first version of the workflow single step

This commit is contained in:
Miriam Baglioni 2023-12-07 09:59:52 +01:00
parent c5b7253130
commit 616622d2bb
38 changed files with 2863 additions and 19 deletions

View File

@ -167,4 +167,9 @@ public class Utils implements Serializable {
});
return projectMap;
}
public static List<String> getCommunityIdList(String baseURL) throws IOException {
return getValidCommunities(baseURL).stream()
.map(community -> community.getId()).collect(Collectors.toList());
}
}

View File

@ -151,7 +151,13 @@ public class SparkBulkTagJob {
.write()
.mode(SaveMode.Overwrite)
.option("compression", "gzip")
.json(outputPath + e.name());
.json(outputPath + e.name());//writing the tagging in the working dir for entity
readPath(spark, outputPath + e.name(), resultClazz) //copy the tagging in the actual result output path
.write()
.mode(SaveMode.Overwrite)
.option("compression","gzip")
.json(inputPath + e.name());
});
}

View File

@ -66,7 +66,7 @@ public class PrepareDatasourceCountryAssociation {
conf,
isSparkSessionManaged,
spark -> {
removeOutputDir(spark, outputPath);
//removeOutputDir(spark, outputPath);
prepareDatasourceCountryAssociation(
spark,
Arrays.asList(parser.get("whitelist").split(";")),

View File

@ -97,6 +97,12 @@ public class SparkCountryPropagationJob {
.mode(SaveMode.Overwrite)
.json(outputPath);
readPath(spark, outputPath, resultClazz)
.write()
.mode(SaveMode.Overwrite)
.option("compression","gzip")
.json(sourcePath);
}
private static <R extends Result> MapFunction<Tuple2<R, ResultCountrySet>, R> getCountryMergeFn() {

View File

@ -92,6 +92,12 @@ public class SparkResultToCommunityFromOrganizationJob {
.mode(SaveMode.Overwrite)
.option("compression", "gzip")
.json(outputPath + e.name());
readPath(spark, outputPath + e.name(), resultClazz)
.write()
.mode(SaveMode.Overwrite)
.option("compression","gzip")
.json(inputPath + e.name());
}
});

View File

@ -53,7 +53,7 @@ public class PrepareResultCommunitySet {
log.info("outputPath: {}", outputPath);
final String baseURL = parser.get("baseURL");
log.info("baseUEL: {}", baseURL);
log.info("baseURL: {}", baseURL);
final CommunityEntityMap projectsMap = Utils.getCommunityProjects(baseURL);

View File

@ -102,6 +102,12 @@ public class SparkResultToCommunityFromProject implements Serializable {
.mode(SaveMode.Overwrite)
.option("compression", "gzip")
.json(outputPath + e.name());
readPath(spark, outputPath + e.name(), resultClazz)
.write()
.mode(SaveMode.Overwrite)
.option("compression","gzip")
.json(inputPath + e.name());
}
});

View File

@ -4,9 +4,11 @@ package eu.dnetlib.dhp.resulttocommunityfromsemrel;
import static eu.dnetlib.dhp.PropagationConstant.*;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession;
import java.io.IOException;
import java.util.Arrays;
import java.util.List;
import eu.dnetlib.dhp.api.Utils;
import org.apache.commons.io.IOUtils;
import org.apache.spark.SparkConf;
import org.apache.spark.sql.*;
@ -26,11 +28,6 @@ import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
public class PrepareResultCommunitySetStep1 {
private static final Logger log = LoggerFactory.getLogger(PrepareResultCommunitySetStep1.class);
private static final String COMMUNITY_LIST_XQUERY = "for $x in collection('/db/DRIVER/ContextDSResources/ContextDSResourceType')"
+ " where $x//CONFIGURATION/context[./@type='community' or ./@type='ri']"
+ " and $x//CONFIGURATION/context/param[./@name='status']/text() != 'hidden'"
+ " return $x//CONFIGURATION/context/@id/string()";
/**
* associates to each result the set of community contexts they are associated to; associates to each target of a
* relation with allowed semantics the set of community context it could possibly inherit from the source of the
@ -88,10 +85,10 @@ public class PrepareResultCommunitySetStep1 {
final List<String> allowedsemrel = Arrays.asList(parser.get("allowedsemrels").split(";"));
log.info("allowedSemRel: {}", new Gson().toJson(allowedsemrel));
final String isLookupUrl = parser.get("isLookUpUrl");
log.info("isLookupUrl: {}", isLookupUrl);
final String baseURL = parser.get("baseURL");
log.info("baseURL: {}", baseURL);
final List<String> communityIdList = getCommunityList(isLookupUrl);
final List<String> communityIdList = getCommunityList(baseURL);
log.info("communityIdList: {}", new Gson().toJson(communityIdList));
final String resultType = resultClassName.substring(resultClassName.lastIndexOf(".") + 1).toLowerCase();
@ -159,9 +156,8 @@ public class PrepareResultCommunitySetStep1 {
.json(outputResultPath);
}
public static List<String> getCommunityList(final String isLookupUrl) throws ISLookUpException {
ISLookUpService isLookUp = ISLookupClientFactory.getLookUpService(isLookupUrl);
return isLookUp.quickSearchProfile(COMMUNITY_LIST_XQUERY);
public static List<String> getCommunityList(final String baseURL) throws IOException {
return Utils.getCommunityIdList(baseURL);
}
}

View File

@ -100,6 +100,12 @@ public class SparkResultToCommunityThroughSemRelJob {
.mode(SaveMode.Overwrite)
.option("compression", "gzip")
.json(outputPath);
readPath(spark, outputPath, resultClazz)
.write()
.mode(SaveMode.Overwrite)
.option("compression","gzip")
.json(inputPath);
}
private static <R extends Result> MapFunction<Tuple2<R, ResultCommunityList>, R> contextUpdaterFn() {

View File

@ -1,7 +1,7 @@
[
{
"paramName":"is",
"paramLongName":"isLookUpUrl",
"paramName":"bu",
"paramLongName":"baseURL",
"paramDescription": "URL of the isLookUp Service",
"paramRequired": true
},

