2019-12-17 16:24:49 +01:00
|
|
|
<workflow-app name="index_infospace_graph" xmlns="uri:oozie:workflow:0.5">
|
|
|
|
<parameters>
|
|
|
|
<property>
|
|
|
|
<name>hive_db_name</name>
|
|
|
|
<description>the target hive database name</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>
|
2020-01-21 16:35:44 +01:00
|
|
|
<property>
|
|
|
|
<name>spark2YarnHistoryServerAddress</name>
|
|
|
|
<description>spark 2.* yarn history server address</description>
|
|
|
|
</property>
|
|
|
|
<property>
|
|
|
|
<name>spark2EventLogDir</name>
|
|
|
|
<description>spark 2.* event log dir location</description>
|
|
|
|
</property>
|
2019-12-17 16:24:49 +01:00
|
|
|
</parameters>
|
|
|
|
|
2020-01-30 17:46:29 +01:00
|
|
|
<start to="adjancency_lists"/>
|
2019-12-17 16:24:49 +01:00
|
|
|
|
|
|
|
<kill name="Kill">
|
|
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
|
|
</kill>
|
|
|
|
|
2020-01-30 17:46:29 +01:00
|
|
|
<action name="adjancency_lists">
|
2019-12-17 16:24:49 +01:00
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
|
|
<name-node>${nameNode}</name-node>
|
|
|
|
<master>yarn-cluster</master>
|
|
|
|
<mode>cluster</mode>
|
2020-01-30 17:46:29 +01:00
|
|
|
<name>build_adjacency_lists</name>
|
|
|
|
<class>eu.dnetlib.dhp.graph.SparkXmlRecordBuilderJob</class>
|
2019-12-17 16:24:49 +01:00
|
|
|
<jar>dhp-graph-provision-${projectVersion}.jar</jar>
|
2020-01-21 16:35:44 +01:00
|
|
|
<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"
|
|
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
|
|
</spark-opts>
|
2019-12-17 16:24:49 +01:00
|
|
|
<arg>-mt</arg> <arg>yarn-cluster</arg>
|
2020-01-16 11:02:28 +01:00
|
|
|
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
|
2020-01-30 17:46:29 +01:00
|
|
|
<arg>--outputPath</arg><arg>${outputPath}</arg>
|
2019-12-17 16:24:49 +01:00
|
|
|
</spark>
|
|
|
|
<ok to="End"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<end name="End"/>
|
|
|
|
</workflow-app>
|