dnet-hadoop/dhp-workflows/dhp-propagation/src/main/resources/eu/dnetlib/dhp/countrypropagation/oozie_app/workflow.xml

81 lines
3.3 KiB
XML
Raw Normal View History

2020-02-17 18:04:15 +01:00
<workflow-app name="country_propagation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
2020-02-17 18:04:15 +01:00
<name>whitelist</name>
<description>the white list</description>
</property>
<property>
<name>allowedtypes</name>
<description>the allowed types</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>
2020-04-08 10:31:29 +02:00
<property>
<name>sparkExecutorNumber</name>
<description>number of executors used</description>
</property>
<property>
<name>writeUpdate</name>
<description>writes the information found for the update. No double check done if the information is already present</description>
</property>
<property>
<name>saveGraph</name>
<description>writes new version of the graph after the propagation step</description>
</property>
</parameters>
2020-02-17 18:04:15 +01:00
<start to="CountryPropagation"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="CountryPropagation">
<spark xmlns="uri:oozie:spark-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>CountryPropagation</name>
<class>eu.dnetlib.dhp.countrypropagation.SparkCountryPropagationJob</class>
2020-02-17 18:04:15 +01:00
<jar>dhp-propagation-${projectVersion}.jar</jar>
2020-04-08 10:31:29 +02:00
<spark-opts>
--num-executors=${sparkExecutorNumber}
--executor-memory=${sparkExecutorMemory}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
2020-04-08 10:31:29 +02:00
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
2020-02-17 18:04:15 +01:00
</spark-opts>
<arg>-mt</arg> <arg>yarn-cluster</arg>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
2020-02-18 17:23:34 +01:00
<arg>--whitelist</arg><arg>${whitelist}</arg>
2020-02-17 18:04:15 +01:00
<arg>--allowedtypes</arg><arg>${allowedtypes}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
2020-04-08 10:31:29 +02:00
<arg>--writeUpdate</arg><arg>${writeUpdate}</arg>
<arg>--saveGraph</arg><arg>${saveGraph}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>