Introduction to HTCondor

This tutorial is a part of GridKa School 2019 and will try to answer the question:

How to distribute your compute tasks and get results with high performance, keeping machines and site admins joyful?

The documentation on this page is part of this repository which also contains files used in the part of the course.

Structure

The tutorial is started with an introductory presentation and complemented by this very documentation and files provided in the repository.

The following outline links to the various exercises of the course.

Outline

1. Accessing the test cluster
2. An interactive job
3. Resource Requests
4. Submitting a first batch job
5. Submitting many jobs
6. Queueing more cleverly
7. Running a DAG

Explanation of common icons

This marks a task.

This marks extra information, or things you may want to investigate to learn some more details.

This highlights a good point to ask questions if something is unclear. Of course, you can ask at any point!

This marks bonus tasks for those who are faster than their colleagues.

Good practice.

Known (often common) bad practice.

Licensing

There are two directories of the repository shared under a different license than the rest of this tutorial:

• files/povray/mini_demo
• files/povray/dice

These are shared under Creative Commons Attribution-Share Alike 3.0 Unported . Detailed attribution is given in the README.md inside each folder, and during the course when the artwork is used.