Jump Over Left Menu
Engineering Computing Newsletter: Issue 40
- EASE Visualisation Community Club - Video Record
- RAL Parallel Processing Evaluation Centre Established
- The Connection Machine of Edinburgh: Access for UK Academics
- E&A Publications
- RAL Software Tools for STEP
- An Equational Reasoning System in Standard ML
- C For Fortran Programmers Course
- Introductory School in CFD
- ERCOFTAC/CFDCC Workshop on Visualisation of Numerically Generated Data
- Graphics and Visualisation: Techniques and Tools
- Workshop on Developing Parallel Engineering Applications
- Visualisation CC Meeting Reports
- Forthcoming Events
EASE Visualisation Community Club - Video Record
Video is an increasingly important medium in scientific visualisation, for both exploration and presentation of data. It is natural also to think of video as a medium for disseminating information about visualisation.
The Visualisation Community Club is therefore planning to use video as a key part of its activities:
- Record of Seminars and Meetings. These will be recorded on video, and edited down to a 30 minute programme - to be available later to attendees as proceedings, and indeed to others who may not have been able to attend
- Annual Video. This will be a short 10 minute tape, highlighting the work of the Club over the year, and showing short clips of effective use of video in engineering research
- Video Collection. This will be a growing archive of successful examples of the use of video in science and engineering
Call For Contributions
Our first project will be the creation of a 1992 video. For this we would like to include examples from the engineering community where video has been used as a key part of the visualisation - perhaps allowing a discovery to be made which would not have been possible otherwise. Please send us, in the first instance, PAL VHS copies of any material you would be willing to have considered for inclusion; if it is part of a long sequence, please indicate where the clip starts and ends. Please include information on the scientific or engineering background, so we know what we are watching!
We will use these copies for deciding the content of the video, and we will then contact you for further details, to get a higher quality original tape and to agree any copyright/acknowledgement We can handle (in order of preference) for this higher quality tape:
- Hi-Band U-Matic SP
- Betacam SP
- Lo-Band U-Matic
all in PAL format Tapes may have a soundtrack - please tell us if they do. A 0db test tone and colour bars at the beginning of the tape are useful for quality control.
Ken Brodlie, University of Leeds
RAL Parallel Processing Evaluation Centre Established
A Parallel Processing Evaluation Centre has been established at RAL under the EASE Programme to assist SERC-funded researchers who are using or would like to use parallel processing in their research. The service is aimed particularly at people who wish to assess the suitability of parallel processing for engineering applications and the choice of appropriate parallel hardware and software platforms.
The Centre is equipped with a range of T800 transputer and i860 kit together with associated parallel software development tools. Equipment based on the T9000 transputer will be added once this becomes available next year. Use of the equipment is available free on an appointment basis to the SERC-supported community, with staff on hand to assist users in getting started.
The staff of the Centre are also available, by appointment, to discuss users' requirements, to advise on the choice of approaches to parallelism, to recommend appropriate equipment, and to provide advice to current or prospective grant holders. It would be helpful when making appointments if users can provide a brief description of their interest.
Chris Wadsworth, RAL
The Connection Machine of Edinburgh: Access for UK Academics
The University of Edinburgh has received funding from the Advisory Board for the Research Councils to support a service for U.K. academics on a 16,384 processor Connection Machine. This machine is currently the most powerful parallel super computer in the U.K. with a peak performance of 8 Gflops.
The Connection Machine is a data parallel computer with 512 Mbyte of memory, a 10 Gbyte DataVault, and frame buffer. The machine is hosted by two Sun 4/370 workstations, permitting users to develop code under SunOS. The system is accessible on JANET and internet.
The Connection Machine can be programmed in data parallel Fortran (a subset of Fortran-90), C (ANSI C with parallel extensions) or Lisp. There is also a comprehensive library of scientific functions providing random number generators, matrix/vector operations, linear solvers, FFTs, etc. Porting many serial codes to the Connection Machine is a straightforward process.
Prism is an X-windows environment which permits users to profile and debug codes and to visualise program data graphically. Prism may be used by those accessing the Connection Machine directly or through the internet.
