ACADEMIA
MATLAB Runs on Enabling Grids for E-sciencE
The software is a high-level language and interactive environment that enables users to perform computationally intensive tasks faster than with traditional programming languages such as C, C++, and Fortran. Widely regarded as a powerful piece of simulation software, for use in everything from optimizing rocket launch control settings to vector analysis, it is now fully compatible with any grid computing system using gLite middleware.
"Our motivation for incorporating this tool was rather straightforward. MATLAB is one of the most popular and important general-purpose scientific software tools. A large number of grid users from a multitude of scientific disciplines have been requesting for a long time to be able to run MATLAB applications within EGEE's gLite infrastructure," says Vangelis Floros, who helps to coordinate and manage the EGEE user community and application support activities.
"Through our collaboration with The Mathworks (creators of MATLAB) we managed to achieve a very efficient integration between the two platforms -- MDCS and gLite -- on the technical level. Usage of the software is still in trial phase but currently we are working very closely with them to define a grid-friendly licensing model in order to be able to offer it very soon for production-level exploitation."
Sample Cases
The software can be used in a number of places. A few examples include:
Finance -- Spyros Skouras of the Athens University of Business Economics and Business is examining financial market data in extremely fine detail.
"To examine a market's microstructure, we use data that are timestamped at the millisecond level and measure tens of variables for thousands of financial instruments," says Skouras. "Such a fine-grain view can shed light on issues of relevance to academics as well as professional traders who can translate milliseconds into profits. But it also means we work with unusually large data sets that are not easily manageable."
Skouras recently began using MATLAB on EGEE in his research, which he says is well suited to grid technology, because the same analysis is performed on data for each instrument. He believes many researchers working in finance, economics, econometrics, statistics and other fields may become users of EGEE if they become aware of how accessible the grid is, now that the interfacing can be done with MATLAB tools they are already familiar with.
Making better lasers -- Research fellow Pavlo Ivanov from the University of Bristol is seeking to improve fiber-optic telecommunication networks by making better semiconductor lasers with the use of MATLAB on EGEE.
"Making a new laser is very expensive," says Ivanov. "Improving design by trial and error would be a disaster. Laser researchers need to optimize their designs prior to fabrication and MATLAB can help me understand how the lasers are behaving."
Laser-created light is used to carry information on fiber-optic communication lines. If Ivanov can, as he hopes, build a high-power, single-mode laser, it could translate into better quality and longer distance telecommunications. Ivanov hopes to present his most recent findings at the upcoming EGEE'09 conference in Barcelona this September.
Source: International Science Grid this Week and EGEE’s homepage (www.eu-egee.org)