N8 HPC are Hosting a Free Azure Workshop - University of Manchester

posted by Michael K Griffiths

On Wednesday 25 March there will be a free one-day workshop in Azure taking place at the University of Manchester.

Delivered by specialists from Microsoft’s Education Team the session is intended for research infrastructure engineers, systems administrators, cloud architects as well as data & software engineers, data scientists and AI engineers.

The day will feature practical tasks in Azure covering areas such as:

•       High Performance Computing
•       High Throughput Computing
•       Machine Learning
•       Cognitive Service and Deep Learning

You can read full details of the sessions and register for a place here: https://www.eventbrite.co.uk/e/cloud-computing-with-microsoft-azure-tickets-95205593493

The workshop will be free to attend and only open to those working and studying at one of the N8 Research Partnership universities.

Places are strictly limited and will be available on a first-come-first-served basis.

RSE Seminar 25th Feb 2020: from laptop to HPC to cloud: how to containerise your workflow to make it portable and reproducible

posted by Michael K Griffiths

This month in the RSE seminar series, Martin Callaghan, Research
Computing Manager at the University of Leeds, is coming to talk 
about how research software/workflows can be made portable and
reproducible by wrapping them in containers, plus tools for making the
process of containerisation easier.  Martin's full of great ideas and
is an engaging speaker so this should be great!

When: Tues 25th February 2020 - 12:00-13:00
Where: Lewin Room (COM-G12), Computer Science
Refreshments: coffee and cake

Abstract:
In this talk Martin will discuss a number of use-cases for
containerising your research codes and applications using both the
Docker and Singularity ecosystems.

Martin will demonstrate the NVIDIA HPC Container Maker language (write
your container definitions once to create both Docker and Singularity
definitions) and how to combine Git, Github, Dockerhub and Singularity
Hub to create a continuous test and development pipeline for your
containers which you can then go on to use on your Desktop, in the
Cloud or on HPC.

Bio:
Martin's Research Computing Manager and leads the Research Computing
team at the University of Leeds.

His team provides Research Infrastructures (including Cloud and High
Performance Computing), Programming and Software Development
consultancy across the University’s diverse research community.

His role is part Research Software Engineer, part trainer, part
consultant and part manager, including running Research Computing’s
comprehensive training programme for research staff and students.

His personal research interests are in text analytics, particularly
using neural networks to summarise text at scale.

HPC for N8 Universities Northern Intensive Computing Environment

posted 24 Feb 2020, 08:37 by Michael K Griffiths   [ updated 24 Feb 2020, 08:40 ]

Researchers are set to benefit from access to a High Performance Computing Facility (HPC), following a £3.1 million investment from the Engineering and Physical Sciences Research Council (EPSRC).


Based at the University of Durham, the new Northern Intensive Computing Environment (NICE) will provide a shared facility for academic and industry researchers from all of the N8 universities, including here at Sheffield. The other N8 universities are Durham, Newcastle, Lancaster, Leeds, Liverpool, Manchester and York.

The Tier-2 supercomputing services will provide researchers with invaluable access to powerful systems to support groundbreaking work in areas ranging from Artificial Intelligence (AI), energy storage and supply and therapeutic drug design.

The primary focus of NICE is to support experimental researchers who generate large datasets for analysis, such as researchers using Cryo-Electron Microscopy facilities within Molecular Biology and Biotechnology. The system will also be well suited for use by researchers across all disciplines who perform modelling using machine learning.

The unique GPU architecture of NICE will allow researchers to tackle much larger problem sizes than can be tackled by existing GPU systems, allowing larger data sets to be analysed and the simulation of more complex systems.

The new NICE service will use the same technology as that used in the current leading supercomputers in the world, extending the capability of accelerated computing. The technology has been chosen with the aim of combining experimental, modelling and machine learning approaches, bringing these communities together to address new challenges.

This will mean that machine learning can be used alongside modelling and simulation to better understand the vast data sets now being generated by experimentalists through, for example, national facilities such as Diamond and the Sir Henry Royce Institute and international facilities such as the European Synchrotron Radiation Facility. This approach will enable scientists to, for example, advance the imaging techniques necessary to produce the next generation of X-ray instruments and develop future students working with deep learning techniques at the interface of algorithms and High Performance Computing.

Our University was also part of another successful HPC award JADE2. JADE2 is a UK Tier-2 resource, funded by EPSRC, led by the University of Oxford and hosted at the Hartree Centre. JADE2 aims to build on the success of the current JADE facility - a unique national resource providing state of the art GPU computing facilities to world leading experts in the areas of Artificial Intelligence/Machine Learning (AI/ML) and molecular dynamics (MD) research - by providing increased computational capabilities to these communities and delivering a stronger, more robust service to address the lessons learned from the initial service.

High Performance Computing (HPC) facilities provided by IT Services are freely available to all staff and research students at the University.

As well as fast number crunching capabilities, it also provides many specialist software packages that are not normally available on the public workstations.


Hear more about NICE at the ResearchComputing@Sheffield 2020 annual event

RC@Sheffield 2020: Innovations in Data Analytics & High Performance Computing (HPC)

Wednesday 22 April 2020, 9am - 4pm
Firth Hall, Firth Court

This event presents aspects of research computing within the University from standard desktop to high performance computing applications, using open source and commercial code, for statistical analyses, bioinformatics, engineering simulations, medical applications and deep learning.

The day will consist of talks, including an update from the N8 CIR (N8 Centre of Excellence in Computationally Intensive Research), computation workflow walk-throughs and an opportunity to take part in two lightning talk sessions (register for your two minute research computing plug on the day).

All staff and students welcome.

For further information, contact Des Ryan at des.ryan@sheffield.ac.uk


Research Computing at Sheffield 2020 Save the Date 22nd April 2020

posted 24 Feb 2020, 07:56 by Michael K Griffiths

Save the date for the Research Computing full day event (09:00 - 16:00) on Wednesday 22 April at Firth Hall, TUoS - RC@Sheffield 2020: Innovations in Data Analytics & High Performance Computing (HPC)

R - Users Group Tuesday, February 4th, 2020: 1730

posted 28 Jan 2020, 04:07 by Michael K Griffiths

R-Shiny for Decision Making and Debugging in RStudio

Tuesday, February 4, 2020, 5.30pm
The Red Deer, 18 Pitt Street, Sheffield

R-Shiny for Decision Making

A short presentation on the use of R-Shiny to aid decision makers. Two examples are open source; including a study of the social and geographic disparities in the access to and the participation in parkrun events in England. The project aimed to support parkrun’s planned expansion, by providing recommendations for 200 new event locations, maximising access for deprived communities.

http://iol-map.shef.ac.uk/

Robert Smith is a PhD Student in Public Health Economics & Decision Science, ScHARR.

Getting to know the debugger in RStudio

If you're writing moderately complicated code in R there are times when your code throws an error but you don't know why: the error occurs inside a function or loop, the error message is obtuse or unhelpful, one of the many parameter sets you're processing is causing a problem... In this talk I'll introduce the debugger in RStudio, which helps you to get inside your code and diagnose problems from within.

Tamora James is a PhD researcher in Animal and Plant Sciences with a former life as a web and software developer. She is a co-organiser of the Sheffield R Users Group and a contributor to the British Ecological Society publication ‘A Guide to Reproducible Code in Ecology and Evolution’.

Further details and registration:

Sheffield R Users Group
@Sheffield_R_

Love Data Week 2020

posted 16 Jan 2020, 01:23 by Michael K Griffiths

We are putting together a couple of events for Love Data Week (10-14th February) here are some of the events which are taking place at The University of Sheffield

10am-12 11th February - DDP workshop on Looking after your Research Data - FULLY BOOKED - spaces available for subsequent sessions - http://ris.dept.shef.ac.uk/skills_seminars/listings/view/122
12-2pm 12th February - Data management plan drop-in session - https://lms.shef.ac.uk/index.php#STS
12-2pm 13th February - Data Conversations: Delving into data visualisations - https://www.eventbrite.co.uk/e/data-conversations-delving-into-data-visualisations-tickets-89967480155
10am-12 14th February - Bring Your Own Data workshop - https://www.eventbrite.co.uk/e/bring-your-own-data-workshop-tickets-89957440125


N8 CIR 2020 ReproHack Northern Tour

posted 14 Jan 2020, 00:00 by Michael K Griffiths

The N8 CIR, in collaboration with Anna Krystalli, a Research Software Engineer at the University of Sheffield, are organising a series of ReproHacks,  one-day reproducibility hackathons, across N8 universities.  These events bring together researchers, RSEs and data scientists to reproduce papers from published scientific code and data in a low pressure environment. They also provide an opportunity to network and learn from colleagues with different research and software experience. 

The 2020 N8 ‘Northern Tour’ events are scheduled to take place between January and March at:
 
·         Newcastle University – Tuesday 21 January
·         University of Leeds – Friday 14 February
·         University of Liverpool – Tuesday 25 February
·         University of Sheffield  – Tuesday 10 March 

·         University of Manchester – Thursday 12 March

 
Registration for these events, with the exception of Sheffield, is open now. You can read more about these events and what to expect, as well as book a place here: https://n8cir.org.uk/news/reprohacks/
 
ReproHacks only work when those participating have research code and data to experiment with. The more papers that are submitted, the more useful the events are so if you have . You can submit papers for consideration here: https://sheffield-university.shinyapps.io/n8cir-reprohacks/
 
Don’t worry if there isn’t one happening at your institution, these five events are open to anyone working or studying at one of the N8 institutions. If your institution isn’t hosting a ReproHack then travel bursaries are available to PGRs and university staff. They will be offered on a first come, first served basis. Please e-mail enquiries@n8cir.org.uk for details of how to apply.

Course in Parallel Computing with Graphical Processing Units

posted 13 Jan 2020, 23:51 by Michael K Griffiths

The following 12 week course in "Parallel Computing with Graphical Processing Units" is available to PhD students and research staff (including academic staff).

Course Details: https://paulrichmond.shef.ac.uk/teaching/COM4521/

The module is traditionally taught as a 4th year undergraduate and masters course but is typically has a high update of research students and staff who are wanting to learn about shared memory parallel and GPU computing.

The course goes beyond a simple 1 or 2 day introduction and provides advanced material on GPU programming and optimisation. There is a 2 hour lecture each week followed by a 2 hour lab class which can be conducted remotely or in person (subject to lab space availability). There is also a course google group monitored by lab demonstrators and members of the RSE team.

PhD students should enroll as part of their DDP via the following form: https://www.sheffield.ac.uk/rs/ddpportal/enquiry

Staff and research students should contact the course leader Paul Richmond (p.richmond@sheffield.ac.uk) to notify of their enrolment.

Note: PhD students are not required to undertake the assessment but if enrolled are expected to participate in the course.

Announcements for short introduction courses (1 day) on CUDA will be made via the RSE mailing list.

[N8CIR] Introduction to COMSOL Multiphysics

posted 29 Nov 2019, 05:34 by Michael K Griffiths

On 12 December the University of York is running an introduction to COMSOL Multiphysics. At the workshop, you will see the capabilities and workflow of the COMSOL Multiphysics® software and learn through a guided hands-on exercise. You will leave with new skills to work on your own applications using your free two-week COMSOL Multiphysics trial.

Full details of the event, as well as registration details can be found on the N8 CIR website here: https://n8cir.org.uk/events/introduction-comsol-multiphysics/

The N8CIR list is archived at:
http://listserv.manchester.ac.uk/cgi-bin/wa?A0=N8CIR

High Performance Computing Carpentry Workshop for Women in HPC

posted 13 Nov 2019, 04:29 by Desmond M Ryan

N8 CIR is delighted to support a new HPC Carpentry Workshop initiative

Taking place on Thursday and Friday 30 & 31 January, the two day
course has been organised by Mozhgan Kabiri Chimeh of the University
of Sheffield. It is being run in collaboration with Women in HPC and
the Software Sustainability Institute with the goal of supporting and
promoting women to develop the skills needed to use High Performance
Computing (HPC) in their research.

The course introduces the basics of HPC and how it can benefit and
accelerate research. After this introduction the course will focus on
practical activities including remotely connecting to HPC systems and
transferring data to them, using the Bash command line, using a
scheduler and working effectively on a shared system and using
software modules to access different HPC software.

Participants will need to bring a laptop computer with them to the
course. Full details of the software requirements can be found via
Eventbrite page.

Places on the course are FREE and lunch and refreshments are included.
The course is primarily aimed at women and will be taught and
supported by an all-woman team:

·         Mozhgan Kabiri Chimeh (University of Sheffield)

·         Anna Brown (University of Oxford, University of Southampton)

·         Fouzhan Hosseini (Numerical Algorithms Group)

·         Weronika Fillinger (EPCC)

·         Neelofar Banglawala (EPCC)

Until Friday 22 November registration is open exclusively to those
working and studying at universities within the N8 Research
Partnership. Full details of the course, including software and
hardware requirements, can be found on Eventbrite here:
https://www.eventbrite.co.uk/e/hpc-carpentry-workshop-2-days-tickets-80825839261

The N8CIR list is archived at:


1-10 of 141