View File

@ -9,8 +9,8 @@
<description>the semantic relationships allowed for propagation</description>
</property>
<property>
<name>isLookUpUrl</name>
<description>the isLookup service endpoint</description>
<name>baseURL</name>
<description>the baseurl for the comminity APIs</description>
</property>
<property>
<name>outputPath</name>
@ -116,7 +116,7 @@
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
<arg>--outputPath</arg><arg>${workingDir}/preparedInfo/targetCommunityAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
<arg>--isLookUpUrl</arg><arg>${isLookUpUrl}</arg>
<arg>--baseURL</arg><arg>${baseURL}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>

View File

@ -0,0 +1,30 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>hiveMetastoreUris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>hiveJdbcUrl</name>
<value>jdbc:hive2://iis-cdh5-test-m3.ocean.icm.edu.pl:10000</value>
</property>
<property>
<name>hiveDbName</name>
<value>openaire</value>
</property>
<property>
<name>oozie.launcher.mapreduce.user.classpath.first</name>
<value>true</value>
</property>
</configuration>

View File

@ -0,0 +1,10 @@
## This is a classpath-based import file (this header is required)
orcid_propagation classpath eu/dnetlib/dhp/wf/subworkflows/orcidtoresultfromsemrel/oozie_app
bulk_tagging classpath eu/dnetlib/dhp/wf/subworkflows/bulktag/oozie_app
affiliation_inst_repo classpath eu/dnetlib/dhp/wf/subworkflows/resulttoorganizationfrominstrepo/oozie_app
entity_semantic_relation classpath eu/dnetlib/dhp/wf/subworkflows/entitytoorganizationfromsemrel/oozie_app
community_organization classpath eu/dnetlib/dhp/wf/subworkflows/resulttocommunityfromorganization/oozie_app
result_project classpath eu/dnetlib/dhp/wf/subworkflows/projecttoresult/oozie_app
community_project classpath eu/dnetlib/dhp/wf/subworkflows/resulttocommunityfromproject/oozie_app
community_sem_rel classpath eu/dnetlib/dhp/wf/subworkflows/resulttocommunityfromsemrel/oozie_app
country_propagation classpath eu/dnetlib/dhp/wf/subworkflows/countrypropagation/oozie_app

View File

@ -0,0 +1,324 @@
<workflow-app name="enrichment_main" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>allowedsemrelsorcidprop</name>
<description>the semantic relationships allowed for propagation</description>
</property>
<property>
<name>allowedsemrelsresultproject</name>
<description>the allowed semantics </description>
</property>
<property>
<name>allowedsemrelscommunitysemrel</name>
<description>the semantic relationships allowed for propagation</description>
</property>
<property>
<name>datasourceWhitelistForCountryPropagation</name>
<description>the white list</description>
</property>
<property>
<name>allowedtypes</name>
<description>the allowed types</description>
</property>
<property>
<name>outputPath</name>
<description>the output path</description>
</property>
<property>
<name>organizationtoresultcommunitymap</name>
<description>organization community map</description>
</property>
<property>
<name>pathMap</name>
<description>the json path associated to each selection field</description>
</property>
<property>
<name>blacklist</name>
<description>list of datasources in blacklist for the affiliation from instrepo propagation</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>
</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>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="resumeFrom"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<decision name="resumeFrom">
<switch>
<case to="bulk_tagging">${wf:conf('resumeFrom') eq 'BulkTagging'}</case>
<case to="affiliation_inst_repo">${wf:conf('resumeFrom') eq 'AffiliationInstitutionalRepository'}</case>
<case to="affiliation_semantic_relation">${wf:conf('resumeFrom') eq 'AffiliationSemanticRelation'}</case>
<case to="community_organization">${wf:conf('resumeFrom') eq 'CommunityOrganization'}</case>
<case to="result_project">${wf:conf('resumeFrom') eq 'ResultProject'}</case>
<case to="community_project">${wf:conf('resumeFrom') eq 'CommunityProject'}</case>
<case to="community_sem_rel">${wf:conf('resumeFrom') eq 'CommunitySemanticRelation'}</case>
<case to="country_propagation">${wf:conf('resumeFrom') eq 'CountryPropagation'}</case>
<default to="orcid_propagation"/>
</switch>
</decision>
<action name="orcid_propagation">
<sub-workflow>
<app-path>${wf:appPath()}/orcid_propagation
</app-path>
<propagate-configuration/>
<configuration>
<property>
<name>sourcePath</name>
<value>${sourcePath}</value>
</property>
<property>
<name>allowedsemrels</name>
<value>${allowedsemrelsorcidprop}</value>
</property>
<property>
<name>outputPath</name>
<value>${outputPath}</value>
</property>
</configuration>
</sub-workflow>
<ok to="bulk_tagging" />
<error to="Kill" />
</action>
<action name="bulk_tagging">
<sub-workflow>
<app-path>${wf:appPath()}/bulk_tagging
</app-path>
<propagate-configuration/>
<configuration>
<property>
<name>sourcePath</name>
<value>${outputPath}</value>
</property>
<property>
<name>baseURL</name>
<value>${baseURL}</value>
</property>
<property>
<name>pathMap</name>
<value>${pathMap}</value>
</property>
</configuration>
</sub-workflow>
<ok to="affiliation_inst_repo" />
<error to="Kill" />
</action>
<action name="affiliation_inst_repo">
<sub-workflow>
<app-path>${wf:appPath()}/affiliation_inst_repo
</app-path>
<propagate-configuration/>
<configuration>
<property>
<name>sourcePath</name>
<value>${outputPath}</value>
</property>
<property>
<name>blacklist</name>
<value>${blacklist}</value>
</property>
</configuration>
</sub-workflow>
<ok to="affiliation_semantic_relation" />
<error to="Kill" />
</action>
<action name="affiliation_semantic_relation">
<sub-workflow>
<app-path>${wf:appPath()}/affiliation_semantic_relation
</app-path>
<propagate-configuration/>
<configuration>
<property>
<name>sourcePath</name>
<value>${outputPath}</value>
</property>
</configuration>
</sub-workflow>
<ok to="community_organization" />
<error to="Kill" />
</action>
<action name="community_organization">
<sub-workflow>
<app-path>${wf:appPath()}/community_organization
</app-path>
<propagate-configuration/>
<configuration>
<property>
<name>sourcePath</name>
<value>${outputPath}</value>
</property>
<property>
<name>baseURL</name>
<value>${baseURL}</value>
</property>
</configuration>
</sub-workflow>
<ok to="result_project" />
<error to="Kill" />
</action>
<action name="result_project">
<sub-workflow>
<app-path>${wf:appPath()}/result_project
</app-path>
<propagate-configuration/>
<configuration>
<property>
<name>sourcePath</name>
<value>${outputPath}</value>
</property>
<property>
<name>allowedsemrels</name>
<value>${allowedsemrelsresultproject}</value>
</property>
</configuration>
</sub-workflow>
<ok to="community_project" />
<error to="Kill" />
</action>
<action name="community_project">
<sub-workflow>
<app-path>${wf:appPath()}/community_project
</app-path>
<propagate-configuration/>
<configuration>
<property>
<name>sourcePath</name>
<value>${outputPath}</value>
</property>
</configuration>
</sub-workflow>
<ok to="community_sem_rel" />
<error to="Kill" />
</action>
<action name="community_sem_rel">
<sub-workflow>
<app-path>${wf:appPath()}/community_sem_rel
</app-path>
<propagate-configuration/>
<configuration>
<property>
<name>sourcePath</name>
<value>${outputPath}</value>
</property>
<property>
<name>allowedsemrels</name>
<value>${allowedsemrelscommunitysemrel}</value>
</property>
<property>
<name>baseURL</name>
<value>${baseURL}</value>
</property>
</configuration>
</sub-workflow>
<ok to="country_propagation" />
<error to="Kill" />
</action>
<action name="country_propagation">
<sub-workflow>
<app-path>${wf:appPath()}/country_propagation
</app-path>
<propagate-configuration/>
<configuration>
<property>
<name>sourcePath</name>
<value>${outputPath}</value>
</property>
<property>
<name>whitelist</name>
<value>${datasourceWhitelistForCountryPropagation}</value>
</property>
<property>
<name>allowedtypes</name>
<value>${allowedtupes}</value>
</property>
</configuration>
</sub-workflow>
<ok to="End" />
<error to="Kill" />
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,54 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
</property>
<property>
<name>sparkExecutorNumber</name>
<value>4</value>
</property>
<property>
<name>spark2EventLogDir</name>
<value>/user/spark/spark2ApplicationHistory</value>
</property>
<property>
<name>sparkDriverMemory</name>
<value>15G</value>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>6G</value>
</property>
<property>
<name>sparkExecutorCores</name>
<value>1</value>
</property>
</configuration>

