From 8742934843df1de1ff9163a7f4f2a5188dba3db3 Mon Sep 17 00:00:00 2001
From: Claudio Atzori <claudio.atzori@isti.cnr.it>
Date: Fri, 18 Nov 2022 11:32:22 +0100
Subject: [PATCH 1/2] added spark.sql.shuffle.partitions in the last join phase
 of the result to community through semantic relation propagation

---
 .../dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml    | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml b/dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml
index 81b51443c..26c7d1788 100644
--- a/dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml
@@ -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>

From ed64618235caec74176ee8b5046b2bf8c8a808df Mon Sep 17 00:00:00 2001
From: Claudio Atzori <claudio.atzori@isti.cnr.it>
Date: Fri, 18 Nov 2022 16:06:51 +0100
Subject: [PATCH 2/2] increased spark.sql.shuffle.partitions in the last join
 phase of the result (publication) to community through semantic relation
 propagation

---
 .../dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml b/dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml
index 26c7d1788..58d5bb26d 100644
--- a/dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-enrichment/src/main/resources/eu/dnetlib/dhp/resulttocommunityfromsemrel/oozie_app/workflow.xml
@@ -260,7 +260,7 @@
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
                 --conf spark.dynamicAllocation.enabled=true
                 --conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
-                --conf spark.sql.shuffle.partitions=5000
+                --conf spark.sql.shuffle.partitions=10000
             </spark-opts>
             <arg>--preparedInfoPath</arg><arg>${workingDir}/preparedInfo/mergedCommunityAssoc</arg>
             <arg>--sourcePath</arg><arg>${sourcePath}/publication</arg>