forked from D-Net/dnet-hadoop
99 lines
4.2 KiB
XML
99 lines
4.2 KiB
XML
<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>
|
|
<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>
|
|
</parameters>
|
|
|
|
<start to="reuse_records"/>
|
|
|
|
<decision name="reuse_records">
|
|
<switch>
|
|
<case to="adjancency_lists">${wf:conf('reuseRecords') eq false}</case>
|
|
<case to="to_solr_index">${wf:conf('reuseRecords') eq true}</case>
|
|
<default to="adjancency_lists"/>
|
|
</switch>
|
|
</decision>
|
|
|
|
<kill name="Kill">
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
</kill>
|
|
|
|
<action name="adjancency_lists">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
<name-node>${nameNode}</name-node>
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>build_adjacency_lists</name>
|
|
<class>eu.dnetlib.dhp.graph.SparkXmlRecordBuilderJob</class>
|
|
<jar>dhp-graph-provision-${projectVersion}.jar</jar>
|
|
<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.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
</spark-opts>
|
|
<arg>-mt</arg> <arg>yarn</arg>
|
|
<arg>-is</arg> <arg>${isLookupUrl}</arg>
|
|
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
|
|
<arg>--outputPath</arg><arg>${outputPath}</arg>
|
|
</spark>
|
|
<ok to="to_solr_index"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="to_solr_index">
|
|
<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>to_solr_index</name>
|
|
<class>eu.dnetlib.dhp.graph.SparkXmlIndexingJob</class>
|
|
<jar>dhp-graph-provision-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-memory ${sparkExecutorMemory}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.dynamicAllocation.maxExecutors=${sparkExecutorCoresForIndexing}
|
|
--conf spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener"
|
|
--conf spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener"
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
</spark-opts>
|
|
<arg>-mt</arg> <arg>yarn</arg>
|
|
<arg>-is</arg> <arg>${isLookupUrl}</arg>
|
|
<arg>--sourcePath</arg><arg>${outputPath}/xml</arg>
|
|
<arg>--format</arg><arg>${format}</arg>
|
|
<arg>--batchSize</arg><arg>${batchSize}</arg>
|
|
</spark>
|
|
<ok to="End"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<end name="End"/>
|
|
</workflow-app> |