View File

@ -0,0 +1,66 @@
<workflow-app name="bulk_tagging" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>pathMap</name>
<description>the json path associated to each selection field</description>
</property>
<property>
<name>baseURL</name>
<description>The URL to access the community APIs</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="exec_bulktag"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="exec_bulktag">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>bulkTagging-publication</name>
<class>eu.dnetlib.dhp.bulktag.SparkBulkTagJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--num-executors=${sparkExecutorNumber}
--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}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/</arg>
<arg>--workingPath</arg><arg>${workingDir}/bulktag/</arg>
<arg>--pathMap</arg><arg>${pathMap}</arg>
<arg>--baseURL</arg><arg>${baseURL}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,58 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
</property>
<property>
<name>spark2EventLogDir</name>
<value>/user/spark/spark2ApplicationHistory</value>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
</property>
<property>
<name>sparkExecutorNumber</name>
<value>4</value>
</property>
<property>
<name>sparkDriverMemory</name>
<value>15G</value>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>6G</value>
</property>
<property>
<name>sparkExecutorCores</name>
<value>1</value>
</property>
<property>
<name>spark2MaxExecutors</name>
<value>50</value>
</property>
</configuration>

View File

@ -0,0 +1,316 @@
<workflow-app name="country_propagation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>whitelist</name>
<description>the white list</description>
</property>
<property>
<name>allowedtypes</name>
<description>the allowed types</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="prepare_datasource_country_association"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="prepare_datasource_country_association">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>PrepareDatasourceCountryAssociation</name>
<class>eu.dnetlib.dhp.countrypropagation.PrepareDatasourceCountryAssociation</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--whitelist</arg><arg>${whitelist}</arg>
<arg>--allowedtypes</arg><arg>${allowedtypes}</arg>
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
</spark>
<ok to="fork_prepare_result_country"/>
<error to="Kill"/>
</action>
<fork name="fork_prepare_result_country">
<path start="prepareresult_publication"/>
<path start="prepareresult_dataset"/>
<path start="prepareresult_otherresearchproduct"/>
<path start="prepareresult_software"/>
</fork>
<action name="prepareresult_publication">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>prepareResultCountry-Publication</name>
<class>eu.dnetlib.dhp.countrypropagation.PrepareResultCountrySet</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
</spark>
<ok to="wait_prepare"/>
<error to="Kill"/>
</action>
<action name="prepareresult_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>prepareResultCountry-Dataset</name>
<class>eu.dnetlib.dhp.countrypropagation.PrepareResultCountrySet</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
</spark>
<ok to="wait_prepare"/>
<error to="Kill"/>
</action>
<action name="prepareresult_otherresearchproduct">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>prepareResultCountry-ORP</name>
<class>eu.dnetlib.dhp.countrypropagation.PrepareResultCountrySet</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
</spark>
<ok to="wait_prepare"/>
<error to="Kill"/>
</action>
<action name="prepareresult_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>prepareResultCountry-Software</name>
<class>eu.dnetlib.dhp.countrypropagation.PrepareResultCountrySet</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
</spark>
<ok to="wait_prepare"/>
<error to="Kill"/>
</action>
<join name="wait_prepare" to="fork_join_apply_country_propagation"/>
<fork name="fork_apply_country_propagation">
<path start="propagation_publication"/>
<path start="propagation_dataset"/>
<path start="propagation_otherresearchproduct"/>
<path start="propagation_software"/>
</fork>
<action name="propagation_publication">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>countryPropagationForPublications</name>
<class>eu.dnetlib.dhp.countrypropagation.SparkCountryPropagationJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="propagation_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>countryPropagationForDataset</name>
<class>eu.dnetlib.dhp.countrypropagation.SparkCountryPropagationJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="propagation_otherresearchproduct">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>countryPropagationForORP</name>
<class>eu.dnetlib.dhp.countrypropagation.SparkCountryPropagationJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="propagation_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>countryPropagationForSoftware</name>
<class>eu.dnetlib.dhp.countrypropagation.SparkCountryPropagationJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
<arg>--workingPath</arg><arg>${workingDir}/country</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<join name="wait" to="reset_workingDir"/>
<action name="reset_workingDir">
<fs>
<delete path="${workingDir}"/>
<mkdir path="${workingDir}"/>
</fs>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,50 @@
[
{
"paramName":"gp",
"paramLongName":"graphPath",
"paramDescription": "the path of the sequencial file to read",
"paramRequired": true
},
{
"paramName":"h",
"paramLongName":"hive_metastore_uris",
"paramDescription": "the hive metastore uris",
"paramRequired": true
},
{
"paramName":"lp",
