This heading covers those SERC scientific activities that were not devolved to direct control by Boards when the arrangements for managing computing were revised at the beginning of 1985-86. At the time they included two main items: the SERC computing infrastructure and networking facilities that are used by all Boards, and the Rutherford Appleton Laboratory mainframe computing service which is funded by charging Boards and others for usage. There have been no major changes in the nature and scale of these activities in the past year. In this second year of charging for use of the mainframes, the income has been sufficient to cover all the recurrent operating costs but, as in the first year, not to cover all the funding required to offset depreciation on installed equipment and to provide for planned developments.
This constitutes a new item under the Central Computing heading and the most significant event has been the delivery and setting to work of a Cray X-MP/48 computer which SERC will operate on behalf of the wider UK academic community. The provision of this machine follows the recommendations of a report prepared in 1985 for the Advisory Board for the Research Councils, the University Grants Committee and the Computer Board on the Future provision for advanced research computing.
The report found that there were major opportunities for advancing knowledge by computational methods in almost every branch of science. The power of modern supercomputers makes possible realistic simulations of complex physical phenomena to an accuracy that was unattainable on previous generations of computers. Such simulations can be used to check the validity of theoretical models or to conduct simulated experiments on physical systems for which laboratory experimentation is impossible. The fields to which such computations can be applied are as diverse as galactic evolution, chemical reactions of molecules, the circulation of the oceans and the atmosphere, and the design of pharmaceutical products.
The Cray X-MP/48 is one of the most powerful computers it is possible to obtain. In principle it is capable of performing about a thousand million arithmetic operations per second. In practice one is unlikely to attain this extraordinary level of performance because one would have to make every individual section of the computer simultaneously perform at full speed, and although programs can be written to do this they tend to have little relationship to real life applications. Nevertheless, by careful tuning and optimisation of existing scientific programs it is possible to attain a substantial fraction of this theoretical performance for real problems and this is considerably more than can be obtained from conventional mainframe computers.
The computer achieves its high performance in three main ways. First it is constructed from very fast components and all possible steps have been taken in its design to reduce the transit time for electrical signals to move from one point to another. The computer is therefore very compact - it occupies only 6 square metres of floor space. Its compactness leads to formidable cooling problems and the machine is cooled by freon refrigerant rather than by the air or water used by more conventional computers.
Secondly, the machine contains special hardware designed to handle vectors (strings of numbers) in much the same way that a conventional computer handles individual numbers. This can provide large gains in performance when dealing with repetitive operations on arrays of numbers, as for example in doing the kind of matrix algebra which features prominently in solving sets of equations.
Thirdly, the machine contains many functional units which can operate simultaneously, and clearly the more one can exploit this parallelism the greater the overall performance that can be delivered.
The above features apply to any Cray computer. The X-MP/48 model contains a further element of large-scale parallelism in that it consists effectively of four complete Cray computers (processors) in one box. It is possible to run the four processors almost as four independent machines or, by adding extra instructions to users' programs, one can make all four processors cooperate on a single job.
The X-MP/48 computer has 8 million 64 bit words of memory and a 32 million-word solid state device (SSD) to be used for data files which are frequently required by user programs. Use of the SSD can make it feasible to perform certain types of calculations which might otherwise be impractical if one had access only to traditional, and much slower, disk drives for such data storage.
The Cray also has its own conventional disk drive units and it will have access to other facilities which are now available on the IBM mainframe computing facility at Rutherford Appleton Laboratory. This IBM system will act as a 'front-end' to the Cray. It will provide the route into the Cray from the Joint Academic Network (JANET), it will be the machine on which much of the preparatory work is done by users before they submit their jobs to the Cray, and it will provide large-scale storage facilities for long term data storage. In time, further front-end machines providing somewhat different facilities may be added as funds become available.
The Cray computer is housed at the Atlas Centre of the Rutherford Appleton Laboratory. The computer was delivered to the Centre in December 1986 and, after a period of installation, commissioning and testing, a computing service was offered to users from the beginning of February 1987. By the end of March 1987 about 20 groups had been authorised via peer review to use the machine and a further 80 groups had submitted applications. The take-up of work on the machine has been rapid and a number of substantial scientific projects are under way.
The provision of this computer is one element of a national strategy for advanced research computing. The other elements are the enhancement of JANET, and the provision of more specialised computers, such as array processors, to complement the supercomputer. The overall programme of work on advanced research computing will be coordinated by a new Joint Policy Committee on which there will be representatives of the Advisory Board for the Research Councils, the Computer Board and the University Grants Committee.
The enhancements to JANET are already in progress and will substantially increase the network's ability to transfer large volumes of data quickly. On specialised computers, a Floating Point Systems T-20 parallel architecture machine has been installed at the Daresbury Laboratory. This will be used to investigate the feasibility of performing large-scale scientific computations on an array of relatively inexpensive processors. The Science Board plans to spend about £900,000 in 1987-88 on machines of this type against submissions from university departments, and there may be related activities under the Engineering Board's Transputer Initiative.
