From b0eba210c068219580cfa78c17aa23f2e1e170f8 Mon Sep 17 00:00:00 2001
From: Claudio Atzori <claudio.atzori@isti.cnr.it>
Date: Mon, 10 Jun 2024 16:15:07 +0200
Subject: [PATCH] [actionset promotion] use sparkExecutorMemory to define also
 the memoryOverhead

---
 dhp-shade-package/dependency-reduced-pom.xml  | 113 ++++++++++++++++++
 .../wf/dataset/oozie_app/workflow.xml         |   2 +
 .../wf/datasource/oozie_app/workflow.xml      |   1 +
 .../wf/main/oozie_app/workflow.xml            |   1 +
 .../wf/organization/oozie_app/workflow.xml    |   1 +
 .../oozie_app/workflow.xml                    |   4 +-
 .../wf/project/oozie_app/workflow.xml         |   1 +
 .../wf/publication/oozie_app/workflow.xml     |   6 +-
 .../wf/relation/oozie_app/workflow.xml        |   3 +-
 .../wf/software/oozie_app/workflow.xml        |   4 +-
 10 files changed, 131 insertions(+), 5 deletions(-)
 create mode 100644 dhp-shade-package/dependency-reduced-pom.xml

diff --git a/dhp-shade-package/dependency-reduced-pom.xml b/dhp-shade-package/dependency-reduced-pom.xml
new file mode 100644
index 000000000..04843072f
--- /dev/null
+++ b/dhp-shade-package/dependency-reduced-pom.xml
@@ -0,0 +1,113 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
+  <parent>
+    <artifactId>dhp</artifactId>
+    <groupId>eu.dnetlib.dhp</groupId>
+    <version>1.2.5-SNAPSHOT</version>
+  </parent>
+  <modelVersion>4.0.0</modelVersion>
+  <artifactId>dhp-shade-package</artifactId>
+  <description>This module create a jar of all module dependencies</description>
+  <build>
+    <plugins>
+      <plugin>
+        <artifactId>maven-shade-plugin</artifactId>
+        <executions>
+          <execution>
+            <phase>package</phase>
+            <goals>
+              <goal>shade</goal>
+            </goals>
+            <configuration>
+              <transformers>
+                <transformer>
+                  <mainClass>eu.dnetlib.dhp.oa.dedup.SparkCreateSimRels</mainClass>
+                </transformer>
+                <transformer />
+                <transformer>
+                  <resource>META-INF/cxf/bus-extensions.txt</resource>
+                </transformer>
+              </transformers>
+              <filters>
+                <filter>
+                  <artifact>*:*</artifact>
+                  <excludes>
+                    <exclude>META-INF/maven/**</exclude>
+                    <exclude>META-INF/*.SF</exclude>
+                    <exclude>META-INF/*.DSA</exclude>
+                    <exclude>META-INF/*.RSA</exclude>
+                  </excludes>
+                </filter>
+              </filters>
+              <relocations>
+                <relocation>
+                  <pattern>com</pattern>
+                  <shadedPattern>repackaged.com.google.common</shadedPattern>
+                  <includes>
+                    <include>com.google.common.**</include>
+                  </includes>
+                </relocation>
+              </relocations>
+            </configuration>
+          </execution>
+        </executions>
+      </plugin>
+    </plugins>
+  </build>
+  <dependencies>
+    <dependency>
+      <groupId>org.projectlombok</groupId>
+      <artifactId>lombok</artifactId>
+      <version>1.18.28</version>
+      <scope>provided</scope>
+    </dependency>
+    <dependency>
+      <groupId>org.junit.jupiter</groupId>
+      <artifactId>junit-jupiter</artifactId>
+      <version>5.6.1</version>
+      <scope>test</scope>
+      <exclusions>
+        <exclusion>
+          <artifactId>junit-jupiter-api</artifactId>
+          <groupId>org.junit.jupiter</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>junit-jupiter-params</artifactId>
+          <groupId>org.junit.jupiter</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>junit-jupiter-engine</artifactId>
+          <groupId>org.junit.jupiter</groupId>
+        </exclusion>
+      </exclusions>
+    </dependency>
+    <dependency>
+      <groupId>org.mockito</groupId>
+      <artifactId>mockito-core</artifactId>
+      <version>3.3.3</version>
+      <scope>test</scope>
+      <exclusions>
+        <exclusion>
+          <artifactId>byte-buddy</artifactId>
+          <groupId>net.bytebuddy</groupId>
+        </exclusion>
+        <exclusion>
+          <artifactId>byte-buddy-agent</artifactId>
+          <groupId>net.bytebuddy</groupId>
+        </exclusion>
+      </exclusions>
+    </dependency>
+    <dependency>
+      <groupId>org.mockito</groupId>
+      <artifactId>mockito-junit-jupiter</artifactId>
+      <version>3.3.3</version>
+      <scope>test</scope>
+    </dependency>
+  </dependencies>
+  <distributionManagement>
+    <site>
+      <id>DHPSite</id>
+      <url>${dhp.site.stage.path}/dhp-common</url>
+    </site>
+  </distributionManagement>
+</project>
diff --git a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/dataset/oozie_app/workflow.xml b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/dataset/oozie_app/workflow.xml
index 5401b45ca..b1bc1d6e1 100644
--- a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/dataset/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/dataset/oozie_app/workflow.xml
@@ -103,6 +103,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
@@ -156,6 +157,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