"paramLongName":"leavesPath",
"paramDescription": "true if the new version of the graph must be saved",
"paramRequired": false
},
{
"paramName":"cp",
"paramLongName":"childParentPath",
"paramDescription": "path where to store/find association from datasource and organization",
"paramRequired": true
},
{
"paramName":"rp",
"paramLongName":"resultOrgPath",
"paramDescription": "path where to store/find already linked results and organizations",
"paramRequired": true
},
{
"paramName": "ssm",
"paramLongName": "isSparkSessionManaged",
"paramDescription": "the path where prepared info have been stored",
"paramRequired": false
},
{
"paramName": "rep",
"paramLongName": "relationPath",
"paramDescription": "the path where to store the selected subset of relations",
"paramRequired": false
},
{
"paramName": "pop",
"paramLongName": "projectOrganizationPath",
"paramDescription": "the number of iterations to be computed",
"paramRequired": true
}
]

View File

@ -0,0 +1,62 @@
[
{
"paramName":"rep",
"paramLongName":"relationPath",
"paramDescription": "the path of the sequencial file to read",
"paramRequired": true
},
{
"paramName":"h",
"paramLongName":"hive_metastore_uris",
"paramDescription": "the hive metastore uris",
"paramRequired": true
},
{
"paramName":"lp",
"paramLongName":"leavesPath",
"paramDescription": "true if the new version of the graph must be saved",
"paramRequired": false
},
{
"paramName":"cp",
"paramLongName":"childParentPath",
"paramDescription": "path where to store/find association from datasource and organization",
"paramRequired": true
},
{
"paramName":"rp",
"paramLongName":"resultOrgPath",
"paramDescription": "path where to store/find already linked results and organizations",
"paramRequired": true
},
{
"paramName": "ssm",
"paramLongName": "isSparkSessionManaged",
"paramDescription": "the path where prepared info have been stored",
"paramRequired": false
},
{
"paramName": "wd",
"paramLongName": "workingDir",
"paramDescription": "true if it is a test running",
"paramRequired": false
},
{
"paramName": "out",
"paramLongName": "outputPath",
"paramDescription": "the path used to store temporary output files",
"paramRequired": true
},
{
"paramName": "it",
"paramLongName": "iterations",
"paramDescription": "the number of iterations to be computed",
"paramRequired": false
},
{
"paramName": "pop",
"paramLongName": "projectOrganizationPath",
"paramDescription": "the number of iterations to be computed",
"paramRequired": true
}
]

View File

@ -0,0 +1,58 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
</property>
<property>
<name>spark2EventLogDir</name>
<value>/user/spark/spark2ApplicationHistory</value>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
</property>
<property>
<name>sparkExecutorNumber</name>
<value>4</value>
</property>
<property>
<name>sparkDriverMemory</name>
<value>15G</value>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>6G</value>
</property>
<property>
<name>sparkExecutorCores</name>
<value>1</value>
</property>
<property>
<name>spark2MaxExecutors</name>
<value>50</value>
</property>
</configuration>

View File

@ -0,0 +1,93 @@
<workflow-app name="affiliation_from_semrel_propagation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="prepare_info"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="prepare_info">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>PrepareResultOrganizationAssociation</name>
<class>eu.dnetlib.dhp.entitytoorganizationfromsemrel.PrepareInfo</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--graphPath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--leavesPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/leavesPath</arg>
<arg>--childParentPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/childParentPath</arg>
<arg>--resultOrgPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/resultOrgPath</arg>
<arg>--projectOrganizationPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/projectOrganizationPath</arg>
<arg>--relationPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/relation</arg>
</spark>
<ok to="apply_resulttoorganization_propagation"/>
<error to="Kill"/>
</action>
<action name="apply_resulttoorganization_propagation">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>resultToOrganizationFromSemRel</name>
<class>eu.dnetlib.dhp.entitytoorganizationfromsemrel.SparkResultToOrganizationFromSemRel</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--relationPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/relation</arg>
<arg>--outputPath</arg><arg>${sourcePath}/relation</arg>
<arg>--leavesPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/leavesPath</arg>
<arg>--childParentPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/childParentPath</arg>
<arg>--resultOrgPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/resultOrgPath</arg>
<arg>--projectOrganizationPath</arg><arg>${workingDir}/entitiesSemanticRelation/preparedInfo/projectOrganizationPath</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--workingDir</arg><arg>${workingDir}/entitiesSemanticRelation/working</arg>
<arg>--iterations</arg><arg>${iterations}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,58 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>spark2EventLogDir</name>
<value>/user/spark/spark2ApplicationHistory</value>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
</property>
<property>
<name>sparkExecutorNumber</name>
<value>4</value>
</property>
<property>
<name>sparkDriverMemory</name>
<value>15G</value>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>6G</value>
</property>
<property>
<name>sparkExecutorCores</name>
<value>1</value>
</property>
<property>
<name>spark2MaxExecutors</name>
<value>50</value>
</property>
</configuration>

