diff --git a/aztk/spark/helpers/create_cluster.py b/aztk/spark/helpers/create_cluster.py
index 7bb3f137..aa2bd128 100644
--- a/aztk/spark/helpers/create_cluster.py
+++ b/aztk/spark/helpers/create_cluster.py
@@ -54,6 +54,7 @@ def __docker_run_cmd(docker_repo: str = None, gpu_enabled: bool = False, file_mo
     cmd.add_option('-e', 'SPARK_JOB_UI_PORT=$SPARK_JOB_UI_PORT')
     cmd.add_option('-p', '8080:8080')       # Spark Master UI
     cmd.add_option('-p', '7077:7077')       # Spark Master
+    cmd.add_option('-p', '7337:7337')       # Spark Shuffle Service
     cmd.add_option('-p', '4040:4040')       # Job UI
     cmd.add_option('-p', '8888:8888')       # Jupyter UI
     cmd.add_option('-p', '8787:8787')       # Rstudio Server
diff --git a/config/AppRegistrations_2.png b/config/AppRegistrations_2.png
deleted file mode 100644
index b9e61db0..00000000
Binary files a/config/AppRegistrations_2.png and /dev/null differ
diff --git a/config/spark-defaults.conf b/config/spark-defaults.conf
index 415df8dc..683b41aa 100644
--- a/config/spark-defaults.conf
+++ b/config/spark-defaults.conf
@@ -25,9 +25,14 @@
 # spark.driver.memory              5g
 # spark.executor.extraJavaOptions  -XX:+PrintGCDetails -Dkey=value -Dnumbers="one two three"
 
+# set "scratch" space for Spark
+spark.local.dir                 /mnt/batch/
+
 # Note: Aztk pre-loads wasb jars, so loading is not necessary
 spark.jars                      /home/spark-current/jars/azure-storage-2.0.0.jar,/home/spark-current/jars/hadoop-azure-2.7.3.jar
 
 # Note: Default filesystem master HA
 spark.deploy.recoveryMode       FILESYSTEM
 spark.deploy.recoveryDirectory  /root/
+
+spark.shuffle.service.enabled   true
diff --git a/node_scripts/install/install.py b/node_scripts/install/install.py
index 3c0ef495..f011f1c4 100644
--- a/node_scripts/install/install.py
+++ b/node_scripts/install/install.py
@@ -29,7 +29,7 @@ def setup_node():
     else:
         setup_as_worker()
         scripts.run_custom_scripts(is_master=False, is_worker=True)
-    
+
     open("/tmp/setup_complete", 'a').close()
 
 
diff --git a/node_scripts/install/spark.py b/node_scripts/install/spark.py
index 71805d12..11ed2342 100644
--- a/node_scripts/install/spark.py
+++ b/node_scripts/install/spark.py
@@ -114,6 +114,15 @@ def start_spark_worker():
     print("Connecting to master with '{0}'".format(" ".join(cmd)))
     call(cmd)
 
+    # enable the shuffle service on all slaves
+    start_shuffle_service()
+
+
+def start_shuffle_service():
+    exe = os.path.join(spark_home, "sbin", "start-shuffle-service.sh")
+    print("Starting the shuffle service with {}".format(exe))
+    call([exe, " &"])
+
 
 def copyfile(src, dest):
     try: