dnet-hadoop/dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/orcid_gen_authors/oozie_app/workflow.xml

83 lines
3.3 KiB
XML

<workflow-app name="Gen Orcid Authors" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>workingPath</name>
<description>the working dir base path</description>
</property>
<property>
<name>token</name>
<description>access token</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>outputPath</name>
<description>the working dir base path</description>
</property>
</parameters>
<start to="ResetWorkingPath"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="ResetWorkingPath">
<fs>
<delete path='${workingPath_activities}/authors'/>
</fs>
<ok to="Gen_Orcid_Authors"/>
<error to="Kill"/>
</action>
<action name="Split_Lambda_File">
<spark xmlns="uri:oozie:spark-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<master>yarn</master>
<mode>cluster</mode>
<name>Split_Lambda_File</name>
<class>eu.dnetlib.doiboost.orcid.SparkPartitionLambdaFile</class>
<jar>dhp-doiboost-1.2.1-SNAPSHOT.jar</jar>
<spark-opts>--num-executors 24 --conf spark.yarn.jars=&quot;hdfs://hadoop-rm1.garr-pa1.d4science.org:8020/user/oozie/share/lib/lib_20180405103059/spark2&quot; --executor-memory=${sparkExecutorMemory} --executor-cores=${sparkExecutorCores} --driver-memory=${sparkDriverMemory}
</spark-opts>
<arg>-w</arg><arg>${workingPath}/</arg>
<arg>-o</arg><arg>authors/</arg>
<arg>-t</arg><arg>${token}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<action name="Gen_Orcid_Authors">
<spark xmlns="uri:oozie:spark-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<master>yarn</master>
<mode>cluster</mode>
<name>Gen_Orcid_Authors</name>
<class>eu.dnetlib.doiboost.orcid.SparkOrcidGenerateAuthors</class>
<jar>dhp-doiboost-1.2.1-SNAPSHOT.jar</jar>
<spark-opts>--num-executors 20 --conf spark.yarn.jars=&quot;hdfs://hadoop-rm1.garr-pa1.d4science.org:8020/user/oozie/share/lib/lib_20180405103059/spark2&quot; --executor-memory=${sparkExecutorMemory} --executor-cores=${sparkExecutorCores} --driver-memory=${sparkDriverMemory}
</spark-opts>
<arg>-w</arg><arg>${workingPath}/</arg>
<arg>-o</arg><arg>authors/</arg>
<arg>-t</arg><arg>${token}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>