Doctoral Academy Programme Bookings


Welcome Guest
Home > Browse > IT Skills > ARC: Advanced Use of 'Hawk' - the Supercomputer > Details
In order to register for a session, you must first log in. You will be returned to the workshop session, once logged in.
Details
ARC: Advanced Use of 'Hawk' - the Supercomputer
Domain: A1 Knowledge base

TACE: A - Acquisition

Format: Workshop

Suitable for: All

The Hawk supercomputer can be used in lots of different ways – this course aims to review some of the techniques that should help users get the best out of the cluster for their own research.
This course will cover a number of features of the system that should help improve the performance of jobs submitted to the cluster, including: using advanced features in the Slurm Job scheduler (including array jobs); learn how to profile and debug your job, and set the correct resource requirements to enable faster turnaround on the supercomputer; develop more efficient job scripts to run your software; provide an overview of the high performance Lustre filesystem and associated best practices when using this partition; use of the GPU partition; and if requested, an overview of revision control software such as Git and importance of documentation and comments in code development.

To attend this work, you must have attended ‘ARC: Supercomputing for Beginners’ or actively be using the Hawk supercomputer.

Presenters


Session