dnet-hadoop/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/step1/oozie_app/workflow.xml

<workflow-app name="Create Raw Graph Step 1: extract Entities in raw graph" xmlns="uri:oozie:workflow:0.5">
    <parameters>
        <property>
            <name>sourcePath</name>
            <description>the working dir base path</description>
        </property>
        <property>
            <name>targetPath</name>
            <description>the graph Raw base path</description>
        </property>
    </parameters>

    <start to="CreateScholix"/>

    <kill name="Kill">
        <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
    </kill>

    <action name="ExtractEntities">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Extract entities in raw graph</name>
            <class>eu.dnetlib.dhp.sx.graph.SparkCreateInputGraph</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.shuffle.partitions=2000
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
            </spark-opts>
            <arg>--master</arg><arg>yarn</arg>
            <arg>--sourcePath</arg><arg>${sourcePath}</arg>
            <arg>--targetPath</arg><arg>${targetPath}</arg>
        </spark>
        <ok to="ResolveRelations"/>
        <error to="Kill"/>
    </action>


    <action name="ResolveRelations">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Resolve Relations in raw graph</name>
            <class>eu.dnetlib.dhp.sx.graph.SparkResolveRelation</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.shuffle.partitions=3000
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
            </spark-opts>
            <arg>--master</arg><arg>yarn</arg>
            <arg>--relationPath</arg><arg>${targetPath}/extracted/relation</arg>
            <arg>--workingPath</arg><arg>${targetPath}/resolved/</arg>
            <arg>--entityPath</arg><arg>${targetPath}/dedup</arg>
        </spark>
        <ok to="CreateSummaries"/>
        <error to="Kill"/>
    </action>


    <action name="CreateSummaries">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Convert Entities to summaries</name>
            <class>eu.dnetlib.dhp.sx.graph.SparkCreateSummaryObject</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.shuffle.partitions=5000
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
            </spark-opts>
            <arg>--master</arg><arg>yarn</arg>
            <arg>--sourcePath</arg><arg>${targetPath}/dedup</arg>
            <arg>--targetPath</arg><arg>${targetPath}/provision/summaries</arg>
        </spark>
        <ok to="CreateScholix"/>
        <error to="Kill"/>
    </action>

    <action name="CreateScholix">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Generate Scholix Dataset</name>
            <class>eu.dnetlib.dhp.sx.graph.SparkCreateScholix</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.shuffle.partitions=6000
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
            </spark-opts>
            <arg>--master</arg><arg>yarn</arg>
            <arg>--summaryPath</arg><arg>${targetPath}/provision/summaries</arg>
            <arg>--targetPath</arg><arg>${targetPath}/provision/scholix</arg>
            <arg>--relationPath</arg><arg>${targetPath}/resolved/resolvedRelation</arg>

        </spark>
        <ok to="End"/>
        <error to="Kill"/>
    </action>




    <end name="End"/>
</workflow-app>