Logo of Science Foundation Ireland  Logo of the Higher Education Authority, Ireland7 CapacitiesGPGPU Research Projects
Ireland's High-Performance Computing Centre | ICHEC
Home | News | Infrastructure | Outreach | Services | Research | Support | Education & Training | Consultancy | About Us | Login

Science Council

The Science Council consists of members from various scientific disciplines whose main role is to assess the scientific merit of applications for access to ICHEC's computing services. See below for the full list of members, as well as detailed information about the Science Council's roles and review guidelines. ICHEC's Science Council is a voluntary, pro-bono council.


Membership

The membership of the Science Council is as follows:

Physical Sciences

  • Dr. Jim Greer [Chair] (2009 - current) - Tyndall National Institute
  • Prof. Jiri Vala (2009 - current) - National University of Ireland, Maynooth
  • Dr. Damien Thompson (2011 - current) - University of Limerick
  • Prof. Michael Pomeroy (2012 - current) - University of Limerick
  • Dr. Niall English (2013 - current) - University College Dublin
  • Prof. Mike Peardon (2012 - current) - Trinity College Dublin

Computer Sciences

  • Prof. Mike Scott (2010 - current) - Dublin City University
  • Dr. John Burns (2012 - current) - Institute of Technology, Tallaght

Life Sciences

  • Dr. Isabel Rozas (2012 - current) - Trinity College Dublin
  • Prof. Cathal Seoighe (2012 - current) - National University of Ireland, Galway

Environmental Sciences

  • Prof. Michael Hartnett (2011 - current) - National University of Ireland, Galway
  • Prof. Chris Bean (2009 - current) - University College Dublin
  • Ray McGrath (2009 - current) - Met Éireann

Engineering Sciences

  • Dr. Nathan Quinlan (2011 - current) - National University of Ireland, Galway
  • Prof. Noel O'Dowd (2009 - current) - University of Limerick

External Assessor


Terms of Reference

  1. The ICHEC Science Council is responsible for assessing the scientific merit of applications from Irish researchers for access to ICHEC's national high-end computing services and is solely responsible for decisions on acceptance or rejection of particular applications.
  2. In reaching its conclusions, the Science Council shall employ a peer review process appropriate for the Class of service requested and as set put in Annex 1 (not included on web page).
  3. The main criterion for access to ICHEC services is the quality of the research it facilitates, where quality is determined in the main by the anticipated scientific impact of the projected research outcomes.
  4. The membership of the Science Council shall be drawn from all of the main disciplinary areas that require high-end computing viz.:
    • Physical and Chemical sciences
    • Life sciences
    • Environmental sciences, and
    • Engineering
    In general, each of these areas shall be represented by at least two members, except where the number of users in a particular area greatly exceeds the number in other areas. In the case of the latter, three or more representatives may be appointed as appropriate. Membership of Council is personal and not institutionally representative. The number of members of the Council will be not less than 9 and not more than 18.
  5. One third of the Membership of the Council shall be replaced at the beginning of each calendar year. By the first Friday of November each year, in consultation with members of the existing Science Council, the Director of ICHEC will bring forward a list of nominations, to replace retiring members of Council at the start of the next calendar year, and for a Chairperson of the Council for the next calendar year. These nominations will be considered for approval of the ICHEC Advisory Board by the second Friday of December. Members shall normally serve for three years. The Chairperson may be re-appointed each year for up to three years.
  6. The Science Council shall meet to consider applications according to the schedule laid out in Annex 1 (not included on web page). Where appropriate, and as detailed in Annex 1 (not included on web page), e-meetings may be convened to consider applications.
  7. A member of the ICHEC Executive (normally the Head of User Support) shall attend all meetings of Council, but shall not have a vote. H/she shall keep the ICHEC Executive informed of Council's deliberations Secretarial support will be provided for the Council.
  8. DECI Project Evaluation: the Terms of Reference and operation of the Science Council satisfy the requirements of PRACE for its tier-1 DECI access programme. On foot of this PRACE has requested that following technical evaluation by ICHEC staff the Council would perform the scientific evaluation of Irish DECI applications. PRACE provides guidelines for this process. The main objectives are to rank the applications in order of scientific merit and provide recommendations for the fraction of requested compute time to award. The guidance of the Council is then taken into close consideration when compute time on various European tier-1 systems is allocated. This process is undertaken twice each year.
  9. These Terms of Reference may be varied from time to time by the Advisory Board.