View File

@ -0,0 +1,369 @@
<workflow-app name="orcid_to_result_from_semrel_propagation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>allowedsemrels</name>
<description>the semantic relationships allowed for propagation</description>
</property>
<property>
<name>outputPath</name>
<description>the output path</description>
</property>
</parameters>
<start to="reset_outputpath"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="reset_outputpath">
<fs>
<delete path="${outputPath}"/>
<mkdir path="${outputPath}"/>
</fs>
<ok to="copy_entities"/>
<error to="Kill"/>
</action>
<fork name="copy_entities">
<path start="copy_relation"/>
<path start="copy_organization"/>
<path start="copy_projects"/>
<path start="copy_datasources"/>
</fork>
<action name="copy_relation">
<distcp xmlns="uri:oozie:distcp-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<arg>${nameNode}/${sourcePath}/relation</arg>
<arg>${nameNode}/${outputPath}/relation</arg>
</distcp>
<ok to="copy_wait"/>
<error to="Kill"/>
</action>
<action name="copy_organization">
<distcp xmlns="uri:oozie:distcp-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<arg>${nameNode}/${sourcePath}/organization</arg>
<arg>${nameNode}/${outputPath}/organization</arg>
</distcp>
<ok to="copy_wait"/>
<error to="Kill"/>
</action>
<action name="copy_projects">
<distcp xmlns="uri:oozie:distcp-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<arg>${nameNode}/${sourcePath}/project</arg>
<arg>${nameNode}/${outputPath}/project</arg>
</distcp>
<ok to="copy_wait"/>
<error to="Kill"/>
</action>
<action name="copy_datasources">
<distcp xmlns="uri:oozie:distcp-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<arg>${nameNode}/${sourcePath}/datasource</arg>
<arg>${nameNode}/${outputPath}/datasource</arg>
</distcp>
<ok to="copy_wait"/>
<error to="Kill"/>
</action>
<join name="copy_wait" to="prepare_relations"/>
<action name="join_prepare_publication">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-PreparePhase1-Publications</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.PrepareResultOrcidAssociationStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
--conf spark.sql.shuffle.partitions=3840
--conf spark.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
<arg>--outputPath</arg><arg>${workingDir}/preparedInfo/targetOrcidAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="join_prepare_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-PreparePhase1-Dataset</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.PrepareResultOrcidAssociationStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
<arg>--outputPath</arg><arg>${workingDir}/preparedInfo/targetOrcidAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="join_prepare_otherresearchproduct">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-PreparePhase1-ORP</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.PrepareResultOrcidAssociationStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
<arg>--outputPath</arg><arg>${workingDir}/preparedInfo/targetOrcidAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="join_prepare_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-PreparePhase1-Software</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.PrepareResultOrcidAssociationStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
<arg>--outputPath</arg><arg>${workingDir}/preparedInfo/targetOrcidAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<join name="wait" to="prepare_assoc_step2"/>
<action name="prepare_assoc_step2">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-PreparePhase2</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.PrepareResultOrcidAssociationStep2</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${workingDir}/orcidprop</arg>
<arg>--outputPath</arg><arg>${workingDir}/orcidprop/mergedOrcidAssoc</arg>
</spark>
<ok to="fork-join-exec-propagation"/>
<error to="Kill"/>
</action>
<fork name="fork-join-exec-propagation">
<path start="join_propagate_publication"/>
<path start="join_propagate_dataset"/>
<path start="join_propagate_otherresearchproduct"/>
<path start="join_propagate_software"/>
</fork>
<action name="join_propagate_publication">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-Publication</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.SparkOrcidToResultFromSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
--conf spark.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--possibleUpdatesPath</arg><arg>${workingDir}/orcidprop/mergedOrcidAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
<arg>--outputPath</arg><arg>${outputPath}/publication</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagate_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-Dataset</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.SparkOrcidToResultFromSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
--conf spark.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
</spark-opts>
<arg>--possibleUpdatesPath</arg><arg>${workingDir}/orcidprop/mergedOrcidAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
<arg>--outputPath</arg><arg>${outputPath}/dataset</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagate_otherresearchproduct">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-ORP</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.SparkOrcidToResultFromSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
--conf spark.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
</spark-opts>
<arg>--possibleUpdatesPath</arg><arg>${workingDir}/orcidprop/mergedOrcidAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
<arg>--outputPath</arg><arg>${outputPath}/otherresearchproduct</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagate_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ORCIDPropagation-Software</name>
<class>eu.dnetlib.dhp.orcidtoresultfromsemrel.SparkOrcidToResultFromSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
--conf spark.speculation=false
--conf spark.hadoop.mapreduce.map.speculative=false
--conf spark.hadoop.mapreduce.reduce.speculative=false
</spark-opts>
<arg>--possibleUpdatesPath</arg><arg>${workingDir}/orcidprop/mergedOrcidAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
<arg>--outputPath</arg><arg>${outputPath}/software</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<join name="wait2" to="reset_workingDir"/>
<action name="reset_workingDir">
<fs>
<delete path="${workingDir}"/>
<mkdir path="${workingDir}"/>
</fs>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,63 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
<!-- <value>hadoop-rm3.garr-pa1.d4science.org:8032</value>-->
</property>
<property>
<name>nameNode</name>
<!-- <value>hdfs://hadoop-rm1.garr-pa1.d4science.org:8020</value>-->
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
<!-- <value>thrift://hadoop-edge3.garr-pa1.d4science.org:9083</value>-->
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
</property>
<property>
<name>spark2EventLogDir</name>
<value>/user/spark/spark2ApplicationHistory</value>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
</property>
<property>
<name>sparkExecutorNumber</name>
<value>4</value>
</property>
<property>
<name>sparkDriverMemory</name>
<value>15G</value>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>6G</value>
</property>
<property>
<name>sparkExecutorCores</name>
<value>1</value>
</property>
<property>
<name>spark2MaxExecutors</name>
<value>50</value>
</property>
</configuration>

View File

@ -0,0 +1,94 @@
<workflow-app name="project_to_result_propagation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>allowedsemrels</name>
<description>the allowed semantics </description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="prepare_project_results_association"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="prepare_project_results_association">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>PrepareProjectResultsAssociation</name>
<class>eu.dnetlib.dhp.projecttoresult.PrepareProjectResultsAssociation</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/relation</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--potentialUpdatePath</arg><arg>${workingDir}/resultproject/preparedInfo/potentialUpdates</arg>
<arg>--alreadyLinkedPath</arg><arg>${workingDir}/resultproject/preparedInfo/alreadyLinked</arg>
</spark>
<ok to="apply_propagation"/>
<error to="Kill"/>
</action>
<action name="apply_propagation">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ProjectToResultPropagation</name>
<class>eu.dnetlib.dhp.projecttoresult.SparkResultToProjectThroughSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--outputPath</arg><arg>${sourcePath}/relation</arg>
<arg>--potentialUpdatePath</arg><arg>${workingDir}/resultproject/preparedInfo/potentialUpdates</arg>
<arg>--alreadyLinkedPath</arg><arg>${workingDir}/resultproject/preparedInfo/alreadyLinked</arg>
</spark>
<ok to="reset_workingDir"/>
<error to="Kill"/>
</action>
<action name="reset_workingDir">
<fs>
<delete path="${workingDir}"/>
<mkdir path="${workingDir}"/>
</fs>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,58 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
</property>
<property>
<name>spark2EventLogDir</name>
<value>/user/spark/spark2ApplicationHistory</value>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
</property>
<property>
<name>sparkExecutorNumber</name>
<value>4</value>
</property>
<property>
<name>sparkDriverMemory</name>
<value>15G</value>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>6G</value>
</property>
<property>
<name>sparkExecutorCores</name>
<value>1</value>
</property>
<property>
<name>spark2MaxExecutors</name>
<value>50</value>
</property>
</configuration>

View File

@ -0,0 +1,88 @@
<workflow-app name="community_to_result_propagation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>baseURL</name>
<description>the baseURL from where to reach the community APIs</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="prepare_result_communitylist"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="prepare_result_communitylist">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Prepare-Community-Result-Organization</name>
<class>eu.dnetlib.dhp.resulttocommunityfromorganization.PrepareResultCommunitySet</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=6
--executor-memory=5G
--conf spark.executor.memoryOverhead=3g
--conf spark.sql.shuffle.partitions=3284
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/relation</arg>
<arg>--outputPath</arg><arg>${workingDir}/communityorganization/preparedInfo/resultCommunityList</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--baseURL</arg><arg>${baseURL}</arg>
</spark>
<ok to="exec-propagation"/>
<error to="Kill"/>
</action>
<action name="exec-propagation">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>community2resultfromorganization-Publication</name>
<class>eu.dnetlib.dhp.resulttocommunityfromorganization.SparkResultToCommunityFromOrganizationJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=6
--executor-memory=5G
--conf spark.executor.memoryOverhead=3g
--conf spark.sql.shuffle.partitions=3284
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--preparedInfoPath</arg><arg>${workingDir}/preparedInfo/resultCommunityList</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/</arg>
<arg>--outputPath</arg><arg>${workingDir}/resulttocommunityfromorganization/</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,28 @@
[
{
"paramName":"s",
"paramLongName":"sourcePath",
"paramDescription": "the path of the sequencial file to read",
"paramRequired": true
},
{
"paramName": "out",
"paramLongName": "outputPath",
"paramDescription": "the path used to store temporary output files",
"paramRequired": true
},
{
"paramName": "ssm",
"paramLongName": "isSparkSessionManaged",
"paramDescription": "true if the spark session is managed, false otherwise",
"paramRequired": false
},
{
"paramName": "p",
"paramLongName": "preparedInfoPath",
"paramDescription": "the path where prepared info have been stored",
"paramRequired": true
}
]

View File

@ -0,0 +1,28 @@
[
{
"paramName":"s",
"paramLongName":"sourcePath",
"paramDescription": "the path of the sequencial file to read",
"paramRequired": true
},
{
"paramName": "ssm",
"paramLongName": "isSparkSessionManaged",
"paramDescription": "true if the spark session is managed, false otherwise",
"paramRequired": false
},
{
"paramName": "out",
"paramLongName": "outputPath",
"paramDescription": "the path used to store temporary output files",
"paramRequired": true
},
{
"paramName": "bu",
"paramLongName": "baseURL",
"paramDescription": "the path used to store temporary output files",
"paramRequired": false
}
]

View File

@ -0,0 +1,58 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
</property>
<property>
<name>spark2EventLogDir</name>
<value>/user/spark/spark2ApplicationHistory</value>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
</property>
<property>
<name>sparkExecutorNumber</name>
<value>4</value>
</property>
<property>
<name>sparkDriverMemory</name>
<value>15G</value>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>6G</value>
</property>
<property>
<name>sparkExecutorCores</name>
<value>1</value>
</property>
<property>
<name>spark2MaxExecutors</name>
<value>50</value>
</property>
</configuration>