How do I get access to the Connection Machine? Until 1st December 1992, informal applications for pump priming time are being accepted.
Sun workstations are in widespread use throughout the academic community. The EASE Education & Awareness Group has therefore produced three titles to help meet the demand for information on their use.
The Systems Administrator's Guide (pp116) is a basic introduction to system administration for novice administrators or those curious to know how the system runs.
Introduction to SunOS (pp86) is primarily intended for the research engineering with little or no experience of SunOS.
New Users' Guide (pp71) provides not only guidance but describes the use of the SunView graphical interface.
RAL Software Tools for STEP
The STandard for Exchange of Product data (STEP) is a large ISO effort aimed at integrating CAE applications. Interfaces Group (part of RAL Informatics Department) has been involved in the development of STEP since June 1986, and is involved in a wide range of standards-based projects. During this time, a number of programs have been written which together form a useful toolkit for the developer of STEP-based applications and interfaces. The relationships between the tools are shown in the diagram below.
The tools are:
- STEP exchange file parser: This provides an easy-to-use interface to the information contained in a STEP exchange file. The parser performs syntactic and semantic checking of the exchange file, freeing the applications programmer from this burden. The application runs the parser as a subroutine, then queries the content of the file using a supplied set of subroutines
- EXPRESS compiler: This checks EXPRESS source code for syntactic errors and stores the schema information in a symbol table. A variety of outputs can be requested, including pretty-printed output and a Keyword Definition File, which is used by the exchange file parser when performing its semantic checks. Program hooks are provided to allow the addition of further output modules. The compiler accepts the latest version of EXPRESS
- EXPRESS database system: Currently under development, this database management system provides an interface to data consistent with the STEP Data Access Interface (SDAI) currently being developed within ISO (see The STEP Data Access Interface , ECN July). Using this, applications can read, write and query data as if they are talking to an EXPRESS schema. This means applications can be easily integrated in an open architecture. The program takes as input a data dictionary derived from an EXPRESS schema
- Dictionary File Editor: Dictionary Files and Keyword Definition Files have a strict format which makes them laborious to create by hand. This simple editor allows free format input from the user, and can convert between the two me formats. It can also regenerate EXPRESS source code, and output C struct typedefs, corresponding to EXPRESS entity definitions, in a header file suitable for inclusion in applications programs
The software is written in C and Fortran 77, and is designed to be easily portable between operating systems.
Source code is available, for research purposes only, free of charge.
Mike Mead, Informatics
An Equational Reasoning System in Standard ML
The formal design and verification of software is becoming widely recognised as an essential element of the software engineer's toolkit for the development of correct software. However, the adoption of such techniques has been hampered by the lack of software tools to aid the formal development of complex systems. Equational Reasoning is a formal technique which is particularly well suited for automated reasoning tools to assist the discharge of correctness proofs for software systems.
We all use equations everyday. For instance, we automatically discount adding 0 to a number as this does not give a different number. We can write this as the equation:
x + 0 = x
By changing such an equation into a Rewriting Rule so that it only goes in one direction:
x + 0 -> x
we can give the rule to a computer and allow it to automatically rewrite any expression by it. So by this we can do the rewriting:
(x + y) + 0 + 2 -> (x + y) + 2
Computer programs can be described by sets of equations defining their properties. By using rewriting, the designer of the program can explore the properties of a program and even produce a prototype implementation using the rewriting rules as a program themselves, before writing the final program. However, there are problems with this. Although we may use the above rule to delete the 0, we cannot use it to reintroduce the 0, which on occasion, is what we may need to do. This problem is known as Incompleteness. To overcome this, we can use algorithms which change rules into other rules so all the properties of equations can be worked out by the computer. Another problem is that a rewriting rule like
x + y -> y + x
can be used over and over again. Whenever an expression is rewritten by it, the result can also be rewritten. The computer, acting blindly, will never stop. This is the problem of Termination.
As part of the IEATP sponsored project on Equational Reasoning for LOTOS Verification, Glasgow University is developing an equational reasoning system, in conjunction with the Rutherford Appleton Laboratory. This system is inspired and influenced by the ERIL system previously developed at RAL, retaining and extending its important features but written in Standard ML, rather than the original Prolog.
