diff --git a/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/step3/oozie_app/config-default.xml b/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/resolverelation/oozie_app/config-default.xml similarity index 100% rename from dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/step3/oozie_app/config-default.xml rename to dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/resolverelation/oozie_app/config-default.xml diff --git a/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/resolverelation/oozie_app/workflow.xml b/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/resolverelation/oozie_app/workflow.xml new file mode 100644 index 000000000..e73b61a74 --- /dev/null +++ b/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/resolverelation/oozie_app/workflow.xml @@ -0,0 +1,52 @@ + + + + entityPath + the path of deduplicate Entities + + + relationPath + the path of relation unresolved + + + targetPath + the path of relation unresolved + + + + + + + + Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}] + + + + + + + yarn + cluster + Resolve Relations in raw graph + eu.dnetlib.dhp.sx.graph.SparkResolveRelation + dhp-graph-mapper-${projectVersion}.jar + + --executor-memory=${sparkExecutorMemory} + --executor-cores=${sparkExecutorCores} + --driver-memory=${sparkDriverMemory} + --conf spark.extraListeners=${spark2ExtraListeners} + --conf spark.sql.shuffle.partitions=3000 + --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners} + --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress} + --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir} + + --masteryarn + --relationPath${relationPath} + --workingPath${targetPath} + --entityPath${entityPath} + + + + + + \ No newline at end of file diff --git a/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/step3/oozie_app/workflow.xml b/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/step3/oozie_app/workflow.xml deleted file mode 100644 index 4d54b2afb..000000000 --- a/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/sx/graph/step3/oozie_app/workflow.xml +++ /dev/null @@ -1,61 +0,0 @@ - - - - sourcePath - the source path - - - targetPath - the source path - - - sparkDriverMemory - memory for driver process - - - sparkExecutorMemory - memory for individual executor - - - entity - the entity to be merged - - - - - - - Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}] - - - - - - - - - - - - - - - ${jobTracker} - ${nameNode} - yarn-cluster - cluster - Merge ${entity} - eu.dnetlib.dhp.sx.graph.SparkScholexplorerCreateRawGraphJob - dhp-graph-mapper-${projectVersion}.jar - --executor-memory ${sparkExecutorMemory} --driver-memory=${sparkDriverMemory} ${sparkExtraOPT} - -mt yarn-cluster - --sourcePath${sourcePath}/${entity} - --targetPath${targetPath}/${entity} - --entity${entity} - - - - - - - \ No newline at end of file