• Subscribe

Feature - EDGeS helps us all talk to each other

Feature - EDGeS helps us all talk to each other


With AutoDock, biologists can predict how small molecules will interact with known larger molecules - vital to evaluating candidates for new drugs. Image courtesy www.YASARA.org

Like Bill Murray adrift in Japan in the movie "Lost in Translation," things can go awry when service grids and volunteer grids try to talk to each other.

Service grids, such as EGEE, often involve large institutions whose work method is centered around collaboration - to the point where participants often know each other personally. To ensure security, resources and users pass through an accreditation process, and the resulting "certificate" provides proof that this occurred successfully. The certificate is key to allowing jobs to be submitted and completed.

On the other hand, volunteer systems - such as BOINC or XtremWeb-based desktop grids - consist of large numbers of personal computers linked together by individuals who barely know each other, if at all. They run on a different trust model, where the element trusted is the application itself. Home users receive jobs for their home computers from the application server. On a practical level, these hundreds or thousands of users cannot all ask to be certified - yet they still need to know that their machines will not be harmed by the applications running on it.

The result: Researchers who had an application designed for running jobs through one approach found it did not work on the other.

Until now.

Enabling Desktop Grids for e-Science, or EDGeS, makes it easier for researchers to take applications that have been customized, or "ported," for use on the service grid and make them usable on volunteer, desktop grid systems. This potentially opens up vast new resources, as volunteer systems can involve tens of thousands of previously unavailable central processing units - the microprocessors that are at the heart of a computer.

"For EGEE users, this means that they can, from their familiar gLite environment, submit work to desktop grids . . . and can get results in a much shorter time," says EDGeS project co-ordinator Peter Kacsuk of the Hungarian Academy of Sciences.

ViSAGE provides the ability to conduct complex video analyses in real time by parallelizing the entire process, via the use of EGEE. Image courtesy Correlation Systems Ltd.

To build a bridge

EDGeS' solution brings these two computing approaches together via its "Bridge" - a software package running on a server at EDGeS headquarters in Budapest that allows jobs to pass between these two trust models. It allows information to pass between desktop grids based on BOINC or XtremWeb middleware, and service grids such as EGEE's gLite-based middleware for grid computing.

Examples of it in action in the real world include:

• A platform for processing video streams in a grid environment. Known as ViSAGE (for Video Stream Analysis in a Grid Environment), this lets users analyze complex videos in real time by parallelizing the entire process. This makes the use of powerful image processing algorithms possible in cases where they would normally be impractical for use on real-time data. ViSAGE may one day assist large surveillance systems by automating part of the intruder detection process on multiple video cameras.

ViSAGE was originally developed as part of the Self-Organizing ICT Resource MAnagement (SORMA) project. Compared with applications that process video streams serially, and depending upon the amount of grid resources provided, parallelization can shorten processing time by up to 1000%.

• A "docking" software allows researchers to predict how small molecules, such as substrates or drug candidates, bind to a receptor of known biomolecules. Using AutoDock - a program already ported for use on the grid via EGEE - biologists can predict how small molecules, such as drug candidates, will interact with known larger molecules. But now, these same researchers can harness the power of volunteer desk-top grids as well.

• A platform that simulates the dynamics of laser devices, which are difficult to analyze and not easily parallelized. The platform developed by Cellular Automata-based Laser Dynamics, or CALD, could benefit researchers in many other fields that simulate complex systems: viral infection modelling, tumor growth simulation, materials science, and bio-removal of toxic contaminants, among others.

Currently, hundreds of jobs cross the EDGeS Bridge, but Kacsuk says the number could grow explosively in coming months.

At the moment, EDGeS has connected about 100,000 computers from BOINC and from XtremWeb-based desktop systems to EGEE.

-Danielle Venton, EGEE

Join the conversation

Do you have story ideas or something to contribute? Let us know!

Copyright © 2023 Science Node ™  |  Privacy Notice  |  Sitemap

Disclaimer: While Science Node ™ does its best to provide complete and up-to-date information, it does not warrant that the information is error-free and disclaims all liability with respect to results from the use of the information.

Republish

We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit ScienceNode.org — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on ScienceNode.org” containing a link back to the original article.
  4. The easiest way to get the article on your site is to embed the code below.