Guidelines for review and reviewers

The review process will independently evaluate two aspects of the proposal:

Technical Review

Technical aspects will be considered through a technical evaluation which will assess the feasibility of the proposed work: e.g., availability of resources such as software packages and libraries, adequacy of the proposed approach for the target architecture, characteristics of typical jobs (maximum run-time, number of CPUs for parallel jobs, memory requirements, etc.) to assist ICHEC staff in determining optimal scheduling policies. Possible conflicts with the ICHEC's scheduling policies will also be highlighted in this document (e.g., users wishing to run 4 core jobs for 3 or 4 weeks). Technical evaluations will be undertaken by a support scientist selected from ICHEC staff.

Applications that fail the Technical Review will be discussed with the researcher to see if the application can be successfully modified. If not, the application will be rejected.

Scientific Review

The scientific value of the proposed work will be undertaken through a peer review process managed by ICHEC's Science Council, except where computational resources are sought as part of a project which has already been peer-reviewed by a third-party science funding agency. In the latter case, the Science Council will normally accept the agency's scientific assessment and the proposal will be assessed only on its technical merits. In all other cases, the Science Council alone will decide whether to accept or reject an application, as well as make recommendations to ICHEC's Manager of Operations for the level of resources to be allocated to successful applications.

Applications for each classification are subject to different levels of review in accordance with the processing power requested for the application. Reviewers are selected from among the members of the Science Council by the Chairperson according to disciplinary experience, except for Class A as noted below. In detail:

  • Class A applications will be reviewed by two reviewers, at least one of whom must be based outside of the Republic of Ireland.
  • Class B applications will be reviewed by two reviewers.
  • Class C are exempt from Scientific peer review.

Scientific evaluations will be carried out with a different level of scrutiny depending on the class of applications.

  • Applications for Class C access will have minimum scientific requirements, for which a half-page overview of the research undertaken will be adequate.
  • The scientific case for Class B applications will require a more detailed description of the research, and must satisfy the reviewers that the work described is of sufficient quality to be published in a refereed publication.
  • Finally, Class A applications must include a strong and detailed case for funding clearly establishing the High Impact nature of the problem under investigation. Such applications are expected to lead to high impact publications, and as such must be supported by a strong scientific case.

Class A and B applications will be independently reviewed by two reviewers, at least one of whom must based outside of the Republic of Ireland in the case of Class A applications.

Reviewers will assess the proposal according to the structure specified to applicants (below) and comment on the quality, strengths or shortcomings of each aspect of the proposal. An overall mark should be assigned on the following scale:

9-10 outstanding in all respects and likely to lead to very high-impact publications
7-8 excellent project, likely to lead to publications in leading journals
6 very good, well-balanced application, likely to lead to international peer-reviewed publications
5 good - a reasonable application which may lead to publication and/or conference proceedings
4 application requires substantial further development
2-3 very poorly written or with little scientific merit; should not be funded

The procedure for assessing applications is as follows:

The Chairman allocates each reviewer c. 3 proposals and the reviewer then agrees:

  1. to review the proposal and to acknowledge that it is in their field of competence
  2. to confirm that they do not have a conflict of interest
  3. not to discuss the application with a third party
  4. to deliver the review on time

Reviewers have access to PDF files of the applications and a history of the PI’s use of ICHEC resources, are provided at initial request for review. The review process involves providng detailed comments on the overall merit of the application under the following headings:

  1. Scientific Merit
  2. Reasonable demand of resources
  3. PI/ applicant track record
  4. Presentation of project plan

The review process should produce a mark out of 10. Successful Class A applications should score of at least 8/10 for both reviews to get funded, according to the international norms (INTAS) that ICHEC subscribes to, and successful Class B applications should score at least 6/10 for both reviews.

All deliberations of the Science Council are strictly confidential in order to ensure that applicants have the confidence to present a comprehensive description of their project.