Skip to content

Apache Lucene open-source search software

LUCENE-9651 Update benchmark module docs (#759)

Mike Drobb3906e96eaea
diff --git a/gradle/datasets/external-datasets.gradle b/gradle/datasets/external-datasets.gradle
index f466909dac9..4df47bec41b 100644
--- a/gradle/datasets/external-datasets.gradle
+++ b/gradle/datasets/external-datasets.gradle
@@ -24,7 +24,7 @@ configure(project(":lucene:benchmark")) {
apply plugin: "de.undercouch.download"
ext {
- dataDir = file("data")
+ dataDir = file("work")
}
task getEnWiki(type: Download) {
@@ -120,10 +120,9 @@ configure(project(":lucene:benchmark")) {
task getReuters(type: Download) {
ext {
name = "reuters21578"
- // note: there is no HTTPS url and we don't care because this is merely test/perf data
- src = "http://www.daviddlewis.com/resources/testcollections/reuters21578/${name}.tar.gz"
+ src = "https://kdd.ics.uci.edu/databases/${name}/${name}.tar.gz"
intermediate = file("${dataDir}/${name}.tar.gz")
- dst = file("${dataDir}/${name}")
+ dst = file("${dataDir}/reuters-out")
}
outputs.dir ext.dst
@@ -171,4 +170,4 @@ configure(project(":lucene:benchmark")) {
logger.lifecycle("Downloading data set ${task.ext.name} from ${task.ext.src} to ${task.ext.dst}...")
}
}
-}
\ No newline at end of file
+}
diff --git a/gradle/validation/validate-source-patterns.gradle b/gradle/validation/validate-source-patterns.gradle
index 019373236f4..271084e616a 100644
--- a/gradle/validation/validate-source-patterns.gradle
+++ b/gradle/validation/validate-source-patterns.gradle
@@ -108,6 +108,7 @@ allprojects {
configure(project(':lucene:benchmark')) {
project.tasks.withType(ValidateSourcePatternsTask) {
sourceFiles.exclude 'data/**'
+ sourceFiles.exclude 'work/**'
// Known .txt offenders.
sourceFiles.exclude '**/reuters.first20.lines.txt', '**/trecQRels.txt'
diff --git a/lucene/benchmark/.gitignore b/lucene/benchmark/.gitignore
index 249cda967c1..b394c390548 100644
--- a/lucene/benchmark/.gitignore
+++ b/lucene/benchmark/.gitignore
@@ -1 +1,2 @@
-/data
\ No newline at end of file
+/data
+/work
diff --git a/lucene/benchmark/conf/createLineFile.alg b/lucene/benchmark/conf/createLineFile.alg
index b53456ac994..1a112fd6966 100644
--- a/lucene/benchmark/conf/createLineFile.alg
+++ b/lucene/benchmark/conf/createLineFile.alg
@@ -20,9 +20,9 @@
# This alg will process the Reuters documents feed to produce a
# single file that contains all documents, one per line.
#
-# To use this, first cd to benchmark and then run:
+# To use this run:
#
-# ant run-task -Dtask.alg=conf/createLineFile.alg
+# gradlew :lucene:benchmark:run -Ptask.alg=conf/createLineFile.alg
#
# Then, to index the documents in the line file, see
# indexLineFile.alg.
diff --git a/lucene/benchmark/conf/extractWikipedia.alg b/lucene/benchmark/conf/extractWikipedia.alg
index dfc469ebafe..704e4c34412 100644
--- a/lucene/benchmark/conf/extractWikipedia.alg
+++ b/lucene/benchmark/conf/extractWikipedia.alg
@@ -20,9 +20,9 @@
# This alg will process the Wikipedia documents feed to produce a
# single file that contains all documents, one per line.
#
-# To use this, first cd to benchmark and then run:
+# To use this run:
#
-# ant run-task -Dtask.alg=conf/extractWikipedia.alg
+# gradlew :lucene:benchmark:run -Ptask.alg=conf/extractWikipedia.alg
#
# Then, to index the documents in the line file, see
# indexLineFile.alg.
diff --git a/lucene/benchmark/conf/indexLineFile.alg b/lucene/benchmark/conf/indexLineFile.alg
index 5748d42a718..db43841f8ad 100644
--- a/lucene/benchmark/conf/indexLineFile.alg
+++ b/lucene/benchmark/conf/indexLineFile.alg
@@ -23,10 +23,9 @@
# document to let you more accurately measure time spent analyzing and
# indexing your documents vs time spent creating the documents.
#
-# To use this, you must first run the createLineFile.alg, then cd to
-# benchmark and then run:
+# To use this, you must first run the createLineFile.alg, then run:
#
-# ant run-task -Dtask.alg=conf/indexLineFile.alg
+# gradlew :lucene:benchmark:run -Ptask.alg=conf/indexLineFile.alg
#
analyzer=org.apache.lucene.analysis.core.SimpleAnalyzer
diff --git a/lucene/benchmark/conf/micro-standard.alg b/lucene/benchmark/conf/micro-standard.alg
index ea875c1f1c9..452f95e3c9a 100644
--- a/lucene/benchmark/conf/micro-standard.alg
+++ b/lucene/benchmark/conf/micro-standard.alg
@@ -30,8 +30,7 @@ doc.tokenized=true
doc.term.vector=false
log.step=500
-work.dir=data
-docs.dir=reuters21578
+docs.dir=reuters-out
#content.source=org.apache.lucene.benchmark.byTask.feeds.SingleDocSource
content.source=org.apache.lucene.benchmark.byTask.feeds.ReutersContentSource