View File

@ -0,0 +1,90 @@
<workflow-app name="community_to_result_propagation_project" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>baseURL</name>
<description>the base URL to use to select the right community APIs</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="prepare_result_communitylist"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="prepare_result_communitylist">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Prepare-Community-Result-Organization</name>
<class>eu.dnetlib.dhp.resulttocommunityfromproject.PrepareResultCommunitySet</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=6
--executor-memory=5G
--conf spark.executor.memoryOverhead=3g
--conf spark.sql.shuffle.partitions=3284
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/relation</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitythroughproject/preparedInfo/resultCommunityList</arg>
<arg>--baseURL</arg><arg>${baseURL}</arg>
</spark>
<ok to="exec-propagation"/>
<error to="Kill"/>
</action>
<action name="exec-propagation">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>community2resultfromproject</name>
<class>eu.dnetlib.dhp.resulttocommunityfromproject.SparkResultToCommunityFromProject</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=6
--executor-memory=5G
--conf spark.executor.memoryOverhead=3g
--conf spark.sql.shuffle.partitions=3284
--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.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--preparedInfoPath</arg><arg>${workingDir}/communitythroughproject/preparedInfo/resultCommunityList</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitythroughproject/</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,58 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
</property>
<property>
<name>spark2EventLogDir</name>
<value>/user/spark/spark2ApplicationHistory</value>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
</property>
<property>
<name>sparkExecutorNumber</name>
<value>4</value>
</property>
<property>
<name>sparkDriverMemory</name>
<value>15G</value>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>6G</value>
</property>
<property>
<name>sparkExecutorCores</name>
<value>1</value>
</property>
<property>
<name>spark2MaxExecutors</name>
<value>50</value>
</property>
</configuration>

View File

@ -0,0 +1,305 @@
<workflow-app name="result_to_community_from_semrel_propagation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>allowedsemrels</name>
<description>the semantic relationships allowed for propagation</description>
</property>
<property>
<name>baseURL</name>
<description>the isLookup service endpoint</description>
</property>
<property>
<name>outputPath</name>
<description>the output path</description>
</property>
</parameters>
<start to="fork_prepare_assoc_step1"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<fork name="fork_prepare_assoc_step1">
<path start="join_prepare_publication"/>
<path start="join_prepare_dataset"/>
<path start="join_prepare_otherresearchproduct"/>
<path start="join_prepare_software"/>
</fork>
<action name="join_prepare_publication">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ResultToCommunitySemRel-PreparePhase1-Publications</name>
<class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/targetCommunityAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
<arg>--baseURL</arg><arg>${baseURL}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="join_prepare_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ResultToCommunitySemRel-PreparePhase1-Dataset</name>
<class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/targetCommunityAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
<arg>--baseURL</arg><arg>${baseURL}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="join_prepare_otherresearchproduct">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ResultToCommunitySemRel-PreparePhase1-ORP</name>
<class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/targetCommunityAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
<arg>--baseURL</arg><arg>${baseURL}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<action name="join_prepare_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ResultToCommunitySemRel-PreparePhase1-Software</name>
<class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep1</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/targetCommunityAssoc</arg>
<arg>--allowedsemrels</arg><arg>${allowedsemrels}</arg>
<arg>--baseURL</arg><arg>${baseURL}</arg>
</spark>
<ok to="wait"/>
<error to="Kill"/>
</action>
<join name="wait" to="prepare_assoc_step2"/>
<action name="prepare_assoc_step2">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ResultToCommunityEmRelPropagation-PreparePhase2</name>
<class>eu.dnetlib.dhp.resulttocommunityfromsemrel.PrepareResultCommunitySetStep2</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${workingDir}/communitysemrel/preparedInfo/targetCommunityAssoc</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/mergedCommunityAssoc</arg>
</spark>
<ok to="fork-join-exec-propagation"/>
<error to="Kill"/>
</action>
<fork name="fork-join-exec-propagation">
<path start="join_propagate_publication"/>
<path start="join_propagate_dataset"/>
<path start="join_propagate_otherresearchproduct"/>
<path start="join_propagate_software"/>
</fork>
<action name="join_propagate_publication">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Result2CommunitySemRelPropagation-Publication</name>
<class>eu.dnetlib.dhp.resulttocommunityfromsemrel.SparkResultToCommunityThroughSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--preparedInfoPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/mergedCommunityAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitysemrel/publication</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagate_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Result2CommunitySemRelPropagation-Dataset</name>
<class>eu.dnetlib.dhp.resulttocommunityfromsemrel.SparkResultToCommunityThroughSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--preparedInfoPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/mergedCommunityAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitysemrel/dataset</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagate_otherresearchproduct">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Result2CommunitySemRelPropagation-ORP</name>
<class>eu.dnetlib.dhp.resulttocommunityfromsemrel.SparkResultToCommunityThroughSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--preparedInfoPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/mergedCommunityAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitysemrel/otherresearchproduct</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagate_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Result2CommunitySemRelPropagation-Software</name>
<class>eu.dnetlib.dhp.resulttocommunityfromsemrel.SparkResultToCommunityThroughSemRelJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--preparedInfoPath</arg><arg>${workingDir}/communitysemrel/preparedInfo/mergedCommunityAssoc</arg>
<arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
<arg>--outputPath</arg><arg>${workingDir}/communitysemrel/software</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,58 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
</property>
<property>
<name>spark2EventLogDir</name>
<value>/user/spark/spark2ApplicationHistory</value>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
</property>
<property>
<name>sparkExecutorNumber</name>
<value>4</value>
</property>
<property>
<name>sparkDriverMemory</name>
<value>15G</value>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>6G</value>
</property>
<property>
<name>sparkExecutorCores</name>
<value>1</value>
</property>
<property>
<name>spark2MaxExecutors</name>
<value>50</value>
</property>
</configuration>

