282 lines
12 KiB
XML
282 lines
12 KiB
XML
<workflow-app name="dump_graph_csv" xmlns="uri:oozie:workflow:0.5">
|
|
<parameters>
|
|
<property>
|
|
<name>sourcePath</name>
|
|
<description>the source path</description>
|
|
</property>
|
|
<property>
|
|
<name>outputPath</name>
|
|
<description>the output path</description>
|
|
</property>
|
|
<property>
|
|
<name>communities</name>
|
|
<description>the communities whose products should be dumped</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>oozieActionShareLibForSpark2</name>
|
|
<description>oozie action sharelib for spark 2.*</description>
|
|
</property>
|
|
<property>
|
|
<name>spark2ExtraListeners</name>
|
|
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
|
|
<description>spark 2.* extra listeners classname</description>
|
|
</property>
|
|
<property>
|
|
<name>spark2SqlQueryExecutionListeners</name>
|
|
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
|
|
<description>spark 2.* sql query execution listeners classname</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>
|
|
<global>
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
<name-node>${nameNode}</name-node>
|
|
<configuration>
|
|
<property>
|
|
<name>mapreduce.job.queuename</name>
|
|
<value>${queueName}</value>
|
|
</property>
|
|
<property>
|
|
<name>oozie.launcher.mapred.job.queue.name</name>
|
|
<value>${oozieLauncherQueueName}</value>
|
|
</property>
|
|
<property>
|
|
<name>oozie.action.sharelib.for.spark</name>
|
|
<value>${oozieActionShareLibForSpark2}</value>
|
|
</property>
|
|
</configuration>
|
|
</global>
|
|
<start to="reset_outputpath" />
|
|
|
|
<action name="reset_outputpath">
|
|
<fs>
|
|
<delete path="${outputPath}"/>
|
|
<mkdir path="${outputPath}"/>
|
|
</fs>
|
|
<ok to="dump_communities"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="dump_communities">
|
|
<java>
|
|
<main-class>eu.dnetlib.dhp.oa.graph.dump.csv.DumpCommunities</main-class>
|
|
<arg>--outputPath</arg><arg>${outputPath}/community</arg>
|
|
<arg>--nameNode</arg><arg>${nameNode}</arg>
|
|
<arg>--communities</arg><arg>${communities}</arg>
|
|
</java>
|
|
<ok to="select_result_dump_relation"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
|
|
<action name="select_result_dump_relation">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>select results ids connected to communities and dump relation </name>
|
|
<class>eu.dnetlib.dhp.oa.graph.dump.csv.SparkSelectResultsAndDumpRelations</class>
|
|
<jar>dump-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-memory=${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>--sourcePath</arg><arg>${sourcePath}</arg>
|
|
<arg>--workingPath</arg><arg>${outputPath}/workingDir</arg>
|
|
<arg>--outputPath</arg><arg>${outputPath}</arg>
|
|
<arg>--communities</arg><arg>${communities}</arg>
|
|
</spark>
|
|
<ok to="fork_dump_result_author_pid"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<fork name="fork_dump_result_author_pid">
|
|
<path start="dump_publication"/>
|
|
<path start="dump_dataset"/>
|
|
<path start="dump_other"/>
|
|
<path start="dump_software"/>
|
|
</fork>
|
|
|
|
|
|
<action name="dump_publication">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>select results from publication </name>
|
|
<class>eu.dnetlib.dhp.oa.graph.dump.csv.SparkDumpResults</class>
|
|
<jar>dump-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-memory=9G
|
|
--executor-cores=${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=${sparkSqlWarehouseDir}
|
|
--conf spark.sql.shuffle.partitions=3840
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
|
|
<arg>--workingPath</arg><arg>${outputPath}/workingDir</arg>
|
|
<arg>--resultType</arg><arg>publication</arg>
|
|
</spark>
|
|
<ok to="join_dump"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
<action name="dump_dataset">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>select results from dataset </name>
|
|
<class>eu.dnetlib.dhp.oa.graph.dump.csv.SparkDumpResults</class>
|
|
<jar>dump-${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=${sparkSqlWarehouseDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
|
|
<arg>--workingPath</arg><arg>${outputPath}/workingDir</arg>
|
|
<arg>--resultType</arg><arg>dataset</arg>
|
|
</spark>
|
|
<ok to="join_dump"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
<action name="dump_other">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>select results from other </name>
|
|
<class>eu.dnetlib.dhp.oa.graph.dump.csv.SparkDumpResults</class>
|
|
<jar>dump-${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=${sparkSqlWarehouseDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
|
|
<arg>--workingPath</arg><arg>${outputPath}/workingDir</arg>
|
|
<arg>--resultType</arg><arg>otherresearchproduct</arg>
|
|
</spark>
|
|
<ok to="join_dump"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
<action name="dump_software">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>select results from software</name>
|
|
<class>eu.dnetlib.dhp.oa.graph.dump.csv.SparkDumpResults</class>
|
|
<jar>dump-${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=${sparkSqlWarehouseDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
|
|
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
|
|
<arg>--workingPath</arg><arg>${outputPath}/workingDir</arg>
|
|
<arg>--resultType</arg><arg>software</arg>
|
|
</spark>
|
|
<ok to="join_dump"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<join name="join_dump" to="dump_single_results"/>
|
|
|
|
<kill name="Kill">
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
</kill>
|
|
<action name="dump_single_results">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Dump single results </name>
|
|
<class>eu.dnetlib.dhp.oa.graph.dump.csv.SparkMoveOnSigleDir</class>
|
|
<jar>dump-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-memory=${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>--workingPath</arg><arg>${outputPath}/workingDir</arg>
|
|
<arg>--outputPath</arg><arg>${outputPath}</arg>
|
|
</spark>
|
|
<ok to="End"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="make_archive">
|
|
<java>
|
|
<main-class>eu.dnetlib.dhp.oa.graph.dump.MakeTar</main-class>
|
|
<arg>--hdfsPath</arg><arg>${outputPath}</arg>
|
|
<arg>--nameNode</arg><arg>${nameNode}</arg>
|
|
<arg>--sourcePath</arg><arg>${workingDir}/tar</arg>
|
|
</java>
|
|
<ok to="send_zenodo"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
<action name="send_zenodo">
|
|
<java>
|
|
<main-class>eu.dnetlib.dhp.oa.graph.dump.SendToZenodoHDFS</main-class>
|
|
<arg>--hdfsPath</arg><arg>${outputPath}</arg>
|
|
<arg>--nameNode</arg><arg>${nameNode}</arg>
|
|
<arg>--accessToken</arg><arg>${accessToken}</arg>
|
|
<arg>--connectionUrl</arg><arg>${connectionUrl}</arg>
|
|
<arg>--metadata</arg><arg>${metadata}</arg>
|
|
<arg>--conceptRecordId</arg><arg>${conceptRecordId}</arg>
|
|
<arg>--depositionType</arg><arg>${depositionType}</arg>
|
|
<arg>--depositionId</arg><arg>${depositionId}</arg>
|
|
</java>
|
|
<ok to="End"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
<end name="End"/>
|
|
</workflow-app> |