Posted: 29 Nov 2013 | 11:00
ARCHER (Advanced Research Computing High End Resource) is the next national HPC service for academic research. The service comprises a number of components: accommodation provided by the University of Edinburgh; hardware by Cray; systems support by EPCC and Daresbury Laboratory; and user and computational science and engineering support by EPCC.
Posted: 28 Nov 2013 | 10:00
My colleague Mike Jackson recently posted about the DiRAC driving test. DiRAC is the UK's integrated supercomputing facility for theoretical modelling and HPC-based research in particle physics, astronomy and cosmology and is used by numerous researchers with diverse backgrounds. Whilst much of their work is very different, one commonality is that it often requires in-depth technical and software engineering techniques. The idea of the driving test was therefore to ensure that all users have the required knowledge for effective use of the consortium’s machines.
Posted: 21 Nov 2013 | 16:24
In a previous post I introduced the European Exascale Software Initiative (EESI) Co-design deliverable in which EPCC is a joint task leader. We are looking to understand exactly what different people mean by the co-design term, what can be done to encourage and promote co-design and - specifically for exascale - how institutions might apply it to successfully deliver the HPC services of tomorrow.
Posted: 21 Nov 2013 | 11:05
If you have read any of my previous blogs, you will know that I have an interest in the lack of women working in the field of HPC and Physics. Last week I was lucky to be able to attend the third international Gender Summit in Washington DC, which aims to address, discuss and share ideas on how to improve the participation of women in science, technology, engineering and mathematics (STEM).
Posted: 21 Nov 2013 | 09:31
Posted: 19 Nov 2013 | 06:35
Today I attended the "First International Workshop on Energy Efficient Supercomputing (E2SC)"
It was really interesting (and confidence boosting) to find that what we are trying to do in the Adept project is similar in approach to that done by other research labs. The talks could be divided into two categories: modelling the effect of system parameters (such as cache) on energy and performance efficiency; and methods for measuring energy consumption.
Posted: 18 Nov 2013 | 10:20
For the last five years, I've been involved in the development of the CP2K materials science code, mostly in terms of improving its performance and scalability and porting to new systems. During that time a great deal of new functionality has been added, as well as additional input file options, libraries and autotuning capability to enable maximum performance on any given architecture. However, these improvements have come at the cost of increased complexity - and CP2K already has a fearsome reputation among researchers for being very hard to use!
Posted: 17 Nov 2013 | 16:02
Posted: 13 Nov 2013 | 14:27
EUDAT - the European Data Infrastructure project - has reached the end of its second year and has, with some success, distilled the first version of a common, collaborative, horizontal data infrastructure from among the vertical stacks of its various partners.