.byTask.tasks.PerfTask#setup()
+ */
+ public void setup() throws Exception {
+ super.setup();
+ // one time static initializations
+ if (logStep<0) {
+ logStep = getRunData().getConfig().get("doc.delete.log.step",500);
+ }
+ if (deleteStep<0) {
+ deleteStep = getRunData().getConfig().get("doc.delete.step",8);
+ }
+ // set the docid to be deleted
+ docid = (byStep ? lastDeleted + deleteStep : docid);
+ }
+
+ /* (non-Javadoc)
+ * @see PerfTask#tearDown()
+ */
+ public void tearDown() throws Exception {
+ log(++numDeleted);
+ super.tearDown();
+ }
+
+ private void log (int count) {
+ if (logStep>0 && (count%logStep)==0) {
+ System.out.println("--> processed "+count+" docs, last deleted: "+lastDeleted);
+ }
+ }
+
+ /**
+ * Set the params (docid only)
+ * @param params docid to delete, or -1 for deleting by delete gap settings.
+ */
+ public void setParams(String params) {
+ super.setParams(params);
+ docid = (int) Float.parseFloat(params);
+ byStep = (docid < 0);
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/NewRoundTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/NewRoundTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/NewRoundTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/NewRoundTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,44 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+
+/**
+ * Increment the counter for properties maintained by Round Number.
+ * Other side effects: if there are props by round number, log value change.
+ */
+public class NewRoundTask extends PerfTask {
+
+ public NewRoundTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws Exception {
+ getRunData().getConfig().newRound();
+ return 0;
+ }
+
+ /* (non-Javadoc)
+ * @see PerfTask#shouldNotRecordStats()
+ */
+ protected boolean shouldNotRecordStats() {
+ return true;
+ }
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OpenIndexTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OpenIndexTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OpenIndexTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OpenIndexTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,59 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.io.IOException;
+
+import org.apache.lucene.analysis.Analyzer;
+import org.apache.lucene.index.IndexWriter;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.benchmark.byTask.utils.Config;
+
+
+/**
+ * Open an index writer.
+ * Other side effects: index writer object in perfRunData is set.
+ */
+public class OpenIndexTask extends PerfTask {
+
+ public OpenIndexTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws IOException {
+ Directory dir = getRunData().getDirectory();
+ Analyzer analyzer = getRunData().getAnalyzer();
+ IndexWriter writer = new IndexWriter(dir, analyzer, false);
+
+ Config config = getRunData().getConfig();
+
+ boolean cmpnd = config.get("compound",true);
+ int mrgf = config.get("merge.factor",10);
+ int mxbf = config.get("max.buffered",10);
+
+ // must update params for newly opened writer
+ writer.setMaxBufferedDocs(mxbf);
+ writer.setMergeFactor(mrgf);
+ writer.setUseCompoundFile(cmpnd); // this one redundant?
+
+ getRunData().setIndexWriter(writer);
+ return 1;
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OpenReaderTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OpenReaderTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OpenReaderTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OpenReaderTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,43 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.io.IOException;
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.index.IndexReader;
+import org.apache.lucene.store.Directory;
+
+/**
+ * Open an index reader.
+ * Other side effects: index redaer object in perfRunData is set.
+ */
+public class OpenReaderTask extends PerfTask {
+
+ public OpenReaderTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws IOException {
+ Directory dir = getRunData().getDirectory();
+ IndexReader reader = IndexReader.open(dir);
+ getRunData().setIndexReader(reader);
+ return 1;
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OptimizeTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OptimizeTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OptimizeTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/OptimizeTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,40 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.index.IndexWriter;
+
+/**
+ * Optimize the index.
+ * Other side effects: none.
+ */
+public class OptimizeTask extends PerfTask {
+
+ public OptimizeTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws Exception {
+ IndexWriter iw = getRunData().getIndexWriter();
+ iw.optimize();
+ //System.out.println("optimize called");
+ return 1;
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/PerfTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/PerfTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/PerfTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/PerfTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,217 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.benchmark.byTask.stats.Points;
+import org.apache.lucene.benchmark.byTask.stats.TaskStats;
+import org.apache.lucene.benchmark.byTask.utils.Format;
+
+/**
+ * A (abstract) task to be tested for performance.
+ * <br>
+ * Every performance task extends this class, and provides its own doLogic() method,
+ * which performss the actual task.
+ * <br>
+ * Tasks performing some work that should be measured for the task, can overide setup() and/or tearDown() and
+ * placed that work there.
+ */
+public abstract class PerfTask implements Cloneable {
+
+ private PerfRunData runData;
+
+ // propeties that all tasks have
+ private String name;
+ private int depth = 0;
+ private int maxDepthLogStart = 0;
+ protected String params = null;
+
+ protected static final String NEW_LINE = System.getProperty("line.separator");
+
+ /**
+ * Should not be used externally
+ */
+ private PerfTask() {
+ name = Format.simpleName(getClass());
+ if (name.endsWith("Task")) {
+ name = name.substring(0,name.length()-4);
+ }
+ }
+
+ public PerfTask(PerfRunData runData) {
+ this();
+ this.runData = runData;
+ this.maxDepthLogStart = runData.getConfig().get("task.max.depth.log",0);
+ }
+
+ /* (non-Javadoc)
+ * @see java.lang.Object#clone()
+ */
+ protected Object clone() throws CloneNotSupportedException {
+ // tasks having non primitive data structures should overide this.
+ // otherwise parallel running of a task sequence might not run crrectly.
+ return super.clone();
+ }
+
+ /**
+ * Run the task, record statistics.
+ * @return number of work items done by this task.
+ */
+ public final int runAndMaybeStats(boolean reportStats) throws Exception {
+ if (reportStats && depth <= maxDepthLogStart && !shouldNeverLogAtStart()) {
+ System.out.println("------------> starting task: " + getName());
+ }
+ if (shouldNotRecordStats() || !reportStats) {
+ setup();
+ int count = doLogic();
+ tearDown();
+ return count;
+ }
+ setup();
+ Points pnts = runData.getPoints();
+ TaskStats ts = pnts.markTaskStart(this,runData.getConfig().getRoundNumber());
+ int count = doLogic();
+ pnts.markTaskEnd(ts, count);
+ tearDown();
+ return count;
+ }
+
+ /**
+ * Perform the task once (ignoring repetions specification)
+ * Return number of work items done by this task.
+ * For indexing that can be number of docs added.
+ * For warming that can be number of scanned items, etc.
+ * @return number of work items done by this task.
+ */
+ public abstract int doLogic() throws Exception;
+
+ /**
+ * @return Returns the name.
+ */
+ public String getName() {
+ if (params==null) {
+ return name;
+ }
+ return new StringBuffer(name).append('(').append(params).append(')').toString();
+ }
+
+ /**
+ * @param name The name to set.
+ */
+ protected void setName(String name) {
+ this.name = name;
+ }
+
+ /**
+ * @return Returns the run data.
+ */
+ public PerfRunData getRunData() {
+ return runData;
+ }
+
+ /**
+ * @return Returns the depth.
+ */
+ public int getDepth() {
+ return depth;
+ }
+
+ /**
+ * @param depth The depth to set.
+ */
+ public void setDepth(int depth) {
+ this.depth = depth;
+ }
+
+ // compute a blank string padding for printing this task indented by its depth
+ String getPadding () {
+ char c[] = new char[4*getDepth()];
+ for (int i = 0; i < c.length; i++) c[i] = ' ';
+ return new String(c);
+ }
+
+ /* (non-Javadoc)
+ * @see java.lang.Object#toString()
+ */
+ public String toString() {
+ String padd = getPadding();
+ StringBuffer sb = new StringBuffer(padd);
+ sb.append(getName());
+ return sb.toString();
+ }
+
+ /**
+ * @return Returns the maxDepthLogStart.
+ */
+ int getMaxDepthLogStart() {
+ return maxDepthLogStart;
+ }
+
+ /**
+ * Tasks that should never log at start can overide this.
+ * @return true if this task should never log when it start.
+ */
+ protected boolean shouldNeverLogAtStart () {
+ return false;
+ }
+
+ /**
+ * Tasks that should not record statistics can overide this.
+ * @return true if this task should never record its statistics.
+ */
+ protected boolean shouldNotRecordStats () {
+ return false;
+ }
+
+ /**
+ * Task setup work that should not be measured for that specific task.
+ * By default it does nothing, but tasks can implement this, moving work from
+ * doLogic() to this method. Only the work done in doLogicis measured for this task.
+ * Notice that higher level (sequence) tasks containing this task would then
+ * measure larger time than the sum of their contained tasks.
+ * @throws Exception
+ */
+ public void setup () throws Exception {
+ }
+
+ /**
+ * Task tearDown work that should not be measured for that specific task.
+ * By default it does nothing, but tasks can implement this, moving work from
+ * doLogic() to this method. Only the work done in doLogicis measured for this task.
+ * Notice that higher level (sequence) tasks containing this task would then
+ * measure larger time than the sum of their contained tasks.
+ */
+ public void tearDown () throws Exception {
+ }
+
+ /**
+ * Set the params of this task.
+ * Sub classes that supports parameters may overide this method for fetching/processing the params.
+ */
+ public void setParams(String params) {
+ this.params = params;
+ }
+
+ /**
+ * @return Returns the Params.
