Skip to content

Latest commit

 

History

History
63 lines (41 loc) · 2.79 KB

using.rst

File metadata and controls

63 lines (41 loc) · 2.79 KB
Raw

Using the Turing's Cray Urika-GX service

This chapter provides miscellaneous hints and tips on using the Turing's Cray Urika-GX service ("Urika").

See the chapter connecting for instructions on how to connect to Urika.

View documentation

Detailed documentation on Urika and its software and services is available on the Urika home page:

  1. Within a web browser, visit http://urika1.turing.ac.uk/home (see Use web browser to access Urika's applications software).
  2. Click on Learning Resources on the home page, or visit http://urika1.turing.ac.uk/documentation.

View Best Practice Guide

For a guide on best practices in the use of Urika, please see the PRACE "Best Practice Guide - HPC for Data Science on the Cray Urika", January 2019: HTML PDF

Submit jobs

To access notes on submitting jobs using the various job submission mechanisms available on Urika:

  1. view-doc as described above.
  2. Scroll down the page and click on "Job Submission Notes".

Alternatively, within a web browser, visit http://urika1.turing.ac.uk/static/documentation/notebooks/ATI-Job-Submission.pdf.

Open Jupyter notebook

  1. Within a web browser, visit http://urika1.turing.ac.uk/home (see Use web browser to access Urika's applications software).
  2. Click on the "jupyter" icon on the home page.
  3. A Jupyter sign-in page will open in a new browser tab, prompting you for a Username and Password. Enter your Urika username and password.
  4. Click Sign In.
  5. A Jupyter notebook page will appear in a new browser tab, displaying the contents of your home directory on Urika.

Open Apache Mesos resource manager

  1. Connect to Urika via the command-line.

  2. Get your Mesos password by running the following command:

    cat /security/secrets/<your-urika-username>.mesos
  3. Within a web browser, visit http://urika1.turing.ac.uk/home (see Use web browser to access Urika's applications software).
  4. Scroll down the home page until you see a carousel of icons which includes "MESOS".
  5. Click on the "MESOS" icon to select it.
  6. Click on the "MESOS" icon again.
  7. The Mesos dashboard will open in a new browser tab, with a dialog prompting you for a User Name and Password. Enter your Urika username and Mesos password.
  8. The Mesos dashboard will refresh with its current data.