The system has many features which allow the experimenter considerable freedom to reason with equations. The system:
- Allows users to declare their own object language
- Allows rewriting and unification of terms and equations
- Has several completion algorithms
- Has built in equational theories, in particular to handle the non-termination of the commutativity rule
- is built on a hierarchy of types, known as Order-Sorting. This allows greater expressively in defining partial functions and error handling and also efficiency of rewriting
- Allows the user to try different termination methods
The system currently has a menu driven teletype interface. Future plans with this system include: adding a tactic language to allow the user to set up and run provably correct equational reasoning tasks; the extension for more and more powerful termination methods; adding forms of inductive inference (structural and completion based) within an order-sorted framework; adding a theory store to allow the proof of theorems over large scale systems; building a window based user-interface, using Xwindows, to allow a greater ease of use of the system.
Plans for the use of the system include the testing of properties of LOTOS specifications, experiments in parallelisation and the verification of properties and refinement in algebraic specifications.
For more details, please contact the author, Brian Matthews, Department of Computing Science, University of Glasgow.
C For Fortran Programmers Course
This five day course, 18-23 January, is an intensive, hands-on programming course for Fortran programmers wishing to utilise the programming language C. The course has been run successfully several times before both in industry and academia, including within the EASE programme. The course covers the basic features of C, using simple examples from Fortran to ease the initial difficulties encountered with a change of language.
The course aims to impart the spirit of C and some basic software engineering principles (such as data hiding (encapsulation) so that, subsequent to the course, the programmer codes in C and not Fortran C.
Using SUN workstations attendees gain hands on experience of designing, coding and running their own C programs, designed to solve course exercises that re-enforce the course material. The final day's group exercise allows the delegates to gain a feel for those features of C that assist in the group development of software.
Although focusing on the Kernighan and Ritchie standard C (most compilers available accept this standard with some ANSI additions), the course also covers the key features of ANSI C.
The course comes complete with an indexed set of the overheads used during the course.
Due to the popularity of this free course it has been decided that each application to join the course shall be accompanied with a case for attendance (for example, you are an RA and are about to start a SERC funded project that needs the use of C).These cases will form the basis of the anticipated selection process.
Dr. S. K. Robinson, Informatics
Introductory School in CFD
Following the success of its Summer School in Computational Fluid Dynamics held in September1991 (reviewed in issue 35 of the ECN), the CFD Community Cub is organising its First Introductory School in CFD in January 1993.The school will provide a foundation in the physical understanding of fluid flow and the consequences for the numerical solution of fluid flow problems. It is intended to be suitable for recently-started graduate students or others starting in CFD research.
The topics to be covered in the school include the Navier-Stokes equations, the nature of turbulence, simplifications to the equations determined by the physics (e.g. high and low Reynolds number problems and thin shear layers), explicit and implicit time marching methods, turbulence models and their numerical implementation, pressure correction methods and grid generation. The formal sessions will comprise a mixture of review lectures on aspects of these topics in addition to lectures which present current research. This approach reflects the results of the membership survey which indicated that a combined educational and research-oriented school is the preferred option. Lecturers whose participation is confirmed are: Dr D Bray (RMCS), Prof D M Causon (Manchester Poly), S P Fiddes (Bristol), Dr K Knowles (RMCS), Prof J J McGuirk (Loughborough), Dr DG Rhodes (RMCS) and Dr B A Younis (City).
In addition to the formal lectures, there will be practical sessions where all participants will tackle a set of test problems which will be solved using some of the methods presented in the lectures. A network of SUN3s, SUN4s, a Stardent and a Silicon Graphics superworkstation, will form the basis of the computing facilities for the summer school; there will also be a link to the Cray at RAL available. Some commercial packages and the lecturers' own software will be mounted on the systems for use in the practical sessions. A brief introduction to computing under Unix will be provided for those not familiar with this operating system.
The Introductory School will take place at the Cosener's House, Abingdon 4-8 January 1993. The school will be residential and the price is £325 to academics and £650 to non-academics. Some bursaries will be available to cover the cost of attendance by graduate students.
