forked from D-Net/dnet-hadoop
[dedup] increased number of partitions in the consistency phase
This commit is contained in:
parent
bafa2990f3
commit
0a727d325d
|
@ -89,7 +89,7 @@
|
||||||
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
||||||
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
||||||
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
||||||
--conf spark.sql.shuffle.partitions=7680
|
--conf spark.sql.shuffle.partitions=15000
|
||||||
</spark-opts>
|
</spark-opts>
|
||||||
<arg>--graphBasePath</arg><arg>${graphBasePath}</arg>
|
<arg>--graphBasePath</arg><arg>${graphBasePath}</arg>
|
||||||
<arg>--o</arg><arg>${graphOutputPath}</arg>
|
<arg>--o</arg><arg>${graphOutputPath}</arg>
|
||||||
|
@ -114,7 +114,7 @@
|
||||||
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
||||||
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
||||||
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
||||||
--conf spark.sql.shuffle.partitions=7680
|
--conf spark.sql.shuffle.partitions=15000
|
||||||
</spark-opts>
|
</spark-opts>
|
||||||
<arg>--graphInputPath</arg><arg>${graphBasePath}</arg>
|
<arg>--graphInputPath</arg><arg>${graphBasePath}</arg>
|
||||||
<arg>--outputPath</arg><arg>${workingPath}/grouped_entities</arg>
|
<arg>--outputPath</arg><arg>${workingPath}/grouped_entities</arg>
|
||||||
|
|
Loading…
Reference in New Issue