forked from D-Net/dnet-hadoop
added spark.sql.shuffle.partitions in the last join phase of the result to community through semantic relation propagation
This commit is contained in:
parent
13cc592f39
commit
8742934843
|
@ -260,6 +260,7 @@
|
|||
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
||||
--conf spark.dynamicAllocation.enabled=true
|
||||
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
|
||||
--conf spark.sql.shuffle.partitions=5000
|
||||
</spark-opts>
|
||||
<arg>--preparedInfoPath</arg><arg>${workingDir}/preparedInfo/mergedCommunityAssoc</arg>
|
||||
<arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>
|
||||
|
@ -289,6 +290,7 @@
|
|||
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
||||
--conf spark.dynamicAllocation.enabled=true
|
||||
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
|
||||
--conf spark.sql.shuffle.partitions=5000
|
||||
</spark-opts>
|
||||
<arg>--preparedInfoPath</arg><arg>${workingDir}/preparedInfo/mergedCommunityAssoc</arg>
|
||||
<arg>--sourcePath</arg><arg>${sourcePath}/dataset</arg>
|
||||
|
@ -318,6 +320,7 @@
|
|||
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
||||
--conf spark.dynamicAllocation.enabled=true
|
||||
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
|
||||
--conf spark.sql.shuffle.partitions=2000
|
||||
</spark-opts>
|
||||
<arg>--preparedInfoPath</arg><arg>${workingDir}/preparedInfo/mergedCommunityAssoc</arg>
|
||||
<arg>--sourcePath</arg><arg>${sourcePath}/otherresearchproduct</arg>
|
||||
|
@ -347,6 +350,7 @@
|
|||
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
||||
--conf spark.dynamicAllocation.enabled=true
|
||||
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
|
||||
--conf spark.sql.shuffle.partitions=1000
|
||||
</spark-opts>
|
||||
<arg>--preparedInfoPath</arg><arg>${workingDir}/preparedInfo/mergedCommunityAssoc</arg>
|
||||
<arg>--sourcePath</arg><arg>${sourcePath}/software</arg>
|
||||
|
|
Loading…
Reference in New Issue