HPC research

EPCC wins HPC Innovation Excellence Award

Author: Adrian Jackson
Posted: 24 Jun 2014 | 14:10

Electrostatic potential fluctuations in an annular region at mid-radius in the MAST tokamak, from a gyrokinetic simulation of the saturated turbulence using the GS2 code. A wedge of plasma has been removed from the visualisation so as to view the nature of the fluctuations inside the annulus.EPCC is delighted to be part of a team that has won an HPC Innovation Excellence Award. Presented at the International Supercomputing Conference (ISC14) in Leipzig (22-26 June 2014), the awards recognise outstanding application of HPC Computing for Business and Scientific Achievements.

EPCC at ISC'14: come and say hello!

Author: Nick Brown
Posted: 13 Jun 2014 | 12:06

EPCC will be exhibiting in Booth 920.

International Supercomputing Conference (ISC) is one of the main events in the field of high performance computing, networking and storage. It has a diverse programme of keynote talks, research presentations, tutorials, BoFs and satellite events provided by 300 expert speakers along with over 170 exhibitors and 3,000 attendees. As usual, there will be a significant EPCC representation.

When applications go exascale — the CRESTA project

Author: Guest blogger
Posted: 10 Feb 2014 | 09:22

Dr Jason Beech-Brandt, Manager Exascale Research, Europe at Cray writes about the CRESTA project, which is addressing the challenges of exascale computing.

Seymour Cray, the pioneer of supercomputing, famously asked if you would rather plough a field with two strong oxen or 1024 chickens.

Pervasive Parallelism PhD studentships at EPCC

Author: Mark Bull
Posted: 12 Dec 2013 | 11:54

EPCC was recently successful in its joint bid with Informatics for an EPSRC Centre for Doctoral Training in Pervasive Parallelism. The Centre aims to develop research leaders with deep knowledge and awareness across the parallelism spectrum. To support this, it will support about 10 fully-funded, four-year studentships each year over the next 5 years. The majority will be based in Informatics, but we expect 1 or 2 per year to be available at EPCC.

Research in a nutshell: HPC

Author: Iain Bethune
Posted: 2 Dec 2013 | 10:13

In this short video 'Research in a nutshell: HPC', I describe my investigations into how high performance computing can be used to go beyond the limits of theory and experiment in many fields of scientific research.

Also in this series, you can watch Prof. Arthur Trew's Computational Science video where he explains the use of computers to simulate systems that are too large, too fast, too slow, or perhaps just too expensive to experiment on.

Energy efficient supercomputing

Author: Nick Johnson
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.

Kicking-off Adept!

Author: Michele Weiland
Posted: 30 Sep 2013 | 15:58

Adept, EPCC’s latest EU project, started officially on the 1st September and we held our kick-off meeting in Edinburgh a couple of weeks ago. The project partners travelled from Sweden (Uppsala University and Ericsson) and Belgium (Ghent), as well as from just across town (Alpha Data) to take part.

McMPI at the EuroMPI 2013 conference

Author: Daniel Holmes
Posted: 26 Sep 2013 | 10:05

Following my presentation about McMPI at the EuroMPI 2013 conference last week, some people asked me to post the slides. The presentation and the associated paper give a brief introduction to the McMPI software and quickly cover some of my thoughts about threading in MPI.

EPCC joins the OpenACC consortium

Author: Adrian Jackson
Posted: 18 Jul 2013 | 14:00

EPCC has recently joined the OpenACC consortium. OpenACC, a directives-based parallel programming standard, is designed to simplify the programming and utilisation of heterogenous computer systems, where standard CPUs and accelerators (such as GPUs) are combined.   Using OpenACC, developers can specify loops and regions of code in standard C, C++ and Fortran to be offloaded from a CPU to an attached accelerator, and also target parallelism on a range of different CPUs and accelerators without having to modify source code.