dnet-hadoop/dhp-workflows/dhp-dedup/src/main/resources/eu/dnetlib/dhp/dedup/roots/oozie_app/workflow.xml

<workflow-app name="Build Root Records" xmlns="uri:oozie:workflow:0.5">
    <parameters>
        <property>
            <name>graphBasePath</name>
            <description>the raw graph base path</description>
        </property>
        <property>
            <name>isLookUpUrl</name>
            <description>the address of the lookUp service</description>
        </property>
        <property>
            <name>actionSetId</name>
            <description>id of the actionSet</description>
        </property>
        <property>
            <name>workingPath</name>
            <description>path of the working directory</description>
        </property>
        <property>
            <name>dedupGraphPath</name>
            <description>path of the dedup graph</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>
    </parameters>

    <start to="DeleteWorkingPath"/>

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

    <action name="DeleteWorkingPath">
        <fs>
            <delete path='${workingPath}/${actionSetId}/*_mergerel'/>
            <delete path='${workingPath}/${actionSetId}/*_deduprecord'/>
        </fs>
        <ok to="CreateMergeRel"/>
        <error to="Kill"/>
    </action>

    <action name="CreateMergeRel">
        <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>Create Merge Relations</name>
            <class>eu.dnetlib.dhp.dedup.SparkCreateConnectedComponent</class>
            <jar>dhp-dedup-${projectVersion}.jar</jar>
            <spark-opts>--executor-memory ${sparkExecutorMemory} --executor-cores ${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory} --conf
                spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener" --conf
                spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener" --conf
                spark.sql.warehouse.dir="/user/hive/warehouse"
            </spark-opts>
            <arg>-mt</arg><arg>yarn-cluster</arg>
            <arg>--i</arg><arg>${graphBasePath}</arg>
            <arg>--w</arg><arg>${workingPath}</arg>
            <arg>--la</arg><arg>${isLookUpUrl}</arg>
            <arg>--asi</arg><arg>${actionSetId}</arg>
        </spark>
        <ok to="CreateDedupRecord"/>
        <error to="Kill"/>
    </action>

    <action name="CreateDedupRecord">
        <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>Create Dedup Record</name>
            <class>eu.dnetlib.dhp.dedup.SparkCreateDedupRecord</class>
            <jar>dhp-dedup-${projectVersion}.jar</jar>
            <spark-opts>--executor-memory ${sparkExecutorMemory} --executor-cores ${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory} --conf
                spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener" --conf
                spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener" --conf
                spark.sql.warehouse.dir="/user/hive/warehouse"
            </spark-opts>
            <arg>-mt</arg><arg>yarn-cluster</arg>
            <arg>--i</arg><arg>${graphBasePath}</arg>
            <arg>--w</arg><arg>${workingPath}</arg>
            <arg>--la</arg><arg>${isLookUpUrl}</arg>
            <arg>--asi</arg><arg>${actionSetId}</arg>
        </spark>
        <ok to="UpdateEntity"/>
        <error to="Kill"/>
    </action>

    <action name="UpdateEntity">
        <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>Create Dedup Record</name>
            <class>eu.dnetlib.dhp.dedup.SparkUpdateEntity</class>
            <jar>dhp-dedup-${projectVersion}.jar</jar>
            <spark-opts>--executor-memory ${sparkExecutorMemory} --executor-cores ${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory} --conf
                spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener" --conf
                spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener" --conf
                spark.sql.warehouse.dir="/user/hive/warehouse"
            </spark-opts>
            <arg>-mt</arg><arg>yarn-cluster</arg>
            <arg>--i</arg><arg>${graphBasePath}</arg>
            <arg>--w</arg><arg>${workingPath}</arg>
            <arg>--la</arg><arg>${isLookUpUrl}</arg>
            <arg>--asi</arg><arg>${actionSetId}</arg>
            <arg>--o</arg><arg>${dedupGraphPath}</arg>
        </spark>
        <ok to="End"/>
        <error to="Kill"/>
    </action>

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