diff --git a/ansible/roles/monasca/defaults/main.yml b/ansible/roles/monasca/defaults/main.yml
index 0c8b717f0e388541b676b2cdb36e0c5ebb96097c..55c2f125dc2d6da6d7dcad4e2b14f822345692b7 100644
--- a/ansible/roles/monasca/defaults/main.yml
+++ b/ansible/roles/monasca/defaults/main.yml
@@ -64,6 +64,14 @@ monasca_metrics_topic: "metrics"
 monasca_raw_logs_topic: "logs"
 monasca_transformed_logs_topic: "transformed-logs"
 
+# Processing pipeline threads. In a large scale deployment you will likely
+# want to tune these with finer precision. For example, if you have a very
+# high log throughput, the log metrics service consumer may require a
+# higher thread count than the producer. You will also want to ensure that
+# the total number of threads across all instances of a service does not
+# exceed the Kafka topic partition count.
+monasca_log_pipeline_threads: 2
+
 ####################
 # Docker
 ####################