2023-12-07 09:59:52 +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>
|
|
|
|
<name>whitelist</name>
|
|
|
|
<description>the white list</description>
|
|
|
|
</property>
|
|
|
|
<property>
|
|
|
|
<name>allowedtypes</name>
|
|
|
|
<description>the allowed types</description>
|
|
|
|
</property>
|
|
|
|
</parameters>
|
|
|
|
|
|
|
|
<global>
|
|
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
|
|
<name-node>${nameNode}</name-node>
|
|
|
|
<configuration>
|
|
|
|
<property>
|
|
|
|
<name>oozie.action.sharelib.for.spark</name>
|
|
|
|
<value>${oozieActionShareLibForSpark2}</value>
|
|
|
|
</property>
|
|
|
|
</configuration>
|
|
|
|
</global>
|
|
|
|
|
2023-12-20 14:26:54 +01:00
|
|
|
<start to="reset_outputpath"/>
|
2023-12-07 09:59:52 +01:00
|
|
|
|
|
|
|
<kill name="Kill">
|
|
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
|
|
</kill>
|
|
|
|
|
2023-12-20 14:26:54 +01:00
|
|
|
<action name="reset_outputpath">
|
|
|
|
<fs>
|
|
|
|
<delete path="${workingDir}"/>
|
|
|
|
<mkdir path="${workingDir}"/>
|
|
|
|
</fs>
|
|
|
|
<ok to="prepare_datasource_country_association"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
2023-12-07 09:59:52 +01:00
|
|
|
|
|
|
|
|
|
|
|
<action name="prepare_datasource_country_association">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
|
|
|
<name>PrepareDatasourceCountryAssociation</name>
|
|
|
|
<class>eu.dnetlib.dhp.countrypropagation.PrepareDatasourceCountryAssociation</class>
|
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--driver-memory=${sparkDriverMemory}
|
|
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
|
|
|
|
<arg>--whitelist</arg><arg>${whitelist}</arg>
|
|
|
|
<arg>--allowedtypes</arg><arg>${allowedtypes}</arg>
|
|
|
|
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
|
|
|
|
</spark>
|
|
|
|
<ok to="fork_prepare_result_country"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<fork name="fork_prepare_result_country">
|
|
|
|
<path start="prepareresult_publication"/>
|
|
|
|
<path start="prepareresult_dataset"/>
|
|
|
|
<path start="prepareresult_otherresearchproduct"/>
|
|
|
|
<path start="prepareresult_software"/>
|
|
|
|
</fork>
|
|
|
|
|
|
|
|
<action name="prepareresult_publication">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
|
|
|
<name>prepareResultCountry-Publication</name>
|
|
|
|
<class>eu.dnetlib.dhp.countrypropagation.PrepareResultCountrySet</class>
|
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--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.speculation=false
|
|
|
|
--conf spark.hadoop.mapreduce.map.speculative=false
|
|
|
|
--conf spark.hadoop.mapreduce.reduce.speculative=false
|
|
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
|
|
|
|
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
|
|
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
|
|
|
|
</spark>
|
|
|
|
<ok to="wait_prepare"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<action name="prepareresult_dataset">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
|
|
|
<name>prepareResultCountry-Dataset</name>
|
|
|
|
<class>eu.dnetlib.dhp.countrypropagation.PrepareResultCountrySet</class>
|
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--driver-memory=${sparkDriverMemory}
|
|
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
|
|
--conf spark.speculation=false
|
|
|
|
--conf spark.hadoop.mapreduce.map.speculative=false
|
|
|
|
--conf spark.hadoop.mapreduce.reduce.speculative=false
|
|
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
|
|
|
|
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
|
|
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
|
|
|
|
</spark>
|
|
|
|
<ok to="wait_prepare"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<action name="prepareresult_otherresearchproduct">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
|
|
|
<name>prepareResultCountry-ORP</name>
|
|
|
|
<class>eu.dnetlib.dhp.countrypropagation.PrepareResultCountrySet</class>
|
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--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.speculation=false
|
|
|
|
--conf spark.hadoop.mapreduce.map.speculative=false
|
|
|
|
--conf spark.hadoop.mapreduce.reduce.speculative=false
|
|
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
|
|
|
|
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
|
|
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
|
|
|
|
</spark>
|
|
|
|
<ok to="wait_prepare"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<action name="prepareresult_software">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
|
|
|
<name>prepareResultCountry-Software</name>
|
|
|
|
<class>eu.dnetlib.dhp.countrypropagation.PrepareResultCountrySet</class>
|
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--driver-memory=${sparkDriverMemory}
|
|
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
|
|
--conf spark.speculation=false
|
|
|
|
--conf spark.hadoop.mapreduce.map.speculative=false
|
|
|
|
--conf spark.hadoop.mapreduce.reduce.speculative=false
|
|
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
|
|
|
|
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
|
|
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
|
|
|
|
</spark>
|
|
|
|
<ok to="wait_prepare"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<join name="wait_prepare" to="fork_join_apply_country_propagation"/>
|
|
|
|
|
|
|
|
<fork name="fork_apply_country_propagation">
|
|
|
|
<path start="propagation_publication"/>
|
|
|
|
<path start="propagation_dataset"/>
|
|
|
|
<path start="propagation_otherresearchproduct"/>
|
|
|
|
<path start="propagation_software"/>
|
|
|
|
</fork>
|
|
|
|
|
|
|
|
<action name="propagation_publication">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
|
|
|
<name>countryPropagationForPublications</name>
|
|
|
|
<class>eu.dnetlib.dhp.countrypropagation.SparkCountryPropagationJob</class>
|
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--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.speculation=false
|
|
|
|
--conf spark.hadoop.mapreduce.map.speculative=false
|
|
|
|
--conf spark.hadoop.mapreduce.reduce.speculative=false
|
|
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
|
|
|
|
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
|
|
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
|
|
|
|
|
|
|
|
</spark>
|
|
|
|
<ok to="wait"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<action name="propagation_dataset">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
|
|
|
<name>countryPropagationForDataset</name>
|
|
|
|
<class>eu.dnetlib.dhp.countrypropagation.SparkCountryPropagationJob</class>
|
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--driver-memory=${sparkDriverMemory}
|
|
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
|
|
--conf spark.speculation=false
|
|
|
|
--conf spark.hadoop.mapreduce.map.speculative=false
|
|
|
|
--conf spark.hadoop.mapreduce.reduce.speculative=false
|
|
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
|
|
|
|
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
|
|
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
|
|
|
|
|
|
|
|
</spark>
|
|
|
|
<ok to="wait"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<action name="propagation_otherresearchproduct">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
|
|
|
<name>countryPropagationForORP</name>
|
|
|
|
<class>eu.dnetlib.dhp.countrypropagation.SparkCountryPropagationJob</class>
|
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--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.speculation=false
|
|
|
|
--conf spark.hadoop.mapreduce.map.speculative=false
|
|
|
|
--conf spark.hadoop.mapreduce.reduce.speculative=false
|
|
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
|
|
|
|
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
|
|
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
|
|
|
|
|
|
|
|
</spark>
|
|
|
|
<ok to="wait"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<action name="propagation_software">
|
|
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
|
|
<master>yarn</master>
|
|
|
|
<mode>cluster</mode>
|
|
|
|
<name>countryPropagationForSoftware</name>
|
|
|
|
<class>eu.dnetlib.dhp.countrypropagation.SparkCountryPropagationJob</class>
|
|
|
|
<jar>dhp-enrichment-${projectVersion}.jar</jar>
|
|
|
|
<spark-opts>
|
|
|
|
--executor-cores=${sparkExecutorCores}
|
|
|
|
--executor-memory=${sparkExecutorMemory}
|
|
|
|
--driver-memory=${sparkDriverMemory}
|
|
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
|
|
--conf spark.speculation=false
|
|
|
|
--conf spark.hadoop.mapreduce.map.speculative=false
|
|
|
|
--conf spark.hadoop.mapreduce.reduce.speculative=false
|
|
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
|
|
</spark-opts>
|
|
|
|
<arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
|
|
|
|
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
|
|
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
|
|
|
|
</spark>
|
|
|
|
<ok to="wait"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
|
|
|
|
<join name="wait" to="reset_workingDir"/>
|
|
|
|
<action name="reset_workingDir">
|
|
|
|
<fs>
|
|
|
|
<delete path="${workingDir}"/>
|
|
|
|
<mkdir path="${workingDir}"/>
|
|
|
|
</fs>
|
|
|
|
<ok to="End"/>
|
|
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
<end name="End"/>
|
|
|
|
|
|
|
|
</workflow-app>
|