Conor Fitzsimons, Informatics
ERCOFTAC/CFDCC Workshop on Visualisation of Numerically Generated Data
This workshop was held from 15-19 June, 1992, supported by the COMETT programme of the EC and was attended by approximately 50 people (50/50 industry/research). The general discussion at the end of the workshop culminated in the agreement to develop a special interest group in visualisation through ERCOFTAC.
The aim of the workshop was to provide a one week course of intensive instruction on the visualisation of numerically generated data with special reference to CFD. The workshop centred around the presentations of two types of software: a number of CFD packages and generic visualisation systems. In addition to these presentations some general lectures and keynote addresses. were also given.
The CFD presentations included FLOW3D (I P Jones, AEA Technology), FLAVIA (G Bugeda, Barcelona), CF View (D Vucci, Brussels), HIGHEND (H C T Pagendam, Deutsch Forschungsanstalt) and GRAFFITI (D C Hatch, Rolls Royce). The presentation on FLOW3D addressed the pre-processing requirements of automated grid generation and CAD/CAM integration, some post-processing requirements including contours and vectors on planes and the displaying of particle tracks. Also shown was a video of the modelling of the King's Cross fire using FLOW3D, depicting the model of the Picadilly Line tunnel, escalator tunnel and booking office. The simulation showed how the flames curled across the ceiling of the escalator tunnel and the so-called trench effect.
The commercial visualisation systems presented included PV-Wave from Tessella Support Systems plc, the Data Visualizer from Wavefront Technologies and the MPGS System from Cray Research. One feature of PV-Wave is the facility of being able to read any data format (e.g. FORTRAN formatted or unformatted). The Data Visualizer allows access to vector and scalar data from any workstation or computer on a given network, accommodate any size data set (the only constraint is the size of the virtual memory of the hardware being used) and handle unstructured grids. MPGS is orientated toward CFD and finite element problems on super computers and interfaces to packages such as FLUENT, PHOENICS, FIDAP and ABAQUS.
Professor D B Spalding (CHAM), in the first keynote address, showed that Notebook PCs are now powerful enough and CFD software flexible enough to enable live demonstrations of flow simulations to be performed in the classroom. Professor J C R Hunt (Met Office) discussed the advances in fluid mechanics arising from the analysis of the visualisation of numerically and experimentally generated data.
After an intense and varied week of seeing a number of interesting visualisation products, software packages and more general lectures, it proved to be a highly successful workshop with good organisational work by Denys Nicholas, RAL.
Manjit Boparai, Informatics
Graphics and Visualisation: Techniques and Tools
A Novel One Week Course for Postgraduate Students of All Disciplines
The ability to visualize data is becoming a key aspect of many areas of research. There are striking examples (such as study of the ozone layer) where scientific discoveries have only been possible through exploring visual representations of observed or simulated data. It is also becoming increasingly important to present the end-results of scientific research in the most effective manner - good use of graphics can make a powerful impact The Advisory Group on Computer Graphics (AGOCG) is organising an intensive course on Graphics and Visualization - Tools and Techniques, aimed at postgraduate students and research assistants of any discipline who wish to use computer graphics for the exploration and presentation of their data. The mix of different disciplines is intentional and will be a feature of the course. Attendees should be computer literate, preferably with experience of Unix workstations. Programming will be kept to a minimum, and so extensive programming expertise. is not a requirement.
The objectives of this one week residential course are to:
- Give students an appreciation of effective ways of displaying data visually - including good graphic design, use of colour, how to visualize multivariate data, and use of animation
- Introduce some of the basic ideas in 3D computer graphics for modelling and rendering, including how to represent objects in a computer, how to show what they look like from different angles and some of the tricks of the trade which enhance the appearance of the resulting picture
- Give students practical experience, using systems they are likely have available at their home site
This will be achieved by a mixture of lectures and practical sessions in which students will use a range of software for visualization and graphics - including a state of the art visualization system such as AVS or Iris Explorer. The practical work will steadily introduce students to these systems, working singly and in groups, and will take the form of a series of challenges with specific objectives.
The course leaders are
- Ken Brodlie (University of Leeds)
- Terry Hewitt (University of Manchester)
- Phil Willis (University of Bath)
Julian Gallop (Rutherford Appleton Laboratory) will lead a team of demonstrators who will provide assistance in the practical sessions. All four are leading UK figures in graphics and visualization, with many years experience in lecturing, research and support of users in different disciplines. Practical work will be carried out in a laboratory of 20 Silicon Graphics Indigo workstations, running Unix.
The course will be held at the University of Leeds, on 4-8 January 1993. The cost will be £280 per person inclusive of accommodation.
Ken Brodlie, University of Leeds
Workshop on Developing Parallel Engineering Applications
The EASE Programme is organising a Workshop on Developing Parallel Engineering Applications to be held at The Cosener's House, Abingdon, on 23-24 February 1993.
Faced with the variety of parallel machines, programming models, parallel language extensions, development environments and tools, it is not surprising that application producers have yet to take up parallel processing technology in any wholehearted way despite its technical promise of affordable scalable performance. Perceived difficulties in developing applications to run on parallel systems has also limited exploitation of the technology to date.
The past two years have seen an increasing degree of commonality of approach across parallel systems, e.g. in message passing based on through-routing, in the emergence of systems with virtual shared memory, and in the growing convergence of parallel and distributed systems. Programming parallel applications has certainly become easier and more portable as a result, in most cases with little or no loss of performance.
The objectives of the Workshop are to:
- Identify the state of the art in these areas
- Pool experience from a wide range of parallel systems
- Identify particular needs and priorities for developing engineering applications
- Predict and potentially influence future trends
- Provide a forum for exchange of views between users, application developers, and parallel computing specialists.
Topics of interest include:
- Portability and future-proofing
- Portable approaches to parallel programming
- Practical experience with engineering applications
- Development environments and tools
- Run-time environments
- Virtual parallel architectures
The programme will consist of selected presentations from position papers, a small number of invited presentations, discussion sessions in subgroups, and a closing plenary session. The number of places is limited (about 40) to foster active participation by all. A booklet of position papers will be distributed three weeks before the Workshop. A report will be produced containing the conclusions of the discussion sessions.
The Workshop is funded by the EASE Programme and will start mid-morning on 23 February and will finish mid-afternoon on 24 February. Accommodation for the night of 22 February preceding the Workshop can also be arranged.
B W Henderson Workshop Secretary RAL
Visualisation CC Meeting Reports
Effective Use or Colour, Hardcopy and Video Facilities in Visualisation, RAL, 17 March 1992.
Prof J Lansdown (Middlesex Polytechnic) spoke on the usefulness of colour as a visualisation tool and described how we perceive colour with some examples of good and bad colour usage.
Mr L W McDonald's (Crosfield) spoke on how and why is colour used for communicating information, how can the user select and mix colour for particular purposes and what are the perceptual methods for organising and manipulating colour. A set of guidelines on how to use colour properly in computer graphics can be found in his July 1990 Displays article.
Mr R L Middleton (Edinburgh University) gave a thorough description of the techniques for producing colour printouts.
Mr C D Osland (RAL) spoke on the pitfalls of recording visualisation results on video. The problems include: thin line flicker, wagon-wheel effects, hatching producing fringes and poor saturation of colours.
Mrs M M Hindmarsh (Newcastle University) spoke on her experiences of making a teaching video. One could only conclude: do not produce a teaching video if you do not absolutely need it!
Mr Mike Lamming (Xerox EuroParc) took a familiar problem why a print image is disappointing compared with the corresponding display image - and provided a practical solution. More details can be found in his October 1990 ACM Transactions on Graphics article.
Dr D Clarke's talk was about the need for people who are involved in computing, and who wish to make a video, to benefit from those who have long experience of video production.
Visualisation in Engineering Meeting. Sheffield University, 28 May 1992
Attended by over 40 people, the entire event was recorded on video.
Prof Swithenbanks' talk on work at Sheffield University concluded that benchmark results show top range computer systems are not needed for visualisation in engineering.
