Contact us Heritage collections Image license terms
HOME ACL ACD C&A INF SE ENG Alvey Transputers Literature
Further reading □ OverviewContents1. Summary2. Terms of reference and method of working3. Background4. Problems5. Board submissions and user's views6. Options and Comments7. Conclusions and RecommendationsA. Statistical data on SERC computingB. Submissions from Boards etcC. Case for supercomputer
CCD CISD Harwell Archives Contact us Heritage archives Image license terms

Search

   
InformaticsLiteratureReportsCRWP
InformaticsLiteratureReportsCRWP
ACL ACD C&A INF CCD CISD Archives
Further reading

Overview
Contents
1. Summary
2. Terms of reference and method of working
3. Background
4. Problems
5. Board submissions and user's views
6. Options and Comments
7. Conclusions and Recommendations
A. Statistical data on SERC computing
B. Submissions from Boards etc
C. Case for supercomputer

1. Summary

The problems of providing and organising SERC's central and distributed computing facilities for academic research are examined. Four options are considered for the future organisation of these computing facilities, and for their administration and funding, which could provide a solution to some of these problems.

The role of SERC in supporting computing for academic research in the UK in relation to the roles of other funding bodies is described in Section 3. The problems currently besetting the provision of computing resources are considered in Section 4. Among the problems discussed are: the unsatisfactory nature of the present administrative arrangements and funding mechanisms within SERC, and the blurred responsibility within SERC between central support and direct provision for certain types of computing provision. The wider problems are also discussed of divided responsibility for computing provision and for long-term strategic planning between the various funding agencies involved in research computing, of the effects of rapid technological development, of coordination of hardware and software provision and support, and of user support. Section 5 summarises the evidence which has been taken by the Working Party from Boards, user support groups, and other bodies.

The four options which have been examined in seeking a robust solution to SERC's computing problems are discussed in Section 6. Options I and II are discussed as possible solutions to the short-term internal problems of SERC. Option I involves the devolution of all responsibility for computing to Boards with no central provision. Option II involves the reorganisation of SERC computing so that facility-related computing and some other aspects of computing such as ICF are devolved to Boards and central provision is retained for infrastructure and networking. Options III and IV are considered as possible solutions to the problem of future provision of large state-of-the-art computers, either funded solely by SERC (option III) or jointly by all the Research Councils (option IV).

In Section 7 the recommendations of the Working Party are given in detail. Option II is put forward as the preferred solution to SERC's short-term problems, and specific recommendations are made as to the future organisation of computing resources. Option IV proposes collaboration between SERC and the other research councils in the provision of a national resource for research computing, including large state-of-the-art computers.

The specific recommendations made by the Working Party are as follows:

  1. Option II - "Central provision coupled to Boards' requirements" should be adopted (7.1). Within this option:
    1. The AS 7000 at DL should become the responsibility of the Science Board
    2. The RAL mainframes should be run on a direct-charging basis
    3. The ICF should become the responsibility of the Engineering Board with support for non-ICF and VAX machines devolved as far as possible to Boards
    4. The single-user systems programme should become the responsibility of the Engineering Board with contributions from other Boards directly linked to the systems installed in their area
    5. Support for infrastructure and networking should be a direct Council responsibility.
  2. A full-time Director of SERC Computing should be appointed (7.2).
  3. The Central Computing Committee should be retained with revised terms of reference.
  4. SERC should not pay for vector processing charges at National Centres.
  5. The new administrative arrangements relating to Option II should be brought in by April 1985.
  6. A joint Computer Board/Research Councils Working Party should be set up to give urgent consideration to the future provision of advanced research computing (7.3).
⇑ Top of page
© Chilton Computing and UKRI Science and Technology Facilities Council webmaster@chilton-computing.org.uk
Our thanks to UKRI Science and Technology Facilities Council for hosting this site