enrico.ottonello
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
2,596 Commits 14 Branches 0 Tags 72 MiB
Commit Graph

47 Commits

Author SHA1 Message Date
Alessia Bardi 09fc7e2f78 serialization of validated flag on relationships 2021-02-10 11:22:09 +01:00
Claudio Atzori 0374d34c3e introduced configuration param outputFormat: HDFS | SOLR 2020-11-19 10:34:28 +01:00
Claudio Atzori d9e07a242b extended XmlIndexingJob to accept an optional parameter: outputPath. When present, forces the job to write its output on the specified HDFS location 2020-11-18 14:34:55 +01:00
Claudio Atzori 1871d1c6f6 solve error java.lang.NoSuchFieldError: INSTANCE when instantiating Solr client 2020-08-14 11:18:30 +02:00
Claudio Atzori 3a11a387a9 data provision workflow enhancement: added nodes to perform DELETE BY QUERY before the indexing begins and COMMIT after the indexing is completed 2020-08-03 14:28:08 +02:00
Claudio Atzori cc5d13da85 introduced parameter shouldIndex (true|false) 2020-07-16 13:46:39 +02:00
Claudio Atzori 06c1913062 added different limits for grouping by source and by target, incremented spark.sql.shuffle.partitions for the join operations 2020-07-10 19:03:33 +02:00
Claudio Atzori b21866a2da allow to set different to relations cut points by source and by target; adjusted weight assigned to relationship types 2020-07-10 13:59:48 +02:00
Claudio Atzori b383ed42fa pass optional parameter relationFilter to the PrepareRelationJob implementation 2020-07-07 14:21:28 +02:00
Claudio Atzori 7817338e05 added test to verify the relation pre-processing 2020-06-26 17:58:33 +02:00
Claudio Atzori 216975c4ec restored complete provision workflow 2020-06-25 12:55:52 +02:00
Claudio Atzori 0e723d378b added default from vocab for missing instance.refereed; remove spurious prefixes from orcid values; WIP: prepare relation job 2020-06-24 18:34:42 +02:00
Claudio Atzori 05f269a1c0 kryo based parallel implementation of CreateRelatedEntitiesJob_phase2, now works by OafType; introduced custom aggregator in AdjacencyListBuilderJob 2020-06-01 00:32:42 +02:00
Claudio Atzori 6f5f498c78 restored common properties driving executor-cores and executor-memory in join_organization_relations wf node 2020-05-29 11:22:00 +02:00
Claudio Atzori b2f9564f13 WIP: fixed PrepareRelationsJob; parallel implementation of CreateRelatedEntitiesJob_phase2, now works by OafType; introduced custom aggregator in AdjacencyListBuilderJob 2020-05-29 10:58:15 +02:00
Claudio Atzori 5dea155a87 increased number of partitions produced by the join_all_entities phase as well as spark.sql.shuffle.partitions in adjancency_lists phase 2020-05-28 13:49:59 +02:00
Claudio Atzori cfd753217c repartition the join_entities in 24k files 2020-05-27 12:44:01 +02:00
Claudio Atzori 4e36d689dd fixed XML serialization for children sub-elements (duplicates & externalreferences) 2020-05-26 18:30:40 +02:00
Claudio Atzori b33dd58be4 replaced parameter 'reuseRecords' with 'resumeFrom', allowing to restart the provision workflow execution from any step, useful for manual submissions or debugging 2020-05-22 08:50:06 +02:00
Claudio Atzori dbfb9c19fe minor changes 2020-05-21 10:00:14 +02:00
Claudio Atzori d7d2a0637f added extra parameters to the provision indexing workflow 2020-05-20 14:55:38 +02:00
Claudio Atzori 0bdfbb0a57 reintroduced RDD based relation cut off procedure 2020-05-19 15:02:21 +02:00
Claudio Atzori 8c67073a07 force speculative execution to false 2020-05-08 09:42:21 +02:00
Claudio Atzori bac37b3973 fixed children expansion in XML records 2020-05-04 11:51:17 +02:00
Claudio Atzori 0b55795d4d small adjustments in the provisioning workflow 2020-04-21 16:15:04 +02:00
Claudio Atzori 77f59b1b10 dataset based provision WIP 2020-04-06 19:37:27 +02:00
Claudio Atzori ca345aaad3 dataset based provision WIP 2020-04-06 15:33:31 +02:00
Claudio Atzori c8f4b95464 dataset based provision WIP 2020-04-06 08:59:58 +02:00
Claudio Atzori eb2f5f3198 dataset based provision WIP 2020-04-04 17:41:31 +02:00
Claudio Atzori 3d1b637cab dataset based provision WIP 2020-04-04 14:03:43 +02:00
Claudio Atzori 24b2c9012e dataset based provision WIP 2020-04-02 18:44:09 +02:00
Claudio Atzori daa26acc9d dataset based provision WIP, fixed spark2EventLogDir 2020-04-02 16:15:50 +02:00
Claudio Atzori 9c7092416a dataset based provision WIP 2020-04-01 19:07:30 +02:00
Claudio Atzori adcdd2d05e WIP: reimplementing the adjacency list construction process using spark Datasets 2020-04-01 14:56:57 +02:00
Claudio Atzori 0fbec69b82 use oozie prepare statement to cleanup working directories 2020-03-30 19:48:41 +02:00
Claudio Atzori f3f9affd49 allow dynamic executors to build XML records 2020-03-30 13:12:11 +02:00
Claudio Atzori 673e744649 moved openaire specific implementations under dedicated package eu.dnetlib.dhp.oa 2020-03-27 10:42:17 +01:00
Claudio Atzori abe8fb69a2 added global properties, moved postprocessing script inside the oozie_app directory 2020-03-18 15:43:54 +01:00
Claudio Atzori 1e563bc15e introduced distinct properties driving the resouce usage for the XML record creation and the indexing phase 2020-03-04 10:55:11 +01:00
Claudio Atzori bc7cfd5975 indexing workflow WIP: fixed projects fundingtree xml conversion, prioritized links between results and projects when limiting them to 100 in the join procedure 2020-03-02 17:03:07 +01:00
Claudio Atzori 56d1810a66 working procedure for records indexing using Spark, via lib com.lucidworks.spark:spark-solr 2020-02-14 12:28:52 +01:00
Claudio Atzori 1fee6e2b7e implemented XML records construction and serialization, indexing WIP 2020-02-13 16:53:27 +01:00
Claudio Atzori 49ef2f4eb1 removed input parameter specification, SparkXmlRecordBuilderJob doesn't need hive 2020-01-30 18:20:26 +01:00
Claudio Atzori b2691a3b0a save adjacency list as JoinedEntity 2020-01-30 17:46:29 +01:00
Claudio Atzori 799929c1e3 joining entities using T x R x S method with groupByKey 2020-01-21 16:35:44 +01:00
Claudio Atzori 97c239ee0d WIP: trying to find a way to build the records for the index 2020-01-16 12:02:28 +02:00
Claudio Atzori 7ba586d2e5 oozie workflow aimed to build the adjacency lists representation of the graph, needed to build the records to be indexed 2019-12-17 16:24:49 +01:00