+ */
+ public String getParams() {
+ return params;
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/ReadTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/ReadTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/ReadTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/ReadTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,124 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.benchmark.byTask.feeds.QueryMaker;
+import org.apache.lucene.document.Document;
+import org.apache.lucene.index.IndexReader;
+import org.apache.lucene.search.Hits;
+import org.apache.lucene.search.IndexSearcher;
+import org.apache.lucene.search.Query;
+import org.apache.lucene.store.Directory;
+
+
+
+/**
+ * Read index (abstract) task.
+ * Sub classes implement withSearch(), withWarm(), withTraverse() and withRetrieve()
+ * methods to configure the actual action.
+ * Other side effects: none.
+ */
+public abstract class ReadTask extends PerfTask {
+
+ public ReadTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws Exception {
+ int res = 0;
+ boolean closeReader = false;
+
+ // open reader or use existing one
+ IndexReader ir = getRunData().getIndexReader();
+ if (ir == null) {
+ Directory dir = getRunData().getDirectory();
+ ir = IndexReader.open(dir);
+ closeReader = true;
+ //res++; //this is confusing, comment it out
+ }
+
+ // optionally warm and add num docs traversed to count
+ if (withWarm()) {
+ Document doc = null;
+ for (int m = 0; m < ir.maxDoc(); m++) {
+ if (!ir.isDeleted(m)) {
+ doc = ir.document(m);
+ res += (doc==null ? 0 : 1);
+ }
+ }
+ }
+
+ if (withSearch()) {
+ res++;
+ IndexSearcher searcher = new IndexSearcher(ir);
+ QueryMaker queryMaker = getQueryMaker();
+ Query q = queryMaker.makeQuery();
+ Hits hits = searcher.search(q);
+ //System.out.println("searched: "+q);
+
+ if (withTraverse()) {
+ Document doc = null;
+ if (hits != null && hits.length() > 0) {
+ for (int m = 0; m < hits.length(); m++) {
+ int id = hits.id(m);
+ res++;
+
+ if (withRetrieve()) {
+ doc = ir.document(id);
+ res += (doc==null ? 0 : 1);
+ }
+ }
+ }
+ }
+
+ searcher.close();
+ }
+
+ if (closeReader) {
+ ir.close();
+ }
+ return res;
+ }
+
+ /**
+ * Return query maker used for this task.
+ */
+ public abstract QueryMaker getQueryMaker();
+
+ /**
+ * Return true if search should be performed.
+ */
+ public abstract boolean withSearch ();
+
+ /**
+ * Return true if warming should be performed.
+ */
+ public abstract boolean withWarm ();
+
+ /**
+ * Return true if, with search, results should be traversed.
+ */
+ public abstract boolean withTraverse ();
+
+ /**
+ * Return true if, with search & results traversing, docs should be retrieved.
+ */
+ public abstract boolean withRetrieve ();
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/RepAllTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/RepAllTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/RepAllTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/RepAllTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,43 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.benchmark.byTask.stats.Report;
+
+/**
+ * Report all statistics with no aggregations.
+ * Other side effects: None.
+ */
+public class RepAllTask extends ReportTask {
+
+ public RepAllTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws Exception {
+ Report rp = getRunData().getPoints().reportAll();
+
+ System.out.println();
+ System.out.println("------------> Report All ("+rp.getSize()+" out of "+rp.getOutOf()+")");
+ System.out.println(rp.getText());
+ System.out.println();
+ return 0;
+ }
+
+}
Added: lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/RepSelectByPrefTask.java
URL: http://svn.apache.org/viewvc/lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/RepSelectByPrefTask.java?view=auto&rev=495834
==============================================================================
--- lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/RepSelectByPrefTask.java (added)
+++ lucene/java/trunk/contrib/benchmark/src/java/org/apache/lucene/benchmark/byTask/tasks/RepSelectByPrefTask.java Fri Jan 12 20:08:23 2007
@@ -0,0 +1,44 @@
+package org.apache.lucene.benchmark.byTask.tasks;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import org.apache.lucene.benchmark.byTask.PerfRunData;
+import org.apache.lucene.benchmark.byTask.stats.Report;
+
+/**
+ * Report by-name-prefix statistics with no aggregations.
+ * Other side effects: None.
+ */
+public class RepSelectByPrefTask extends RepSumByPrefTask {
+
+ public RepSelectByPrefTask(PerfRunData runData) {
+ super(runData);
+ }
+
+ public int doLogic() throws Exception {
+ Report rp = getRunData().getPoints().reportSelectByPrefix(prefix);
+
+ System.out.println();
+ System.out.println("------------> Report Select By Prefix ("+prefix+") ("+
+ rp.getSize()+" about "+rp.getReported()+" out of "+rp.getOutOf()+")");
+ System.out.println(rp.getText());
+ System.out.println();
+
+ return 0;
+ }
+}