BrBETA_dnet-hadoop/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/oa/graph/hive/oozie_app/workflow.xml

<workflow-app name="import_graph_as_hive_DB" xmlns="uri:oozie:workflow:0.5">

    <parameters>
        <property>
            <name>inputPath</name>
            <description>the source path</description>
        </property>
        <property>
            <name>hiveDbName</name>
            <description>the target hive database name</description>
        </property>
        <property>
            <name>hiveJdbcUrl</name>
            <description>hive server jdbc url</description>
        </property>
        <property>
            <name>hiveMetastoreUris</name>
            <description>hive server metastore URIs</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>
        <property>
            <name>sparkSqlWarehouseDir</name>
            <value>spark 2.* db directory location</value>
        </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_DB"/>

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

    <action name="reset_DB">
        <hive2 xmlns="uri:oozie:hive2-action:0.1">
            <configuration>
                <property>
                    <name>hive.metastore.uris</name>
                    <value>${hiveMetastoreUris}</value>
                </property>
            </configuration>
            <jdbc-url>${hiveJdbcUrl}/${hiveDbName}</jdbc-url>
            <script>lib/scripts/reset_db.sql</script>
            <param>hiveDbName=${hiveDbName}</param>
        </hive2>
        <ok to="fork_import"/>
        <error to="Kill"/>
    </action>

    <fork name="fork_import">
        <path start="import_publication"/>
        <path start="import_dataset"/>
        <path start="import_orp"/>
        <path start="import_software"/>
        <path start="import_datasource"/>
        <path start="import_organization"/>
        <path start="import_project"/>
        <path start="import_relation"/>
    </fork>

    <action name="import_publication">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Import table publication</name>
            <class>eu.dnetlib.dhp.oa.graph.hive.GraphHiveTableImporterJob</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/publication</arg>
            <arg>--hiveDbName</arg><arg>${hiveDbName}</arg>
            <arg>--className</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
            <arg>--hiveMetastoreUris</arg><arg>${hiveMetastoreUris}</arg>
        </spark>
        <ok to="join_import"/>
        <error to="Kill"/>
    </action>

    <action name="import_dataset">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Import table dataset</name>
            <class>eu.dnetlib.dhp.oa.graph.hive.GraphHiveTableImporterJob</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/dataset</arg>
            <arg>--hiveDbName</arg><arg>${hiveDbName}</arg>
            <arg>--className</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
            <arg>--hiveMetastoreUris</arg><arg>${hiveMetastoreUris}</arg>
        </spark>
        <ok to="join_import"/>
        <error to="Kill"/>
    </action>

    <action name="import_orp">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Import table otherresearchproduct</name>
            <class>eu.dnetlib.dhp.oa.graph.hive.GraphHiveTableImporterJob</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/otherresearchproduct</arg>
            <arg>--hiveDbName</arg><arg>${hiveDbName}</arg>
            <arg>--className</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
            <arg>--hiveMetastoreUris</arg><arg>${hiveMetastoreUris}</arg>
        </spark>
        <ok to="join_import"/>
        <error to="Kill"/>
    </action>

    <action name="import_software">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Import table software</name>
            <class>eu.dnetlib.dhp.oa.graph.hive.GraphHiveTableImporterJob</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/software</arg>
            <arg>--hiveDbName</arg><arg>${hiveDbName}</arg>
            <arg>--className</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
            <arg>--hiveMetastoreUris</arg><arg>${hiveMetastoreUris}</arg>
        </spark>
        <ok to="join_import"/>
        <error to="Kill"/>
    </action>

    <action name="import_datasource">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Import table datasource</name>
            <class>eu.dnetlib.dhp.oa.graph.hive.GraphHiveTableImporterJob</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/datasource</arg>
            <arg>--hiveDbName</arg><arg>${hiveDbName}</arg>
            <arg>--className</arg><arg>eu.dnetlib.dhp.schema.oaf.Datasource</arg>
            <arg>--hiveMetastoreUris</arg><arg>${hiveMetastoreUris}</arg>
        </spark>
        <ok to="join_import"/>
        <error to="Kill"/>
    </action>

    <action name="import_organization">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Import table organization</name>
            <class>eu.dnetlib.dhp.oa.graph.hive.GraphHiveTableImporterJob</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/organization</arg>
            <arg>--hiveDbName</arg><arg>${hiveDbName}</arg>
            <arg>--className</arg><arg>eu.dnetlib.dhp.schema.oaf.Organization</arg>
            <arg>--hiveMetastoreUris</arg><arg>${hiveMetastoreUris}</arg>
        </spark>
        <ok to="join_import"/>
        <error to="Kill"/>
    </action>

    <action name="import_project">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Import table project</name>
            <class>eu.dnetlib.dhp.oa.graph.hive.GraphHiveTableImporterJob</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/project</arg>
            <arg>--hiveDbName</arg><arg>${hiveDbName}</arg>
            <arg>--className</arg><arg>eu.dnetlib.dhp.schema.oaf.Project</arg>
            <arg>--hiveMetastoreUris</arg><arg>${hiveMetastoreUris}</arg>
            <arg>--numPartitions</arg><arg>100</arg>
        </spark>
        <ok to="join_import"/>
        <error to="Kill"/>
    </action>

    <action name="import_relation">
        <spark xmlns="uri:oozie:spark-action:0.2">
            <master>yarn</master>
            <mode>cluster</mode>
            <name>Import table project</name>
            <class>eu.dnetlib.dhp.oa.graph.hive.GraphHiveTableImporterJob</class>
            <jar>dhp-graph-mapper-${projectVersion}.jar</jar>
            <spark-opts>
                --executor-memory=${sparkExecutorMemory}
                --executor-cores=${sparkExecutorCores}
                --driver-memory=${sparkDriverMemory}
                --conf spark.extraListeners=${spark2ExtraListeners}
                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
            </spark-opts>
            <arg>--inputPath</arg><arg>${inputPath}/relation</arg>
            <arg>--hiveDbName</arg><arg>${hiveDbName}</arg>
            <arg>--className</arg><arg>eu.dnetlib.dhp.schema.oaf.Relation</arg>
            <arg>--hiveMetastoreUris</arg><arg>${hiveMetastoreUris}</arg>
        </spark>
        <ok to="join_import"/>
        <error to="Kill"/>
    </action>

    <join name="join_import" to="PostProcessing"/>

    <action name="PostProcessing">
        <hive2 xmlns="uri:oozie:hive2-action:0.1">
            <configuration>
                <property>
                    <name>hive.metastore.uris</name>
                    <value>${hiveMetastoreUris}</value>
                </property>
            </configuration>
            <jdbc-url>${hiveJdbcUrl}/${hiveDbName}</jdbc-url>
            <script>lib/scripts/postprocessing.sql</script>
            <param>hiveDbName=${hiveDbName}</param>
        </hive2>
        <ok to="End"/>
        <error to="Kill"/>
    </action>

    <end name="End"/>

</workflow-app>