here are a few spark notebooks for trying out the technology. I am running the,m with docker on MAcOS
how to set up (on MacOs):
- get Docker for Mac - get the beta release here
- start with this script:
$ ./start.sh
- get Docker for Mac - get the beta release here
- get scala notebook image (or newer version from http://spark-notebook.io/, e.g.:
$ docker pull andypetrella/spark-notebook:0.6.3-scala-2.11.7-spark-1.6.1-hadoop-2.7.2-with-hive-with-parquet
- put your data files into the /data subdirectory (data files are excluded from git to keep them private)
- start up with:
$ docker run -p 9000:9000 -p 4040-4045:4040-4045 -v `pwd`/data:/opt/docker/data -v `pwd`/notebooks:/opt/docker/notebooks/mysnb andypetrella/spark-notebook:0.6.3-scala-2.11.7-spark-1.6.1-hadoop-2.7.2-with-hive-with-parquet
With this command you expose the right set of ports (9000, 4040-4045) and mount ./data and ./notebooks folders into the container. You will find your own notebooks in the ./mysnb folder when you open the notebook with the browser on http://localhost:9000 or with
$ open http://localhost:9000