forked from D-Net/dnet-hadoop
105 lines
4.0 KiB
XML
105 lines
4.0 KiB
XML
<workflow-app name="Update Graph" xmlns="uri:oozie:workflow:0.5">
|
|
<parameters>
|
|
<property>
|
|
<name>graphBasePath</name>
|
|
<description>the raw graph base path</description>
|
|
</property>
|
|
<property>
|
|
<name>workingPath</name>
|
|
<description>path of the working directory</description>
|
|
</property>
|
|
<property>
|
|
<name>dedupGraphPath</name>
|
|
<description>path of the dedup graph</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>
|
|
</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>
|
|
</configuration>
|
|
</global>
|
|
|
|
<start to="UpdateEntity"/>
|
|
|
|
<kill name="Kill">
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
</kill>
|
|
|
|
<action name="UpdateEntity">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<prepare>
|
|
<delete path='${dedupGraphPath}'/>
|
|
</prepare>
|
|
<master>yarn-cluster</master>
|
|
<mode>cluster</mode>
|
|
<name>Update Entity</name>
|
|
<class>eu.dnetlib.dhp.dedup.SparkUpdateEntity</class>
|
|
<jar>dhp-dedup-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-memory ${sparkExecutorMemory}
|
|
--executor-cores ${sparkExecutorCores}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener"
|
|
--conf spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener"
|
|
--conf spark.sql.warehouse.dir="/user/hive/warehouse"
|
|
</spark-opts>
|
|
<arg>-mt</arg><arg>yarn-cluster</arg>
|
|
<arg>--i</arg><arg>${graphBasePath}</arg>
|
|
<arg>--w</arg><arg>${workingPath}</arg>
|
|
<arg>--o</arg><arg>${dedupGraphPath}</arg>
|
|
</spark>
|
|
<ok to="PropagateRelation"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="PropagateRelation">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<prepare>
|
|
<delete path='${dedupGraphPath}/relation'/>
|
|
</prepare>
|
|
<master>yarn-cluster</master>
|
|
<mode>cluster</mode>
|
|
<name>Update Relations</name>
|
|
<class>eu.dnetlib.dhp.dedup.SparkPropagateRelation</class>
|
|
<jar>dhp-dedup-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-memory ${sparkExecutorMemory}
|
|
--executor-cores ${sparkExecutorCores}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener"
|
|
--conf spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener"
|
|
--conf spark.sql.warehouse.dir="/user/hive/warehouse"
|
|
</spark-opts>
|
|
<arg>-mt</arg><arg>yarn-cluster</arg>
|
|
<arg>--i</arg><arg>${graphBasePath}</arg>
|
|
<arg>--o</arg><arg>${dedupGraphPath}</arg>
|
|
<arg>--w</arg><arg>${workingPath}</arg>
|
|
</spark>
|
|
<ok to="End"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<end name="End"/>
|
|
</workflow-app> |