Skip to main content

Latest Tweets

  • New post: High-performance computing for Economists 2017 edition
    about 4 weeks ago via VirtualRDC
  • New post: ANNOUNCEMENT: EMERGENCY DOWNTIME on ecco cluster July 21-22 2017
    about 4 weeks ago via VirtualRDC
  • Page edited: World Statistical Congress 2015 (ISI2015) session on "Synthetic establishment microdata around the
    about 3 months ago via VirtualRDC

[ECCO+SDS] Customizing Python environment on ECCO and SDS

Print Friendly, PDF & Email

We updated our Anaconda Python environment yesterday. This note is how to customize that environment for your own research, on ECCO and SDS.

To do so is straightforward. Look up the basics about how to create Anaconda environments here and here. You want to create your personal environment in $HOME (which nobody else can access) and your collaborative research environment (that you share with your co-authors on ECCO) in /ssgprojects/(PROJECT_PATH) (which can and is shared with others in your group). The example below uses the former, but you can easily see how to extend that to the shared directories.

First, (in an interactive session), load the Anaconda Python module:

module load python/anaconda

Second, create the new environment:

conda create -p $HOME/anaconda/envs/nobelprize anaconda

(anaconda as the last argument means it will replicate the entire root environment, which as of this writing is Python 2.7 plus a wide number of packages).

Third, use the following command in your qsub scripts:

source /etc/profile.d/
module load python/anaconda
source activate $HOME/anaconda/envs/nobelprize

Fourth, that's it. Enjoy.