forked from D-Net/dnet-hadoop
added relations to the graph2hive import workflow
This commit is contained in:
parent
9d028ffe1c
commit
18f46e47b9
|
@ -102,6 +102,7 @@
|
|||
<path start="import_datasource"/>
|
||||
<path start="import_organization"/>
|
||||
<path start="import_project"/>
|
||||
<path start="import_relation"/>
|
||||
</fork>
|
||||
|
||||
<action name="import_publication">
|
||||
|
@ -286,6 +287,32 @@
|
|||
<error to="Kill"/>
|
||||
</action>
|
||||
|
||||
<action name="import_relation">
|
||||
<spark xmlns="uri:oozie:spark-action:0.2">
|
||||
<master>yarn</master>
|
||||
<mode>cluster</mode>
|
||||
<name>Import table project</name>
|
||||
<class>eu.dnetlib.dhp.oa.graph.hive.GraphHiveTableImporterJob</class>
|
||||
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
|
||||
<spark-opts>
|
||||
--executor-memory=${sparkExecutorMemory}
|
||||
--executor-cores=${sparkExecutorCores}
|
||||
--driver-memory=${sparkDriverMemory}
|
||||
--conf spark.extraListeners=${spark2ExtraListeners}
|
||||
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
||||
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
||||
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
||||
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
|
||||
</spark-opts>
|
||||
<arg>--inputPath</arg><arg>${inputPath}/relation</arg>
|
||||
<arg>--hiveDbName</arg><arg>${hiveDbName}</arg>
|
||||
<arg>--className</arg><arg>eu.dnetlib.dhp.schema.oaf.Relation</arg>
|
||||
<arg>--hiveMetastoreUris</arg><arg>${hiveMetastoreUris}</arg>
|
||||
</spark>
|
||||
<ok to="join_import"/>
|
||||
<error to="Kill"/>
|
||||
</action>
|
||||
|
||||
<join name="join_import" to="PostProcessing"/>
|
||||
|
||||
<action name="PostProcessing">
|
||||
|
|
Loading…
Reference in New Issue