create a separate wf for resolving relation

This commit is contained in:
Sandro La Bruzzo 2021-07-23 11:40:06 +02:00
parent 43e9380cd3
commit ca74e8dd02
3 changed files with 52 additions and 61 deletions

View File

@ -0,0 +1,52 @@
<workflow-app name="Resolve Relation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>entityPath</name>
<description>the path of deduplicate Entities</description>
</property>
<property>
<name>relationPath</name>
<description>the path of relation unresolved</description>
</property>
<property>
<name>targetPath</name>
<description>the path of relation unresolved</description>
</property>
</parameters>
<start to="ResolveRelations"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="ResolveRelations">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Resolve Relations in raw graph</name>
<class>eu.dnetlib.dhp.sx.graph.SparkResolveRelation</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.shuffle.partitions=3000
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--master</arg><arg>yarn</arg>
<arg>--relationPath</arg><arg>${relationPath}</arg>
<arg>--workingPath</arg><arg>${targetPath}</arg>
<arg>--entityPath</arg><arg>${entityPath}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -1,61 +0,0 @@
<workflow-app name="Create Raw Graph Final Step: Construct the Scholexplorer Raw Graph" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>targetPath</name>
<description>the source path</description>
</property>
<property>
<name>sparkDriverMemory</name>
<description>memory for driver process</description>
</property>
<property>
<name>sparkExecutorMemory</name>
<description>memory for individual executor</description>
</property>
<property>
<name>entity</name>
<description>the entity to be merged</description>
</property>
</parameters>
<start to="DeleteTargetPath"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="DeleteTargetPath">
<fs>
<mkdir path="${targetPath}"/>
<delete path='${targetPath}/${entity}'/>
</fs>
<ok to="MergeDLIEntities"/>
<error to="Kill"/>
</action>
<action name="MergeDLIEntities">
<spark xmlns="uri:oozie:spark-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>Merge ${entity}</name>
<class>eu.dnetlib.dhp.sx.graph.SparkScholexplorerCreateRawGraphJob</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts> --executor-memory ${sparkExecutorMemory} --driver-memory=${sparkDriverMemory} ${sparkExtraOPT}</spark-opts>
<arg>-mt</arg> <arg>yarn-cluster</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/${entity}</arg>
<arg>--targetPath</arg><arg>${targetPath}/${entity}</arg>
<arg>--entity</arg><arg>${entity}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>