diff --git a/README.md b/README.md
index 4265f94f..7e81d907 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
 Docker Compose Files
 ===
-Some typical docker compose templates.
+Some typical docker compose examples.
 
 # Install Docker and Docker Compose
 Take ubuntu for example
@@ -14,11 +14,17 @@ $ sudo pip install docker-compose
 See [https://docs.docker.com/compose/](https://docs.docker.com/compose/).
 
 
-# templates
+# Examples
 
 ## consul-discovery
 Using consul to make a service-discoverable architecture.
 
+## elk_netflow
+Elk cluster, with netflow support.
+```sh
+docker-compose scale es=3
+```
+
 ## mongo_cluster
 Start 3 mongo instance to make a replica set.
 
@@ -38,8 +44,8 @@ Use nginx as a proxy with authentication for backend application.
 ## registry_mirror
 docker registry mirror, with redis as the backend cache.
 
-## elk
-Elk cluster, with netflow support
+## spark_cluster
+Spark cluster with master and worker nodes
 ```sh
-docker-compose scale es=3
+docker-compose scale worker=2
 ```
diff --git a/spark_cluster/docker-compose.yml b/spark_cluster/docker-compose.yml
new file mode 100644
index 00000000..013e56ec
--- /dev/null
+++ b/spark_cluster/docker-compose.yml
@@ -0,0 +1,26 @@
+# http://github.com/yeasy/docker-compose-files
+# This compose file will start spark master node and the worker node.
+# All nodes will become a cluster automatically.
+# You can run: docker-compose scale worker=2
+
+master:
+  image: sequenceiq/spark:1.4.0
+  hostname: master
+  ports:
+  - "4040:4040"
+  - "8042:8042"
+  - "7077:7077"
+  - "8088:8088"
+  - "8080:8080"
+  restart: always
+  #mem_limit: 1024m
+  command: bash /usr/local/spark/sbin/start-master.sh && ping localhost > /dev/null
+
+worker:
+  image: sequenceiq/spark:1.4.0
+  links:
+  - master:master
+  expose:
+  - "8081"
+  restart: always
+  command: bash /usr/local/spark/sbin/start-slave.sh spark://master:7077 && ping localhost >/dev/null