diff --git a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/datasource/oozie_app/workflow.xml b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/datasource/oozie_app/workflow.xml
index f9bd66ae3..9a84f4708 100644
--- a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/datasource/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/datasource/oozie_app/workflow.xml
@@ -95,6 +95,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
diff --git a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/main/oozie_app/workflow.xml b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/main/oozie_app/workflow.xml
index 393f04e89..65ddd402b 100644
--- a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/main/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/main/oozie_app/workflow.xml
@@ -125,6 +125,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
diff --git a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/organization/oozie_app/workflow.xml b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/organization/oozie_app/workflow.xml
index ebfdeee31..0d73b498d 100644
--- a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/organization/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/organization/oozie_app/workflow.xml
@@ -95,6 +95,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
diff --git a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/otherresearchproduct/oozie_app/workflow.xml b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/otherresearchproduct/oozie_app/workflow.xml
index 02399ed9b..ca8362c9b 100644
--- a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/otherresearchproduct/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/otherresearchproduct/oozie_app/workflow.xml
@@ -103,6 +103,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
@@ -155,11 +156,12 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-                --conf spark.sql.shuffle.partitions=2560
+                --conf spark.sql.shuffle.partitions=8000
             </spark-opts>
             <arg>--inputGraphTablePath</arg><arg>${workingDir}/otherresearchproduct</arg>
             <arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
diff --git a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/project/oozie_app/workflow.xml b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/project/oozie_app/workflow.xml
index 57c2357b4..37310da79 100644
--- a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/project/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/project/oozie_app/workflow.xml
@@ -95,6 +95,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
diff --git a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/publication/oozie_app/workflow.xml b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/publication/oozie_app/workflow.xml
index 92b114776..a4b0b237c 100644
--- a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/publication/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/publication/oozie_app/workflow.xml
@@ -103,11 +103,12 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-                --conf spark.sql.shuffle.partitions=7000
+                --conf spark.sql.shuffle.partitions=15000
             </spark-opts>
             <arg>--inputGraphTablePath</arg><arg>${inputGraphRootPath}/publication</arg>
             <arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
@@ -156,11 +157,12 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-                --conf spark.sql.shuffle.partitions=7000
+                --conf spark.sql.shuffle.partitions=15000
             </spark-opts>
             <arg>--inputGraphTablePath</arg><arg>${workingDir}/publication</arg>
             <arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
diff --git a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/relation/oozie_app/workflow.xml b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/relation/oozie_app/workflow.xml
index e9e5f0b45..f72847ba8 100644
--- a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/relation/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/relation/oozie_app/workflow.xml
@@ -95,11 +95,12 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-                --conf spark.sql.shuffle.partitions=10000
+                --conf spark.sql.shuffle.partitions=15000
             </spark-opts>
             <arg>--inputGraphTablePath</arg><arg>${inputGraphRootPath}/relation</arg>
             <arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Relation</arg>
diff --git a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/software/oozie_app/workflow.xml b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/software/oozie_app/workflow.xml
index 1d36ddf94..dbe8a63c1 100644
--- a/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/software/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/software/oozie_app/workflow.xml
@@ -103,6 +103,7 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
@@ -155,11 +156,12 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.executor.memoryOverhead=${sparkExecutorMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-                --conf spark.sql.shuffle.partitions=2560
+                --conf spark.sql.shuffle.partitions=4000
             </spark-opts>
             <arg>--inputGraphTablePath</arg><arg>${workingDir}/software</arg>
             <arg>--graphTableClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>