forked from D-Net/dnet-hadoop
105 lines
4.1 KiB
XML
105 lines
4.1 KiB
XML
<workflow-app name="Create EBI Dataset" xmlns="uri:oozie:workflow:0.5">
|
|
<parameters>
|
|
<property>
|
|
<name>sourcePath</name>
|
|
<description>the Working Path</description>
|
|
</property>
|
|
<property>
|
|
<name>workingPath</name>
|
|
<description>the Working Path</description>
|
|
</property>
|
|
<property>
|
|
<name>targetPath</name>
|
|
<description>the OAF MDStore Path</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>resumeFrom</name>
|
|
<value>DownloadEBILinks</value>
|
|
<description>node to start</description>
|
|
</property>
|
|
</parameters>
|
|
|
|
<start to="resume_from"/>
|
|
|
|
<decision name="resume_from">
|
|
<switch>
|
|
<case to="DownloadEBILinks">${wf:conf('resumeFrom') eq 'DownloadEBILinks'}</case>
|
|
<case to="CreateEBIDataSet">${wf:conf('resumeFrom') eq 'CreateEBIDataSet'}</case>
|
|
<default to="DownloadEBILinks"/>
|
|
</switch>
|
|
</decision>
|
|
|
|
|
|
<kill name="Kill">
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
</kill>
|
|
|
|
|
|
<action name="DownloadEBILinks">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn-cluster</master>
|
|
<mode>cluster</mode>
|
|
<name>Incremental Download EBI Links</name>
|
|
<class>eu.dnetlib.dhp.sx.bio.ebi.SparkDownloadEBILinks</class>
|
|
<jar>dhp-aggregation-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--executor-cores=${sparkExecutorCores}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
--conf spark.sql.shuffle.partitions=2000
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
|
|
<arg>--workingPath</arg><arg>${workingPath}</arg>
|
|
<arg>--master</arg><arg>yarn</arg>
|
|
</spark>
|
|
<ok to="OverrideFolders"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
<action name="OverrideFolders">
|
|
<fs>
|
|
<delete path="${sourcePath}/ebi_links_dataset_old"/>
|
|
<move source="${sourcePath}/ebi_links_dataset" target="${sourcePath}/ebi_links_dataset_old"/>
|
|
<move source="${workingPath}/links_final" target="${sourcePath}/ebi_links_dataset"/>
|
|
</fs>
|
|
<ok to="CreateEBIDataSet"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
<action name="CreateEBIDataSet">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn-cluster</master>
|
|
<mode>cluster</mode>
|
|
<name>Create OAF DataSet</name>
|
|
<class>eu.dnetlib.dhp.sx.bio.ebi.SparkEBILinksToOaf</class>
|
|
<jar>dhp-aggregation-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--executor-cores=${sparkExecutorCores}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.sql.shuffle.partitions=2000
|
|
${sparkExtraOPT}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${sourcePath}/ebi_links_dataset</arg>
|
|
<arg>--targetPath</arg><arg>${targetPath}</arg>
|
|
<arg>--master</arg><arg>yarn</arg>
|
|
</spark>
|
|
<ok to="End"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
<end name="End"/>
|
|
</workflow-app> |