View File

@ -0,0 +1,182 @@
<workflow-app name="affiliation_from_instrepo_propagation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
<property>
<name>blacklist</name>
<description>The list of institutional repositories that should not be used for the propagation</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="prepare_result_organization_association"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="prepare_result_organization_association">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>PrepareResultOrganizationAssociation</name>
<class>eu.dnetlib.dhp.resulttoorganizationfrominstrepo.PrepareResultInstRepoAssociation</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${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>${workingDir}/affiliationInstRepo</arg>
<arg>--blacklist</arg><arg>${blacklist}</arg>
</spark>
<ok to="fork_join_apply_resulttoorganization_propagation"/>
<error to="Kill"/>
</action>
<fork name="fork_join_apply_resulttoorganization_propagation">
<path start="join_propagation_publication"/>
<path start="join_propagation_dataset"/>
<path start="join_propagation_otherresearchproduct"/>
<path start="join_propagation_software"/>
</fork>
<action name="join_propagation_publication">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>resultToOrganizationFromInstRepoPropagationForPublications</name>
<class>eu.dnetlib.dhp.resulttoorganizationfrominstrepo.SparkResultToOrganizationFromIstRepoJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
<arg>--outputPath</arg><arg>${sourcePath}/relation</arg>
<arg>--datasourceOrganizationPath</arg><arg>${workingDir}/affiliationInstRepo/preparedInfo/datasourceOrganization</arg>
<arg>--alreadyLinkedPath</arg><arg>${workingDir}/affiliationInstRepo/preparedInfo/alreadyLinked</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagation_dataset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>resultToOrganizationFromInstRepoPropagationForDataset</name>
<class>eu.dnetlib.dhp.resulttoorganizationfrominstrepo.SparkResultToOrganizationFromIstRepoJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
<arg>--outputPath</arg><arg>${sourcePath}/relation</arg>
<arg>--datasourceOrganizationPath</arg><arg>${workingDir}/affiliationInstRepo/preparedInfo/datasourceOrganization</arg>
<arg>--alreadyLinkedPath</arg><arg>${workingDir}/affiliationInstRepo/preparedInfo/alreadyLinked</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagation_otherresearchproduct">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>resultToOrganizationFromInstRepoPropagationForORP</name>
<class>eu.dnetlib.dhp.resulttoorganizationfrominstrepo.SparkResultToOrganizationFromIstRepoJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
<arg>--outputPath</arg><arg>${sourcePath}/relation</arg>
<arg>--datasourceOrganizationPath</arg><arg>${workingDir}/affiliationInstRepo/preparedInfo/datasourceOrganization</arg>
<arg>--alreadyLinkedPath</arg><arg>${workingDir}/affiliationInstRepo/preparedInfo/alreadyLinked</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<action name="join_propagation_software">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>resultToOrganizationFromInstRepoPropagationForSoftware</name>
<class>eu.dnetlib.dhp.resulttoorganizationfrominstrepo.SparkResultToOrganizationFromIstRepoJob</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
</spark-opts>
<arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
<arg>--outputPath</arg><arg>${sourcePath}/relation</arg>
<arg>--datasourceOrganizationPath</arg><arg>${workingDir}/affiliationInstRepo/preparedInfo/datasourceOrganization</arg>
<arg>--alreadyLinkedPath</arg><arg>${workingDir}/affiliationInstRepo/preparedInfo/alreadyLinked</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--resultTableName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
</spark>
<ok to="wait2"/>
<error to="Kill"/>
</action>
<join name="wait2" to="End"/>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,58 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<value>http://iis-cdh5-test-gw.ocean.icm.edu.pl:18089</value>
</property>
<property>
<name>spark2EventLogDir</name>
<value>/user/spark/spark2ApplicationHistory</value>
</property>
<property>
<name>spark2ExtraListeners</name>
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
</property>
<property>
<name>spark2SqlQueryExecutionListeners</name>
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
</property>
<property>
<name>sparkExecutorNumber</name>
<value>4</value>
</property>
<property>
<name>sparkDriverMemory</name>
<value>15G</value>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>6G</value>
</property>
<property>
<name>sparkExecutorCores</name>
<value>1</value>
</property>
<property>
<name>spark2MaxExecutors</name>
<value>50</value>
</property>
</configuration>

View File

@ -0,0 +1,97 @@
<workflow-app name="affiliation_from_semrel_propagation" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>sourcePath</name>
<description>the source path</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="prepare_info"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="prepare_info">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>PrepareResultOrganizationAssociation</name>
<class>eu.dnetlib.dhp.resulttoorganizationfromsemrel.PrepareInfo</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--graphPath</arg><arg>${sourcePath}</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--leavesPath</arg><arg>${workingDir}/affiliationSemanticRelation/preparedInfo/leavesPath</arg>
<arg>--childParentPath</arg><arg>${workingDir}/affiliationSemanticRelation/preparedInfo/childParentPath</arg>
<arg>--resultOrgPath</arg><arg>${workingDir}/affiliationSemanticRelation/preparedInfo/resultOrgPath</arg>
<arg>--relationPath</arg><arg>${workingDir}/affiliationSemanticRelation/preparedInfo/relation</arg>
</spark>
<ok to="apply_resulttoorganization_propagation"/>
<error to="Kill"/>
</action>
<action name="apply_resulttoorganization_propagation">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>resultToOrganizationFromSemRel</name>
<class>eu.dnetlib.dhp.resulttoorganizationfromsemrel.SparkResultToOrganizationFromSemRel</class>
<jar>dhp-enrichment-${projectVersion}.jar</jar>
<spark-opts>
--executor-cores=${sparkExecutorCores}
--executor-memory=${sparkExecutorMemory}
--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.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--relationPath</arg><arg>${workingDir}/affiliationSemanticRelation/preparedInfo/relation</arg>
<arg>--outputPath</arg><arg>${sourcePath}</arg>
<arg>--leavesPath</arg><arg>${workingDir}/affiliationSemanticRelation/preparedInfo/leavesPath</arg>
<arg>--childParentPath</arg><arg>${workingDir}/affiliationSemanticRelation/preparedInfo/childParentPath</arg>
<arg>--resultOrgPath</arg><arg>${workingDir}/affiliationSemanticRelation/preparedInfo/resultOrgPath</arg>
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
<arg>--workingDir</arg><arg>${workingDir}/affiliationSemanticRelation/working</arg>
<arg>--iterations</arg><arg>${iterations}</arg>
</spark>
<ok to="reset_workingDir"/>
<error to="Kill"/>
</action>
<action name="reset_workingDir">
<fs>
<delete path="${workingDir}"/>
<mkdir path="${workingDir}"/>
</fs>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>