208 lines
8.7 KiB
XML
208 lines
8.7 KiB
XML
|
|
<workflow-app name="UnresolvedEntities" xmlns="uri:oozie:workflow:0.5">
|
|
<parameters>
|
|
<property>
|
|
<name>fosPath</name>
|
|
<description>the input path of the resources to be extended</description>
|
|
</property>
|
|
<property>
|
|
<name>outputPath</name>
|
|
<description>the path where to store the actionset</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>sparkExecutorCores</name>
|
|
<description>number of cores used by single executor</description>
|
|
</property>
|
|
<property>
|
|
<name>oozieActionShareLibForSpark2</name>
|
|
<description>oozie action sharelib for spark 2.*</description>
|
|
</property>
|
|
<property>
|
|
<name>spark2ExtraListeners</name>
|
|
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
|
|
<description>spark 2.* extra listeners classname</description>
|
|
</property>
|
|
<property>
|
|
<name>spark2SqlQueryExecutionListeners</name>
|
|
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
|
|
<description>spark 2.* sql query execution listeners classname</description>
|
|
</property>
|
|
<property>
|
|
<name>spark2YarnHistoryServerAddress</name>
|
|
<description>spark 2.* yarn history server address</description>
|
|
</property>
|
|
<property>
|
|
<name>spark2EventLogDir</name>
|
|
<description>spark 2.* event log dir location</description>
|
|
</property>
|
|
</parameters>
|
|
|
|
<global>
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
<name-node>${nameNode}</name-node>
|
|
<configuration>
|
|
<property>
|
|
<name>mapreduce.job.queuename</name>
|
|
<value>${queueName}</value>
|
|
</property>
|
|
<property>
|
|
<name>oozie.launcher.mapred.job.queue.name</name>
|
|
<value>${oozieLauncherQueueName}</value>
|
|
</property>
|
|
<property>
|
|
<name>oozie.action.sharelib.for.spark</name>
|
|
<value>${oozieActionShareLibForSpark2}</value>
|
|
</property>
|
|
|
|
</configuration>
|
|
</global>
|
|
<start to="prepareInfo"/>
|
|
|
|
<kill name="Kill">
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
</kill>
|
|
|
|
|
|
<fork name="prepareInfo">
|
|
<path start="getFOS"/>
|
|
<path start="getSDG"/>
|
|
</fork>
|
|
|
|
<action name="getFOS">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Gets Data from FOS csv file</name>
|
|
<class>eu.dnetlib.dhp.actionmanager.createunresolvedentities.GetFOSSparkJob</class>
|
|
<jar>dhp-aggregation-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-memory=${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}${spark2EventLogDir}
|
|
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${fosPath}</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/input/fos</arg>
|
|
<arg>--delimiter</arg><arg>${delimiter}</arg>
|
|
</spark>
|
|
<ok to="prepareFos"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
|
|
<action name="prepareFos">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Produces the unresolved from FOS</name>
|
|
<class>eu.dnetlib.dhp.actionmanager.createunresolvedentities.PrepareFOSSparkJob</class>
|
|
<jar>dhp-aggregation-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--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}
|
|
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${workingDir}/input/fos</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/prepared</arg>
|
|
|
|
</spark>
|
|
<ok to="join"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="getSDG">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Gets Data from SDG csv file</name>
|
|
<class>eu.dnetlib.dhp.actionmanager.createunresolvedentities.GetSDGSparkJob</class>
|
|
<jar>dhp-aggregation-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--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}
|
|
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${sdgPath}</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/input/sdg</arg>
|
|
</spark>
|
|
<ok to="prepareSDG"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
|
|
<action name="prepareSDG">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Produces the unresolved from FOS</name>
|
|
<class>eu.dnetlib.dhp.actionmanager.createunresolvedentities.PrepareSDGSparkJob</class>
|
|
<jar>dhp-aggregation-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--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}
|
|
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${workingDir}/input/sdg</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/prepared</arg>
|
|
|
|
</spark>
|
|
<ok to="join"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
|
|
<join name="join" to="produceUnresolved"/>
|
|
|
|
<action name="produceUnresolved">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Save the unresolved entities grouping results with the same id</name>
|
|
<class>eu.dnetlib.dhp.actionmanager.createunresolvedentities.SparkSaveUnresolved</class>
|
|
<jar>dhp-aggregation-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--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}
|
|
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${workingDir}/prepared</arg>
|
|
<arg>--outputPath</arg><arg>${outputPath}</arg>
|
|
</spark>
|
|
<ok to="End"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<end name="End"/>
|
|
</workflow-app> |