Dr Sastry and colleagues have developed new techniques to enable scientists to visualise data using distributed resources on a computing Grid. Dr Toby Perring is helping her to develop the techniques for users of the ISIS spallation neutron source at the CCLRC Rutherford Appleton Laboratory (RAL). "We've got really good instruments and need the software to exploit them to the full. The Grid will allow us to do this. All of us who run huge multi-detector instruments see this as the way forward," he says.
Two major new facilities are due to come on line at RAL over the next few years. The Diamond synchrotron X-ray source will start producing data on the crystal structure of many molecules, in particular those of biological significance, by the end of next year. Opportunities to study materials with neutron beams will more than double from 2008 when a second target station comes online at ISIS. Neutrons are especially useful for studying the position and motion of atoms within materials, which govern properties such as magnetism or high temperature superconductivity.
Dr Perring works with one of 19 ISIS instruments, MAPS, to study the motions of magnetic moments and atoms within materials. MAPS detectors take 30 minutes to two days to generate 10 million bits of data from one sample. This data needs to be visualised and compared, or fitted, with theoretical models if it is to be understood and interpreted correctly. "This is a sheer computing power problem. With a single processor we can't fit all our data to the models," says Dr Perring.
Dr Sastry has taken the data-ftting program Dr Perring was using on his desktop and implemented it on the National Grid Service (NGS). RAL manages the NGS which makes distributed computing resources at six universities available to UK academics. Using the GAPtk toolkit, which Dr Sastry and colleagues developed, the output of the data-fitting program can then be visualised using advanced visualisation tools available on the Grid. The whole process can be controlled from a desk or laptop using any desktop visualization tool that the user is familiar with. "Users can visualise and manipulate gigabytes of data as single entities in near real time without expensive specialised hardware", says Dr Sastry. GAPtk is also being used in other applications, including the Integrative Biology e-Science project.
Previously, Dr Perring could fit only parts of the data to models in his single processor machine. Now, he says he can fit all the data and so see the bigger picture. Without the use of the Grid, it would have taken him 50 times longer to do this than now and "I would never have attempted the task", he says.
So far, he has tried out the new methods in two test cases, one on the ferromagnetic properties of pure cobalt. "Now I can get interactive feedback I can see where the model fits and where it doesn't. I can ask 'what if?' questions. I wouldn't have been able to go exploring like this before," he says.
Plans are afoot to roll these new methods out to ISIS users later this year. Meanwhile Dr Perring and Dr Sastry are working on extending them to the analysis of 100s of different data sets simultaneously. The ability to integrate data like this will open up the study of ever more complex materials, such as ferroelectrics and multiferroics. These are good candidate materials for sensors because they change some aspect of their internal properties when external conditions change.
A presentation, "Grid enabling legacy applications for scalability - experiences of a production grid on the UK NGS", based on the work reported here, is scheduled to take place at AHM2005 on Monday 19th September at 17.00-20.00.
To attend the e-Science All Hands meeting, or the media briefing at 3pm on Wednesday 21 September, go to http://www.allhands.org.uk/pr/index.html. Conference website http://www.allhands.org.uk/.
Contacts
Dr Lakshmi Sastry, Rutherford Appleton Laboratory, tel. 01235 445428, e-mail M.Sastry@rl.ac.uk
Dr Toby Perring, Rutherford Appleton Laboratory, tel. 01235 445428, e-mail t.g.perring@rl.ac.uk
Judy Redfearn, e-Science/Research communications officer, JISC/e-Science Core Programme tel. 07768 356309 e-mail: judy.redfearn@epsrc.ac.uk
Links
GAPtk http://www.e-science.clrc.ac.uk/web/projects/gaptk; http://www.e-science.clrc.ac.uk/web/projects/eve;
Diamond http://www.cclrc.ac.uk/Activity/Diamond;
ISIS http://www.isis.rl.ac.uk/aboutIsis/index.htm
MAPS http://www.isis.rl.ac.uk/excitations/maps/
UK e-Science Programme www.rcuk.ac.uk/escience
NGS http://www.e-science.clrc.ac.uk/web/projects/jisc_grid
Notes for Editors
e-Science is the very large scale science that can be carried out by pooling access to very large digital data collections, very large scale computing resources and high performance visualisation held at different sites.
A computing grid refers to geographically dispersed computing resources that are linked together by software known as middleware so that the resources can be shared. The vision is to provide computing resources to the consumer in a similar way to the electric power grid. The consumer can access electric or computing power without knowing which power station or computer it is coming from.
The UK e-Science Programme is a coordinated £230M initiative involving all the Research Councils and the Department of Trade and Industry. It has also leveraged industrial investment of £30M. The Engineering and Physical Sciences Research Council manages the e-Science Core Programme, which is developing generic technologies, on behalf of all the Research Councils.
The UK e-Science Programme as a whole is fostering the development of IT and grid technologies to enable new ways of doing faster, better or different research, with the aim of establishing a sustainable, national e-infrastructure for research and innovation. Further information at www.rcuk.ac.uk/escience.