Developing an HPC/NA Roadmap

Developing an HPC/NA Roadmap

Many applications from different fields share a common numerical algorithmic base. We aim to capture the elements of this common base, identify the status of those elements and in conjunction with the EPSRC Technology and Applications roadmapping activity, determine areas in which the UK should invest in algorithm development. 

A significant sample of applications, from a range of research areas, will be included in the roadmapping activity. The applications chosen will include those in the EPSRC Technology and Applications roadmap, and others that represent upcoming and potentially new HPC areas.

Latest News

First Stage Final Report Published

Roadmap V1.0 Published

  • The initial version of the roadmap is available to download. The document was discussed at the 3rd workshop in London and a revised version will be published shortly. Feedback and comment on the roadmap is welcomed.

3rd Applications/Algorithms Roadmapping Workshop

  • Royal Society, Carlton Terrace, London, 26th - 27th January 2009
  • Report of the 3rd workshop has been published
  • Speaker presentations are now available

Previous Workshop Reports

This first workshop took place in Oxford over 5th-7th November attracting over 40 delegates from academia and industry. The initial roadmap was defined at the second workshop in December in Manchester followed by a further open workshop on 26th/27th January in London. Documents available to download are:

Activity Partners

Prof. A. E. Trefethen, Oxford e-Research Centre, University of Oxford
Prof P. V. Coveney, University College London
Prof N. J. Higham, University of Manchester
Prof I. S. Duff, STFC, Rutherford-Appleton Laboratory
Engineering & Physical Sciences Research Council

Contact

Email     info@oerc.ox.ac.uk

Background

The applications/algorithms roadmapping activity will develop the first instantiation of an high performance numerical algorithm roadmap. The roadmap will identify areas of research and development focus for the next five years including specific algorithmic areas required by applications as well as new architectural issues requiring consideration. The benefits of the roadmap will be:

  • A co-ordinated approach for a numerical algorithm and library development.
  • An internationally leading research and development agenda through the engagement of international experts, complementing and leveraging US and European efforts.
  • The initial roadmap will be constructed through a series of workshops bringing together stakeholders and enabling input from the broader community.

Aims

Many applications from different fields share a common numerical algorithmic base. We aim to capture the elements of this common base, identify the status of those elements and in conjunction with the EPSRC Technology and Applications roadmapping activity, determine areas in which the UK should invest in algorithm development. 

A significant sample of applications, from a range of research areas, will be included in the roadmapping activity. The applications chosen will include those in the EPSRC Technology and Applications roadmap, and others that represent upcoming and potentially new HPC areas.

The applications should provide the basis to understand:

  • The role and limits of a common algorithmic base
  • How this common algorithmic base is currently delivered and how should it be delivered in the future
  • What are the current requirements and limitations of the applications, and how these should be expanded
  • What are the “road-blocks” that limit the scope of the future exploitation of these applications.
  • A better comprehension of the “knowledge gap” between algorithmic developments and scientific deployment
  • How significant computing language as well as other “practical” issues weigh in the delivery of algorithmic content

Activities

Direct engagement with researchers from various areas needs to be actively sought, hopefully tying in with input from industry and commerce.

The initial HPC/NA Roadmap centered around three workshops:

  • Workshop 1: 5-7 November 2008, Oxford e-Research Centre, Oxford University.
    Detailed discussion of current and future numerical aspects and requirements in specific applications and fields
  • Workshop 2: 8-9 December 2008, Manchester University.
    Initial definition of a road map for future development.
  • Workshop 3: 26-27 January 2009, Royal Society, London.
    Iteration on Road map definitions and requirements

Engaging the research community is of course of paramount importance throughout the activity and afterwards.