November 2021

Introduction to Biomedical Image Registration and the Parallel Framework for Image Registration

Author: Gavin Pringle
Posted: 25 Nov 2021 | 16:54

The last CompBioMed e-Seminar of 2021 will take place on the 9th of December, 2021, at 2pm GMT (3pm CET) and it will be an "Introduction to Biomedical Image Registration and the Parallel Framework for Image Registration".  Later, in May 2022, we'll hold an associated hands-on tutorial.

EuroCC@UK: the UK’s National Competence Centre in HPC, HPDA and AI

Author: Lorna Smith
Posted: 8 Nov 2021 | 08:55

EPCC leads EuroCC@ UK in collaboration with STFC Hartree. Both centres are drawing on their complementary expertise to deliver the programme’s goals.

The UK is an advanced country in the context of HPC, with a long history of infrastructure provision and computational science research by academia and industry. Both EPCC and the Hartree Centre have significant experience in supporting and collaborating with academia and industry. Since it was established in 1990, EPCC has collaborated with UK science and industry, as well as providing high-quality accessible training courses. The Hartree Centre is funded to work primarily with industry and has undertaken more than 160 collaborative projects with companies since its foundation in 2012, with its training activities focused on meeting the needs of UK industry.

A brief history of the ARCHER2 Image and Video Competition

Author: Clair Barrass
Posted: 6 Nov 2021 | 08:24

The competition gathers together some of the best images produced by users of the ARCHER2 UK national supercomputing service, which is hosted and managed by EPCC. 

Back in autumn 2014, just a couple of months after I had joined EPCC, I was invited to set up and launch an image competition to showcase the work being done on ARCHER, the UK’s national supercomputing service at that time. Having been immersed in the world of HPC for only a few weeks, I honestly had no idea what to expect. My assumption was that computational research, scientific simulations and crunching of complex equations didn’t really equate with aesthetically pleasing images, artwork and things of beauty. My work centred mainly on service users’ helpdesk, helping to set up projects and passing on problems with codes that weren’t running as hoped to our technical team, and also with administering our training programme, getting new users up and running. I hadn’t really seen much of the kinds of outputs the successful work could lead to.

Two decades of MSc programmes at EPCC

Author: David Henty
Posted: 4 Nov 2021 | 09:38

Back in the late 1990’s we became aware of a need for postgraduate level training in high-performance computing (HPC) and parallel programming. We already had a good portfolio of material from the courses we delivered as part of the UK national supercomputing services (at the time it was based around the Cray T3D/T3E), and when the Engineering and Physical Sciences Research Council (EPSRC) offered Masters Training Programme grants in 2000, we took the opportunity to build on this material and successfully applied for funding to develop an MSc in HPC.

Enabling machine learning for exascale simulations

Author: Anna Roubickova
Posted: 2 Nov 2021 | 14:01

SiMLInt is an interface that increases the efficiency of machine learning (ML) techniques when solving large-scale physical simulations by consuming fewer resources without sacrificing precision.

In a recent breakthrough, Kochkov et al. at Google Research demonstrated the successful use of Convolutional Neural Networks (CNNs) to solve systems of partial differential equations (PDEs) that describe  the evolution of complex physical systems, such as turbulent flows, while using considerably less computational resource.


Blog Archive