This guide explains how to connect to the CrySP RIPPLE compute facility. Additionally, we show a way of how to obtain a comfortable deployment specifically for projects using Python, like executing a Jupyter notebook and connecting via PyCharm. For all others, we only show how to mounting folders via sshfs.

How to Use RIPPLE

Setup RIPPLE

You have to generate an ssh public and private key and send only the public key to Prof. Goldberg. Make sure to put both keys into the "~/.ssh/" folder of your Linux machine. In this tutorial, we assume you named these keys id_rsa and id_rsa.pub.

Connect to RIPPLE directly

RIPPLE can only be accessed through a computer with a static IP that has been whitelisted. This could be the student server "linux.student.cs.uwaterloo.ca" if you are not from CrySP or if you do not have a computer with a static IP.

Connect to RIPPLE through a Jump Server

After booking a RIPPLE computer we can only connect to it via a whitelisted jump server. If you are not from CrySP or do not have a computer this is

most likely the student server with hostname "linux.student.cs.uwaterloo.ca". Otherwise the hostname is your computers static ip address.

Step 1: Figure out your algorithm

Step 2: Understand your algorithm's costs

Step 3: Choose appropriate hardware

Step 4: Implement your algorithm

Step 5: Book and access the hardware

(Information about how to connect to RIPPLE goes here)

Step 6: Monitor your experiment