Blog

9th Workshop on Python for High-Performance and Scientific Computing

Author: Neelofer Banglawala
Posted: 6 Nov 2020 | 15:27

The 9th Workshop on Python for High-Performance and Scientific Computing (PyHPC) at SC20 will once again bring researchers and developers together to share their experiences of using Python across a wide range of disciplines and applications.

EPCC is a member of the PyHPC Organizing Committee and this year we are co-chairing PyHPC's 9th Workshop on Python for High-Performance and Scientific Computing at SC20 on Friday 13 Nov, 10:00–18:30 EST

HPC-Europa3 Transnational Access Meeting (TAM2020)

Author: Catherine Inglis
Posted: 5 Nov 2020 | 11:25

The HPC-Europa3 Transnational Access programme funds short collaborative research visits for computational scientists working in any discipline which can use HPC.

Former HPC-Europa visitors are invited to attend the Transnational Access Meeting (TAM), to present the results of their visits. With approximately 40% of visitors being postgraduates, for many it also provides a supportive environment for their first conference presentation.

Urgent computing: presentations at SC20

Author: Rupert Nash
Posted: 28 Oct 2020 | 14:27

 

EPCC leads a work package of the VESTEC (Visual Exploration and Sampling Toolkit for Extreme Computing) project, which seeks to fuse HPC and real-time data for urgent decision-making for disaster response. Our work has resulted in two accepted papers into SC20 workshops: one around our custom workflow execution system, and another on extending the Common Workflow Language standard to better support parallel application execution.

Our support for industry

Author: Thomas Blyth
Posted: 21 Oct 2020 | 11:26

EPCC continues to support industry in these challenging times, leveraging our three key foundations:

  • The hosting, provision and management of high-performance computing (HPC) and data facilities for academia and business.
  • Research and project support to optimise the computing activities of those organisations.
  • Our unique project-based Applications Group, which provides bespoke novel and high-performance software solutions to academia, industry and commerce.

EPCC and the Machine Intelligence Garage

Author: David Homan
Posted: 14 Oct 2020 | 10:32

During the past year EPCC has increased its engagement with the Machine Intelligence Garage, a UK Government programme that supports innovative startups which make use of machine-learning techniques and artificial intelligence (AI).

Part of the Machine Intelligence Garage's remit is to enable technical support, such as HPC services, which is provided by various partners. For example, its industrial partners include AWS, Google Cloud, and NVIDIA. EPCC is a founding partner and we also offer support and potential collaboration for interested startups.

Virtual event: Doing Data Together - Ethical collaboration through Covid-19 and beyond

Author: Tracy Peet
Posted: 13 Oct 2020 | 11:13

The University of Edinburgh will hold its third Data-Driven Innovation conference (held jointly with The Scotsman newspaper) on Friday 6 November 2020. Join experts from inside the University of Edinburgh and beyond to hear about the enormous challenges – and real opportunities – in our increasingly data-rich world. 

Deadline extended for UrgentHPC workshop at SC20

Author: Nick Brown
Posted: 31 Aug 2020 | 15:33

SC, the world's largest high-performance computing conference, will be held virtually during the week of November 9-13. It is a disappointment not to be going to SC in person, but the flip side is that the conference will be open to a wider audience than would have been possible had it been held in Atlanta as originally planned.

This year I am organising the second run of the Urgent HPC workshop, which is an event aimed at bringing together those who are researching the role of HPC and data science in making urgent decisions to tackle disasters. The event first ran last year at SC19 and comprised a keynote talk by the founder of Technosylva – the world’s leading wildfire simulation code development company – six technical papers, and a panel. Based upon that success we decided to run the workshop again this year, and given all that has happened since then, exploring this topic is more timely than ever before! 

On-demand computing and data solutions for industry at EPCC

Author: Thomas Blyth
Posted: 27 Aug 2020 | 13:54

Cirrus is a Tier-2 state-of-the-art advanced computing facility available to industry and academia. Hosted and managed by EPCC, it provides an ideal platform for solving computational, simulation, modelling, and data science challenges. 

The system was recently upgraded via a £3.5m funding award from the UK's Engineering and Physical Sciences Research Council. This included the addition of 144 NVIDIA V100 GPUs and a 256TB high-performance storage layer for the most demanding data-streaming applications. 

Optimising sparse-matrix vector multiplication on Intel Xeon Gold processors

Author: Guest blogger
Posted: 24 Aug 2020 | 18:11

Edoardo Coronado visited EPCC from 18th May–15th August 2019 under the HPC-Europa3(link is external) programme. Here he gives us an update on his research since his previous blog article.

During my visit to EPCC in 2019 under the auspices of the HPC-Europa3 programme, I worked on optimising routines to perform the Sparse Matrix-Vector (SpMV) product, a common operation in lots of scientific applications, on Intel Xeon Gold processors and Graphics Processing Units (GPUs).

ePython: can we have our cake and eat it?

Author: Guest blogger
Posted: 17 Aug 2020 | 15:02

Maurice Jamieson is a third-year PhD student at the EPCC working on the programmability of micro-core architectures, both in terms of design and implementation, through the development of the ePython programming language.

We recently published a paper at the Scientific Computing with Python (SciPy) 2020 conference that outlined the latest developments to ePython, a version of Python specifically written to leverage micro-core architectures. The paper provides a detailed discussion of the design and implementation of ePython, touching on the latest updates to manage arbitrary large data sets and native code generation. This post will summarise key elements of the paper.

Pages

Blog Archive

Contact

Tracy Peet
+44 (0) 131 650 5362
t.peet@epcc.ed.ac.uk