diff --git a/lucene/benchmark/conf/readContentSource.alg b/lucene/benchmark/conf/readContentSource.alg
index 0759cc9f0ff..f069cf747c4 100644
--- a/lucene/benchmark/conf/readContentSource.alg
+++ b/lucene/benchmark/conf/readContentSource.alg
@@ -22,9 +22,9 @@
# gather baselines for operations like indexing (if reading from the content
# source takes 'X' time, we cannot index faster).
#
-# To use this, first cd to benchmark and then run:
+# To use this run:
#
-# ant run-task -Dtask.alg=conf/readContentSource.alg
+# gradlew :lucene:benchmark:run -Ptask.alg=conf/readContentSource.alg
#
# Where to get documents from:
diff --git a/lucene/benchmark/conf/tokenize.alg b/lucene/benchmark/conf/tokenize.alg
index 2ab105d0401..9d201b8f053 100644
--- a/lucene/benchmark/conf/tokenize.alg
+++ b/lucene/benchmark/conf/tokenize.alg
@@ -20,9 +20,9 @@
# This alg reads all tokens out of a document but does not index them.
# This is useful for benchmarking tokenizers.
#
-# To use this, cd to benchmark and then run:
+# To use this run:
#
-# ant run-task -Dtask.alg=conf/tokenize.alg
+# gradlew :lucene:benchmark:run -Ptask.alg=conf/tokenize.alg
#
content.source=org.apache.lucene.benchmark.byTask.feeds.ReutersContentSource
diff --git a/lucene/benchmark/src/java/org/apache/lucene/benchmark/byTask/package-info.java b/lucene/benchmark/src/java/org/apache/lucene/benchmark/byTask/package-info.java
index e0eed11e38c..ed66c34ca69 100644
--- a/lucene/benchmark/src/java/org/apache/lucene/benchmark/byTask/package-info.java
+++ b/lucene/benchmark/src/java/org/apache/lucene/benchmark/byTask/package-info.java
@@ -81,9 +81,9 @@
* <ul>
* <li>./gradlew -p lucene/benchmark getReuters run <br>
* - would run the <code>micro-standard.alg</code> "algorithm".
- * <li>ant run-task -Dtask.alg=conf/compound-penalty.alg <br>
+ * <li>./gradlew -p lucene/benchmark getReuters run -Ptask.alg=conf/compound-penalty.alg <br>
* - would run the <code>compound-penalty.alg</code> "algorithm".
- * <li>ant run-task -Dtask.alg=[full-path-to-your-alg-file] <br>
+ * <li>./gradlew -p lucene/benchmark getReuters run -Ptask.alg=[full-path-to-your-alg-file] <br>
* - would run <code>your perf test</code> "algorithm".
* <li>java org.apache.lucene.benchmark.byTask.programmatic.Sample <br>
* - would run a performance test programmatically - without using an alg file. This is less
@@ -131,8 +131,8 @@
* benchmark.ext.classpath property:
*
* <ul>
- * <li>ant run-task -Dtask.alg=[full-path-to-your-alg-file] <span style="color:
- * #FF0000">-Dbenchmark.ext.classpath=/mydir/classes </span> -Dtask.mem=512M
+ * <li>./gradlew -p lucene/benchmark run -Ptask.alg=[full-path-to-your-alg-file] <span
+ * style="color: #FF0000">-Dbenchmark.ext.classpath=/mydir/classes </span> -Dtask.mem=512M
* </ul>
*
* <p><u>External tasks</u>: When writing your own tasks under a package other than
@@ -494,7 +494,7 @@
* </pre>
*
* <p>The command line for running this sample: <br>
- * <code>ant run-task -Dtask.alg=conf/sample.alg</code>
+ * <code>./gradlew -p lucene/benchmark getReuters run -Ptask.alg=conf/sample.alg</code>
*
* <p>The output report from running this test contains the following:
*
diff --git a/lucene/benchmark/src/java/org/apache/lucene/benchmark/package-info.java b/lucene/benchmark/src/java/org/apache/lucene/benchmark/package-info.java
index 3f86e012a36..3d4fb8557ca 100644
--- a/lucene/benchmark/src/java/org/apache/lucene/benchmark/package-info.java
+++ b/lucene/benchmark/src/java/org/apache/lucene/benchmark/package-info.java
@@ -35,12 +35,12 @@
* alternate views or to take in command line options. When reporting benchmarking runs you should
* state any alterations you have made.
*
- * <p>To run the short version of the StandardBenchmarker, call "ant run-micro-standard". This
- * should take a minute or so to complete and give you a preliminary idea of how your change affects
- * the code.
+ * <p>To run the short version of the StandardBenchmarker, call "./gradlew -p lucene/benchmark run".
+ * This should take a minute or so to complete and give you a preliminary idea of how your change
+ * affects the code.
*
- * <p>To run the long version of the StandardBenchmarker, call "ant run-standard". This takes
- * considerably longer.
+ * <p>To run the long version of the StandardBenchmarker, call "./gradlew -p lucene/benchmark run
+ * -PtaskAlg=conf/standard.alg". This takes considerably longer, maybe 10 minutes.
*
* <p>The original code for these classes was donated by Andrzej Bialecki at
* http://issues.apache.org/jira/browse/LUCENE-675 and has been updated by Grant Ingersoll to make