Hi Amila 2013/1/3 Amila Surendra <[email protected]>
> *Thanks a lot for your quick reply, but unfortunately aviation/space > related projects are not in our areas of interest and currently we don't > have any application ideas based on Constellation platform.* > You're welcome! :) I also got a lot of help in here, so I try to return it if I can :). I have some ideas, so if you're interested I would like to present it to you. I made some posters that explain some of them :) https://bitly.com/algopost > *The project is done as a requirement of Computer Science and Engineering > Degree at University of Moratuwa as the final year project. We are able to > spend around 5-10 hours a week for the project.* > Ah, okay. That's quite some time. I do Constellation as a young academics project at my university and I also spend this amount of time per week for it and the application I'm currently coding. And when we started with Constellation it took about 6 months from idea to public start. > *Currently we don't have any specific project ideas and we'd like to work > on the areas we have mentioned in the previous mail. However one of our > main requirements is the project should have a research and development > component as its a mandatory requirement of the module. We are specifically > interested in building/improving Data Mining, Parallel Computing and Cloud > platforms. If there are ideas based on existing platforms we are more than > happy to start working on them as soon as possible. * > Yeah for sure, when it's a study project of your courses. But with Data Mining, Prallel Computing and Cloud Computing you describe "methods" and you need those for your task. Because you need a numerical problem, model it, code it and THEN prepare this application to be able to us those methods. Like my TrackJack app for Constellation. I can divide the combined field of task into separated and independently processable workunits. And these WUs are used with the app within the DC environment of BOINC. So you still need to have an objective for your app doing the computation. And it should be a useful objective. For example the example app for BOINC is an app that changes the letters of in a txtfile from small to capital letters. And that's not really worth massive computing time :). So you need to think which numerical problem you want to solve. then it's just a matter for coding, be it for DC or even combined for multicore and gpu parallel computing. But it still needs time and I don't know how much of you will work on it. At least you all can't administrate the boinc server :). > *However, we are still in the process of finding a project idea for us. > So we will keep in touch with your project ideas and if we grow an > interesting on one of already existing project ideas in or come up with a > new project idea, we’d definitely contact you back.* > yeah, would like to read you again and what will be the objective of your app :). > * > * > *Thanks a lot for your time and support* > You're welcome. Bye! Andreas _______________________________________________ boinc_dev mailing list [email protected] http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev To unsubscribe, visit the above URL and (near bottom of page) enter your email address.
