ARCHER News

Tuesday 22nd May 2018

PRACE Preparatory Access Call

Preparatory access is intended for testing and developing codes in order to prepare applications for PRACE Tier-0 project access. In particular, we would like to draw your attention to the Preparatory Access Type C and Type D which includes code development with support from PRACE experts. Type C provides direct access to a Tier-0 system. Type D offers users to start the optimisation work on a PRACE Tier-1 national system to finally reach PRACE Tier-0 system scalability.

Further information is given on the PRACE web page: prace-ri.eu/PRACE-Preparatory-Access.

The Preparatory Access Call is a rolling call, researchers can apply for access all year. There are no closing dates, proposals for Type C and Type D will be reviewed every third month.

The next cut-off date is June 1st, 2018.

We invite all interested scientists, whose projects fit the criteria of the PRACE call to submit their projects. The criteria for the PRACE Calls as well as complete information on the PRACE machines, how to apply and on the PRACE peer review process can be found on the website mentioned above.

Deep Learning and Multi-GPU Training Course, Hartree Centre

Hartree Centre and Atos are inviting you to the Deep Learning and Multi-GPU Training on Tuesday, June 5, 2018 taught by Dr Adam Grzywaczewski, Deep Learning Solution Architect, NVIDIA Ltd. The goal of the course is to provide an overview of the approaches to training neural networks using multiple GPUs.

In this hands-on course you will learn:

  • Key approaches to multi GPU training
  • Key algorithmic and engineering challenges to large scale training
  • Key techniques used to overcome the above-mentioned challenges

On completion, you will be able to effectively parallelize the training of deep neural networks.

> > The course is designed with 2 main groups in mind:
  1. AI practitioners, so people that on a day to day basis develop neural networks and are looking to learn how to scale their code in order to accelerate their work. In particular the audience needs to understand the fundamental principles of training neural networks so understand concepts such as: cost function, backpropagation, Stochastic Gradient Descent and their mathematical foundations.
  2. The support staff responsible for the system, so that they can understand the infrastructure requirements of the AI training process.

To register and for more information go to www.cvent.com/d/wgqqhr.

Corrected module: GROMACS 5.1.4

GROMACS 5.1.4 was partially overwritten with GROMACS 5.0.2 on 27 March 2018. This affects the gmx program (and its aliases) and some topology files:

  • share/gromacs/top/amber03.ff/aminoacids.rtp
  • share/gromacs/top/amber99sb-ildn.ff/aminoacids.rtp
  • share/gromacs/top/charmm27.ff/aminoacids.hdb
  • share/gromacs/top/charmm27.ff/ffnabonded.itp
  • share/gromacs/top/charmm27.ff/ffnonbonded.itp
  • share/gromacs/top/oplsaa.ff/aminoacids.hdb
  • share/gromacs/top/oplsaa.ff/aminoacids.rtp
  • share/gromacs/top/oplsaa.ff/atomtypes.atp
  • share/gromacs/top/oplsaa.ff/ffbonded.itp
  • share/gromacs/top/oplsaa.ff/ffnonbonded.itp
  • share/gromacs/top/oplsaa.ff/ions.itp
  • share/gromacs/top/oplsaa.ff/watermodels.dat
  • share/gromacs/top/residuetypes.dat

The correct Gromacs 5.1.4 will be re-installed between 1500 and 1600 BST 13 June 2018. In the meantime, you can access the correct Gromacs 5.1.4 with:

module load gromacs/5.1.4_original

Update to ARCHER policies

Access policies for ARCHER and the SAFE have recently been updated. You can find the policies at:

www.archer.ac.uk/about-archer/policies/

Webinar: Modern C++ for Computational Scientists

Wednesday 13th June 2018 14:00 BST & 27th June 2018 14:00 BST - Note earlier-than-usual start time

Rupert Nash, EPCC, will present a four-part tutorial, over two afternoons, on the use of Modern C++ for Computational Scientists.

Since the 2011 revision to the C++ language and standard library, the ways it is now being used are quite different. Used well, these features enable the programmer to write elegant, reusable and portable code that runs efficiently on a variety of architectures.

However, it is still a very large and complex tool. This set of online lectures, delivered over two Wednesday afternoons and including practical exercises, will cover a minimal set of features to allow an experienced non-C++ programmer to get to grips with language. These include: overloading, templates, containers, iterators, lambdas and standard algorithms. It concludes with a brief discussion of modern frameworks for portable parallel performance which are commonly implemented in C++.

The course would appeal to existing C++ programmers wanting to learn techniques that are applicable to numerical computing, or C programmers who want to know what parts of the C++ standard they should prioritise when learning new features.

Details and join link : www.archer.ac.uk/training/virtual

Performance Analysis Workshop

19-21 June 2018, UCL

This workshop is organised by VI-HPS for the UK PRACE Advanced Training Centre to:

  • give an overview of the VI-HPS programming tools suite
  • explain the functionality of individual tools, and how to use them effectively
  • offer hands-on experience and expert assistance using the tools

All attendees will be given access to ARCHER, however, all tools covered are designed for portability so will be applicable to any HPC system. On completion participants should be familiar with common performance analysis and diagnosis techniques and how they can be employed in practice (on a range of HPC systems). Those who prepared their own application test cases will have been coached in the tuning of their measurement and analysis, and provided optimization suggestions.

Programme Overview

Presentations and hands-on sessions are on the following topics:

  • BSC tools for trace analysis and performance prediction
  • Score-P instrumentation and measurement
  • Scalasca automated trace analysis
  • MAQAO performance analysis and optimization
  • Full details and registration at www.vi-hps.org/training/tws/tw28.html.

    Upcoming Training Opportunities

    Registration open now

    • Software Carpentry London School of Economics 18-19 June 2018
    • Data Analytics with HPC Belfast 20-21 June 2018
    • Advanced OpenMP Cambridge 17 - 19 July 2018
    • Advanced MPI Milton Keynes 30 - 31 July 2018

    ARCHER Summer School

    • Hands-on Introduction to HPC Edinburgh 2-3 July 2018
    • Message-passing Programming with MPI Edinburgh 4-6 July 2018

    Full details and registration at http://www.archer.ac.uk/training/index.php

    EPCC Training Privacy Statement is now available at http://www.archer.ac.uk/training/privacy.php