dnet-hadoop/dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/collection/mag/oozie_app/workflow.xml

90 lines
3.6 KiB
XML

<workflow-app name="generate_MAG_Datasource" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>magBasePath</name>
<description>The base path of MAG DUMP CSV Tables</description>
</property>
<property>
<name>mdstorePath</name>
<description>The base path of MAG DUMP CSV Tables</description>
</property>
<property>
<name>workingPath</name>
<description>The working path</description>
</property>
<property>
<name>resume_from</name>
<value>generateOAF</value>
<description>start Node</description>
</property>
</parameters>
<start to="resume_from"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<decision name="resume_from">
<switch>
<case to="generateTable">${wf:conf('resume_from') eq 'generateTable'}</case>
<default to="generateOAF"/> <!-- first action to be done when downloadDump is to be performed -->
</switch>
</decision>
<action name="generateTable">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Generate MAG Table</name>
<class>eu.dnetlib.dhp.collection.mag.SparkCreateMagDenormalizedTable</class>
<jar>dhp-aggregation-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.executor.memoryOverhead=2g
--conf spark.sql.shuffle.partitions=3000
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--magBasePath</arg><arg>${magBasePath}</arg>
<arg>--workingPath</arg><arg>${workingPath}</arg>
<arg>--master</arg><arg>yarn</arg>
</spark>
<ok to="generateOAF"/>
<error to="Kill"/>
</action>
<action name="generateOAF">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>MAG TO OAF</name>
<class>eu.dnetlib.dhp.collection.mag.SparkMAGtoOAF</class>
<jar>dhp-aggregation-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.executor.memoryOverhead=2g
--conf spark.sql.shuffle.partitions=3000
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--mdstorePath</arg><arg>${mdstorePath}</arg>
<arg>--workingPath</arg><arg>${workingPath}</arg>
<arg>--master</arg><arg>yarn</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>