dnet-hadoop/dhp-workflows/dhp-graph-provision-scholex.../src/main/resources/eu/dnetlib/dhp/sx/index/oozie_app/workflow.xml

86 lines
3.4 KiB
XML

<workflow-app name="Materialize and Index graph to ElasticSearch" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>workingDirPath</name>
<description>the source path</description>
</property>
<property>
<name>index</name>
<description>the index name</description>
</property>
<property>
<name>esCluster</name>
<description>the Index cluster</description>
</property>
<property>
<name>sparkDriverMemory</name>
<description>memory for driver process</description>
</property>
<property>
<name>sparkExecutorMemory</name>
<description>memory for individual executor</description>
</property>
</parameters>
<start to="DropAndCreateIndex"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="DropAndCreateIndex">
<java>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<main-class>eu.dnetlib.dhp.provision.DropAndCreateESIndex</main-class>
<arg>-i</arg><arg>${index}</arg>
<arg>-c</arg><arg>${esCluster}</arg>
</java>
<ok to="indexSummary"/>
<error to="Kill"/>
</action>
<action name="indexSummary">
<spark xmlns="uri:oozie:spark-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>index summary</name>
<class>eu.dnetlib.dhp.provision.SparkIndexCollectionOnES</class>
<jar>dhp-graph-provision-scholexplorer-${projectVersion}.jar</jar>
<spark-opts>--executor-memory ${sparkExecutorMemory} --driver-memory=${sparkDriverMemory} ${sparkExtraOPT} --conf spark.dynamicAllocation.maxExecutors="8" </spark-opts>
<arg>-mt</arg> <arg>yarn-cluster</arg>
<arg>--sourcePath</arg><arg>${workingDirPath}/summary_json</arg>
<arg>--index</arg><arg>${index}_object</arg>
<arg>--idPath</arg><arg>id</arg>
<arg>--cluster</arg><arg>${esCluster}</arg>
</spark>
<ok to="indexScholix"/>
<error to="Kill"/>
</action>
<action name="indexScholix">
<spark xmlns="uri:oozie:spark-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>index scholix</name>
<class>eu.dnetlib.dhp.provision.SparkIndexCollectionOnES</class>
<jar>dhp-graph-provision-scholexplorer-${projectVersion}.jar</jar>
<spark-opts>--executor-memory ${sparkExecutorMemory} --driver-memory=${sparkDriverMemory} ${sparkExtraOPT} --conf spark.dynamicAllocation.maxExecutors="8" </spark-opts>
<arg>-mt</arg> <arg>yarn-cluster</arg>
<arg>--sourcePath</arg><arg>${workingDirPath}/scholix_json</arg>
<arg>--index</arg><arg>${index}_scholix</arg>
<arg>--idPath</arg><arg>identifier</arg>
<arg>--cluster</arg><arg>${esCluster}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>