forked from D-Net/dnet-hadoop
52 lines
1.9 KiB
XML
52 lines
1.9 KiB
XML
<workflow-app name="Dedup Entities" xmlns="uri:oozie:workflow:0.5">
|
|
<parameters>
|
|
<property>
|
|
<name>relationPath</name>
|
|
<description>the source path</description>
|
|
</property>
|
|
<property>
|
|
<name>mergeRelPath</name>
|
|
<description>the target 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>
|
|
</parameters>
|
|
|
|
<start to="PropagateRelation"/>
|
|
|
|
|
|
<kill name="Kill">
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
</kill>
|
|
|
|
<action name="PropagateRelation">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
<name-node>${nameNode}</name-node>
|
|
<master>yarn-cluster</master>
|
|
<mode>cluster</mode>
|
|
<name>Propagate Dedup Relations</name>
|
|
<class>eu.dnetlib.dedup.SparkPropagateRelationsJob</class>
|
|
<jar>dhp-dedup-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-memory ${sparkExecutorMemory}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--num-executors 100
|
|
--conf spark.yarn.jars="hdfs://hadoop-rm1.garr-pa1.d4science.org:8020/user/oozie/share/lib/lib_20180405103059/spark2"
|
|
</spark-opts>
|
|
<arg>-mt</arg><arg>yarn-cluster</arg>
|
|
<arg>--mergeRelPath</arg><arg>${mergeRelPath}</arg>
|
|
<arg>--relationPath</arg><arg>${relationPath}</arg>
|
|
</spark>
|
|
<ok to="End"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<end name="End"/>
|
|
</workflow-app> |