At DiRAC, we believe one of our most important responsibilities to our user-base is to offer comprehensive training. The better trained our researcher cohort, the more efficiently we can use our HPC systems, thereby increasing our scientific productivity and generating more research from our existing systems. In addition we can deploy more bleeding-edge and innovative HPC technology in our services, increasing our capability and energy efficiency, and enabling calculations which would otherwise not be feasible.
On this page you can find information on the DiRAC Essentials Level Training and other External Opportunities that will be useful to our users and community.
The DiRAC Essentials Level Training (formerly the DiRAC Driving Licence), is a basic introduction to the principles of HPC and the tools needed to work on an HPC system. The course consists of seven modules covering everything from the Unix Environment to Good Network Practices and is presented as a series of links to external sources that we have selected for you. After working through the modules you will know the Essentials of how to work with, and get the best out of, DiRAC’s HPC systems.
The training is mandatory for those studying for advanced degrees, such as M.Sc. and Ph.D., and those employed as Post-Doctoral Research Assistants/Associates who have less than two years HPC experience. The Essentials Level culminates in a basic on-line test and the training material covers all aspects of HPC competency that are included.
Information on accessing the training and registering for a training account can be found here.
Please see below for information on some useful training and networking events that are being hosted external to DiRAC:
Events co-located with the 8th Annual DiRAC Day
Our 8th Annual DiRAC Day event is being held at Swansea University on 12th September 2018 and this year there are three other interesting co-located training and networking events:
Machine Learning Workshop – Applications in Astronomy
10th – 12th September
We are being deluged by data from surveys, taking us far beyond the possibility for human management, even with assistance from citizen science projects. This has motivated a rapidly increasing interest in the use of machine learning to aid in the handling of such data, including feature extraction and statistical inference in particular. The rise of deep learning algorithms such as convolutional neural networks means that many problems associated with large volumes of image data can be largely removed.
This workshop is designed to encourage those who are using such techniques to come together to share their experiences and present ideas on novel applications of machine learning in astronomy. We especially encourage talks on how these methods can be taken forward to new technologies such as LSST, JWST and EUCLID.
The first day will be a tutorial/educational day with longer talks from various invited speakers on general ways to use and improve machine learning techniques. The second two days are a regular workshop with presentations and posters from delegates. Each day includes a buffet lunch and refreshments at break time. There will be an optional meal at a local Indian restaurant on the Tuesday night. See the Workshop Website for registration and further details.