As mentioned in a previous blog post, my parallel tempering code is now working. There is still work to be done in adding new physics, but I am now in a position to benefit from some extra computing power. I’m going to try using Amazon EC2 for this project.
First, I launched an instance running the Amazon Linux AMI (HVM). This blog post may be helpful in setting things up.
Once the instance was created, and my keys were all setup, I SSHed into my instance. My favorite distribution of scientific python is anaconda python. This can be easily downloaded onto the AWS virtual machine instance by
- going to the Anaconda downloads page
- selecting the linux OS
- right-clicking in the browser on the “Linux 64-Bit - Python 2.7” button
- copying the link to the downloadable package
- and pasting the link after
wgetin the SSH session into the VMI.
Once the download has completed, installation is as simple as typing ls to find out the name of the install script, running it, and following the on-screen instructions.
If you said ‘yes’ at the end of installation to have the anaconda python path appended to your .bashrc file, then starting anaconda python is as simple as sourcing the changes to your .bashrc file and typing python:
There are a few packages I need to run my parallel tempering code. I’ll use pip to install these: