JetLag v0.1: The Supercomputing Demo

During our recent Phylanx Kickoff meeting, the team defined a new project to drive our development which we donned the code name JetLag. JetLag represents the production interface of the Phylanx project which incorporates all of its elements behind a user-friendly Jupyter Notebook encapsulated in a Docker image. A user will download the Docker image which includes the entire software stack of the project. They can then open a Jupyter Notebook which will allow them to begin writing and testing decorated NumPy code which will take advantage of the Phylanx. After they are satisfied, they can set up the Tapis backend and then deploy their solution on a cluster where they have access. After any execution of their code (either local and remote) They can visualize the performance results of the run through a link using Traveler tools.

These features represent all of the major deliverables of our project. We were able to get the first version of this running as a demo at Supercomputing 2019 in Denver, Colorado. You can see a recorded demonstration of the technology here: https://cct.lsu.edu/~sbrandt/PhylanxDemo.mp4

We are very excited to share these first results and are looking to release the instructions to build JetLag soon. Stay posted!

 

Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *