55 lines
2.0 KiB
XML
55 lines
2.0 KiB
XML
<workflow-app name="Gen_Doi_Author_List_WF" xmlns="uri:oozie:workflow:0.5">
|
|
<parameters>
|
|
<property>
|
|
<name>workingPath</name>
|
|
<description>the working dir base 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>
|
|
<property>
|
|
<name>sparkExecutorCores</name>
|
|
<description>number of cores used by single executor</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}/doi_author_list'/>
|
|
</fs>
|
|
<ok to="Gen_Doi_Author_List"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="Gen_Doi_Author_List">
|
|
<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_Doi_Author_List</name>
|
|
<class>eu.dnetlib.doiboost.orcid.SparkGenerateDoiAuthorList</class>
|
|
<jar>dhp-doiboost-1.2.1-SNAPSHOT.jar</jar>
|
|
<spark-opts>--num-executors 10 --conf spark.yarn.jars="hdfs://hadoop-rm1.garr-pa1.d4science.org:8020/user/oozie/share/lib/lib_20180405103059/spark2" --executor-memory=${sparkExecutorMemory} --executor-cores=${sparkExecutorCores} --driver-memory=${sparkDriverMemory}
|
|
</spark-opts>
|
|
<arg>-w</arg><arg>${workingPath}/</arg>
|
|
<arg>-o</arg><arg>doi_author_list/</arg>
|
|
</spark>
|
|
<ok to="End"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<end name="End"/>
|
|
</workflow-app> |