Switch branches/tags
Nothing to show
Find file Copy path
b2756e6 Aug 2, 2017
1 contributor

Users who have contributed to this file

80 lines (55 sloc) 2.09 KB

Spinning up multiple Jupyter Notebooks on AWS for a tutorial

(Setup instructions for the sourmash tutorial)

boot an ubuntu/images-testing/hvm-ssd/ubuntu-yakkety-daily-amd64-server-20170201 (ami-008dd060) image, with a ~50 GB local disk and with open ports 8000-9000.

Then, update software:

sudo apt-get -y update

and install docker:

wget -qO- | sudo sh

Also yourself to the docker group:

sudo usermod -aG docker ubuntu

and now log out and log back in.

Next, create a dockerfile that builds on jupyter/datascience-notebook to install the stuff that I want in my custom environment:

cd ~/
mkdir sourmash-nb
cd sourmash-nb
cat > Dockerfile <<EOF
FROM jupyter/datascience-notebook

RUN pip install -U

and run docker build:

docker build -t sourmash-nb .

While that's running, you can also download the sourmash database:

mkdir ~/sourmash-db
cd ~/sourmash-db
curl -O
tar xzf microbe-genbank-sbt-k31-2017.05.09.tar.gz

Now run the notebook on port 8000, linking the large database into the container file system space as /data/genbank (and making it read-only) --

docker run -d -v ~/sourmash-db:/data/genbank:ro -p ${EXT}:8888 sourmash-nb \

Voila, you should now be able to connect to port 8000!

To run more than one, use a for loop:

for EXT in 8000 8001 8002
docker run ...

See the jupyter/datascience-notebook docs for more info on command line arguments to the container.

To kill the docker container, locate its ID and use docker kill <id>. (The ID is output by docker run when -d is used.) If you're running only one container, you can also replace -d with -it and it will run in interactive mode where you can use CTRL-C to quit out of it.