modified resource file, workflow anf default-config. Add 3g of memory Overhead and specified the shuffle partition in the wf confiduration. Removed the multiple instantiation in the wf because of different implementation of the spark job

dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/bulktag/input_bulkTag_parameters.json

@@ -1,10 +1,5 @@
 [
-  {
+
-    "paramName":"is",
-    "paramLongName":"isLookUpUrl",
-    "paramDescription": "URL of the isLookUp Service",
-    "paramRequired": true
-  },
   {
     "paramName":"s",
     "paramLongName":"sourcePath",
@@ -17,12 +12,7 @@
     "paramDescription": "the json path associated to each selection field",
     "paramRequired": true
   },
-  {
+
-    "paramName":"tn",
-    "paramLongName":"resultTableName",
-    "paramDescription": "the name of the result table we are currently working on",
-    "paramRequired": true
-  },
   {
     "paramName": "out",
     "paramLongName": "outputPath",
@@ -35,17 +25,19 @@
     "paramDescription": "true if the spark session is managed, false otherwise",
     "paramRequired": false
   },
-  {
+
-    "paramName": "test",
-    "paramLongName": "isTest",
-    "paramDescription": "Parameter intended for testing purposes only. True if the reun is relatesd to a test and so the taggingConf parameter should be loaded",
-    "paramRequired": false
-  },
   {
     "paramName": "tg",
     "paramLongName": "taggingConf",
     "paramDescription": "this parameter is intended for testing purposes only. It is a possible tagging configuration obtained via the XQUERY. Intended to be removed",
     "paramRequired": false
+  },
+
+  {
+    "paramName": "p",
+    "paramLongName": "production",
+    "paramDescription": "this parameter is intended for testing purposes only. It is a possible tagging configuration obtained via the XQUERY. Intended to be removed",
+    "paramRequired": true
   }
 
 ]

dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/bulktag/oozie_app/config-default.xml

@@ -45,10 +45,14 @@
     </property>
     <property>
         <name>sparkExecutorMemory</name>
-        <value>6G</value>
+        <value>5G</value>
     </property>
     <property>
-        <name>sparkExecutorCores</name>
+        <name>memoryOverhead</name>
-        <value>1</value>
+        <value>3g</value>
+    </property>
+    <property>
+        <name>partitions</name>
+        <value>3284</value>
     </property>
 </configuration>

dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/bulktag/oozie_app/workflow.xml

@@ -4,10 +4,6 @@
             <name>sourcePath</name>
             <description>the source path</description>
         </property>
-        <property>
-            <name>isLookUpUrl</name>
-            <description>the isLookup service endpoint</description>
-        </property>
         <property>
             <name>pathMap</name>
             <description>the json path associated to each selection field</description>
@@ -102,16 +98,9 @@
         <error to="Kill"/>
     </action>
 
-    <join name="copy_wait" to="fork_exec_bulktag"/>
+    <join name="copy_wait" to="exec_bulktag"/>
 
-    <fork name="fork_exec_bulktag">
+    <action name="exec_bulktag">
-        <path start="bulktag_publication"/>
-        <path start="bulktag_dataset"/>
-        <path start="bulktag_otherresearchproduct"/>
-        <path start="bulktag_software"/>
-    </fork>
-
-    <action name="bulktag_publication">
         <spark xmlns="uri:oozie:spark-action:0.2">
             <master>yarn-cluster</master>
             <mode>cluster</mode>
@@ -122,104 +111,23 @@
                 --num-executors=${sparkExecutorNumber}
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
+                --conf spark.executor.memoryOverhead=${memeoryOverhead}
+                --conf spark.sql.shuffle.partitions=${partitions}
                 --driver-memory=${sparkDriverMemory}
                 --conf spark.extraListeners=${spark2ExtraListeners}
                 --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
             </spark-opts>
-            <arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
+            <arg>--sourcePath</arg><arg>${sourcePath}/</arg>
-            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
+            <arg>--outputPath</arg><arg>${outputPath}/</arg>
-            <arg>--outputPath</arg><arg>${outputPath}/publication</arg>
             <arg>--pathMap</arg><arg>${pathMap}</arg>
-            <arg>--isLookUpUrl</arg><arg>${isLookUpUrl}</arg>
+            <arg>--production</arg><arg>${production}</arg>
         </spark>
-        <ok to="wait"/>
+        <ok to="End"/>
         <error to="Kill"/>
     </action>
 
-    <action name="bulktag_dataset">
-        <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
-            <mode>cluster</mode>
-            <name>bulkTagging-dataset</name>
-            <class>eu.dnetlib.dhp.bulktag.SparkBulkTagJob</class>
-            <jar>dhp-enrichment-${projectVersion}.jar</jar>
-            <spark-opts>
-                --num-executors=${sparkExecutorNumber}
-                --executor-memory=${sparkExecutorMemory}
-                --executor-cores=${sparkExecutorCores}
-                --driver-memory=${sparkDriverMemory}
-                --conf spark.extraListeners=${spark2ExtraListeners}
-                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
-                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
-                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-            </spark-opts>
-            <arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
-            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
-            <arg>--outputPath</arg><arg>${outputPath}/dataset</arg>
-            <arg>--pathMap</arg><arg>${pathMap}</arg>
-            <arg>--isLookUpUrl</arg><arg>${isLookUpUrl}</arg>
-        </spark>
-        <ok to="wait"/>
-        <error to="Kill"/>
-    </action>
-
-    <action name="bulktag_otherresearchproduct">
-        <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
-            <mode>cluster</mode>
-            <name>bulkTagging-orp</name>
-            <class>eu.dnetlib.dhp.bulktag.SparkBulkTagJob</class>
-            <jar>dhp-enrichment-${projectVersion}.jar</jar>
-            <spark-opts>
-                --num-executors=${sparkExecutorNumber}
-                --executor-memory=${sparkExecutorMemory}
-                --executor-cores=${sparkExecutorCores}
-                --driver-memory=${sparkDriverMemory}
-                --conf spark.extraListeners=${spark2ExtraListeners}
-                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
-                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
-                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-            </spark-opts>
-            <arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
-            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
-            <arg>--outputPath</arg><arg>${outputPath}/otherresearchproduct</arg>
-            <arg>--pathMap</arg><arg>${pathMap}</arg>
-            <arg>--isLookUpUrl</arg><arg>${isLookUpUrl}</arg>
-        </spark>
-        <ok to="wait"/>
-        <error to="Kill"/>
-    </action>
-
-    <action name="bulktag_software">
-        <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
-            <mode>cluster</mode>
-            <name>bulkTagging-software</name>
-            <class>eu.dnetlib.dhp.bulktag.SparkBulkTagJob</class>
-            <jar>dhp-enrichment-${projectVersion}.jar</jar>
-            <spark-opts>
-                --num-executors=${sparkExecutorNumber}
-                --executor-memory=${sparkExecutorMemory}
-                --executor-cores=${sparkExecutorCores}
-                --driver-memory=${sparkDriverMemory}
-                --conf spark.extraListeners=${spark2ExtraListeners}
-                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
-                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
-                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
-            </spark-opts>
-            <arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
-            <arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
-            <arg>--outputPath</arg><arg>${outputPath}/software</arg>
-            <arg>--pathMap</arg><arg>${pathMap}</arg>
-            <arg>--isLookUpUrl</arg><arg>${isLookUpUrl}</arg>
-        </spark>
-        <ok to="wait"/>
-        <error to="Kill"/>
-    </action>
-
-    <join name="wait" to="End"/>
 
 
     <end name="End"/>