Dr M Sabin (PEGS Ltd) said that a task of visualisation is making apparent the emergent properties of the analysed system. In order to do so, it needs flexibility, programmability and transparent abstraction.
Mr W Roberts (Hydraulics Research Ltd) described his organisation's research areas (free surface flow and transport of other substances by this flow). It is expected that interactive animation in 3D and virtual reality will be the future for visualising time-dependent flows.
Dr D Catherall (Royal Aerospace Establishment) classified data formats and associated visualisation techniques and offered advice on techniques to use.
Prof E T Woodburn (UMIST) explained the use of visualisation in validating mathematical models (of flotation froths) and described visualisation process results compared with experimental results recorded on video.
Mr N Simpson (Rolls-Royce, Derby) explained why they developed their own software (GRAFFITI) and suggested that the benefits from using visualisation were improved understanding of physics, usefulness as a debugging tool, better communication between engineers and improved awareness at senior levels of the organisation.
Mr J Cox (University of Manchester) explained the role of visualisation in the 3D semiconductor device simulation process. Although visualisation systems are interactive, are truly 3D and allow some results to be obtained quickly, they need to be accessible to users and they have limited ability to customise the user interface.
Mr P Hall and Me M Fuller (Sheffield University's VisLab) talked about their work on surface and volume rendering algorithms for scalar and vector fields. Solutions to the problem of visual clutter when visualising vector fields at high sample rate were proposed. Work on perceptual problems of colour and on selective transparency was described.
Information on how to obtain a video copy of the presentation will be advertised in a future Engineering Computing Newsletter.
Mrs R Popovic Informatics
SENAC Lecture Tour
The developers and European distributors of SENAC (University of Waikato, New Zealand, and University of London Computer Centre) are pleased to announce a series of lectures on the SENAC mathematical/numerical system at various academic institutions in the United Kingdom.
SENAC - Software Environment for Numeric and Algebraic Computation - is an environment providing facilities for algebraic, numeric and graphical computation, including a high level easy-to-use interface to the NAG Fortran and Graphics libraries. The system also has its own numeric and graphical routines which contain an interactive form of the popular Numerical Recipes software.
The lectures will introduce SENAC and provide an overview of the algebraic and numeric operational mechanisms and requirements for mixed applications which combine calculus with graphical data representation. Applications of SENAC will also be discussed. Each lecture will illustrate the use of SENAC in a particular discipline area as follows:
- Engineering/Physics/Applied Maths: change of variables, Euler equations, three body problem, statics, PDE Laplace and contouring, pipe network flows, triangulations
- Statistics/Economics/Business: optimization, maximum likelihood, linear programming, spectral analysis, portfolio prediction
- Pure Maths/Computer Science: primality testing and factoring of large integers, symbolic integration, Grobner bases and computation in multinomial ideals, polynomial factorization
Sixth International Conference on Industrial and Engineering Applications of Artificial Intelligence and Expert Systems IEA/AIE-93
Edinburgh, Scotland June 1-4, 1993
Sponsored by the International Society of Applied Intelligence and co-operated with major international organizations and universities including ACM/SIGART, AAAI, IEEE Computer Society, ECCAI, CSCSAI, IAKE, INNS, JSAI, South-Western Texas University.
P. W. H. Chung, Loughborough
Fourth European Software Engineering Conference, Garmisch, Germany
September 13 -17 1993
In 1968, an international workshop was held in Garmisch which coined the term software engineering. In recognition of the 25th anniversary of this key historic event, the 4th European Software Engineering Conference will be held in Garmisch in 1993. Original papers are invited on any aspect of software engineering and papers on the following topics are particularly welcome:
- Software engineering and CSCW
- Software engineering for distributed systems
- Software measurement and experimentation
- Software maintenance
- Requirements engineering
- Critical systems engineering
- Computer-based systems engineering
- Comparisons between development methods and paradigms
- Reports on practical experience with new techniques
Papers should be no longer than 6000 words and should include a title page with a short abstract and the authors' addresses. Papers must be received by the Programme Chair by 23rd November 1992. We welcome correspondence by fax or E-mail but fax or electronic copies of papers will not be accepted.