2023-12-07 09:59:52 +01:00
|
|
|
<workflow-app name="affiliation_from_semrel_propagation" xmlns="uri:oozie:workflow:0.5">
|
|
|
|
<parameters>
|
|
|
|
<property>
|
|
|
|
<name>sourcePath</name>
|
|
|
|
<description>the source path</description>
|
|
|
|
</property>
|
|
|
|
|
|
|
|
</parameters>
|
|
|
|
|
|
|
|
<global>
|
|
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
|
|
<name-node>${nameNode}</name-node>
|
|
|
|
<configuration>
|
|
|
|
<property>
|
|
|
|
<name>oozie.action.sharelib.for.spark</name>
|
|
|
|
<value>${oozieActionShareLibForSpark2}</value>
|
|
|
|
</property>
|
|
|
|
</configuration>
|
|
|
|
</global>
|
|
|
|
|
2023-12-20 14:26:54 +01:00
|
|
|
<start to="reset_outputpath"/>
|
2023-12-07 09:59:52 +01:00
|
|
|
|
|
|
|
<kill name="Kill">
|
|
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
|
|
</kill>
|
|
|
|
|
2023-12-20 14:26:54 +01:00
|
|
|
<action name="reset_outputpath">
|
|
|
|
<fs>
|
|
|
|
<delete path="${workingDir}"/>
|
|
|
|
<mkdir path="${workingDir}"/>
|
|
|
|
</fs>
|
|
|
|
<ok to="prepare_info"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
2023-12-07 09:59:52 +01:00
|
|
|
|
|
|
|
<action name="prepare_info">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
2023-12-18 09:55:27 +01:00
|
|
|
<name>PrepareResultProjectOrganizationAssociation</name>
|
2023-12-07 09:59:52 +01:00
|
|
|
<class>eu.dnetlib.dhp.entitytoorganizationfromsemrel.PrepareInfo</class>
|
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--driver-memory=${sparkDriverMemory}
|
|
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--graphPath</arg><arg>${sourcePath}</arg>
|
|
|
|
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
|
|
|
|
<arg>--leavesPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/leavesPath</arg>
|
|
|
|
<arg>--childParentPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/childParentPath</arg>
|
|
|
|
<arg>--resultOrgPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/resultOrgPath</arg>
|
|
|
|
<arg>--projectOrganizationPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/projectOrganizationPath</arg>
|
|
|
|
<arg>--relationPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/relation</arg>
|
|
|
|
</spark>
|
|
|
|
<ok to="apply_resulttoorganization_propagation"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<action name="apply_resulttoorganization_propagation">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
2023-12-18 09:55:27 +01:00
|
|
|
<name>entityToOrganizationFromSemRel</name>
|
|
|
|
<class>eu.dnetlib.dhp.entitytoorganizationfromsemrel.SparkEntityToOrganizationFromSemRel</class>
|
2023-12-07 09:59:52 +01:00
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--driver-memory=${sparkDriverMemory}
|
|
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
|
|
--conf spark.dynamicAllocation.enabled=true
|
|
|
|
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
|
|
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--relationPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/relation</arg>
|
|
|
|
<arg>--outputPath</arg><arg>${sourcePath}/relation</arg>
|
|
|
|
<arg>--leavesPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/leavesPath</arg>
|
|
|
|
<arg>--childParentPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/childParentPath</arg>
|
|
|
|
<arg>--resultOrgPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/resultOrgPath</arg>
|
|
|
|
<arg>--projectOrganizationPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/projectOrganizationPath</arg>
|
|
|
|
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
|
|
|
|
<arg>--workingDir</arg><arg>${workingDir}/entitiesSemanticRelation/working</arg>
|
|
|
|
<arg>--iterations</arg><arg>${iterations}</arg>
|
|
|
|
</spark>
|
|
|
|
<ok to="End"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<end name="End"/>
|
|
|
|
|
|
|
|
</workflow-app>
|