This is a weblog where you can find information about BOINC, the Berkely Open Infrastructure for Networked Computing. Please use the categories and pages on the right sidebar to navigate. You can also search this place (top right). You can contact me if you want to create your own project or don’t know if your project will fit into the BOINC framework.
If you can’t understand what you have found you may be on a German subpage. I couldn’t translate all content so please contact me if you want specific help.
After a very delicious lunch at a nice Indian restaurant some of the workshop participants gathered together at a pub and managed to arrive to the second day on time.
This day started with the project presentation of Ibercivis by Fermin Serrano. This is a Spanish-Portuguese initiative that provides a huge BOINC umbrella Project for scientists in those countries. It’s almost like WCG but with a more regional context and funded by universities.
Next speaker was Uwe Beckert who described his way from volunteer to project admin of yoyo@home and RNA@home. The last one will publish a paper about some new things they learned using the volunteers computers soon. Also they gone provide an interface for scientists to submit own tasks to the project. Continue Reading »
The first day started with a talk by David Anderson summarizing the development of BOINC over the last year. Especially the new Creditsystem. Another visible addition to the manager is the new notices tab. This will show translatable information for configuring BOINC, project news and even private messages in the new generation of the BOINC manager, currently in testing. Upcoming tasks for the developers are, among other things, the rewrite of the Manager (at least the Simple View) and the integration with Drupal. Continue Reading »
The third day had two astonishing new revelations for the BOINC world. At first there where some talks about virtualizing the scientific application through different virtualization techniques. After the coffee break Daniel Lambrana presented the first BOINC client pre-installed on a USB-Stick that can be distributed by a project and can directly contribute the unused computer cycles of the computer it gets plugged into. There is a Windows and Linux version installed that work on Windows XP and almost all Gnome driven Linux distributions. Windows Vista is currently not supported because the autostart required for this setup is not working. The prototype is connecting the user to the extremadurathome.org project that also has a good video explaining Volunteer Computing.
After that there was a presentation about a Campus-wide Supercomputer at the Pontifical Catholic University of Peru. Genghis Rios Kruger presented the way he and his group of computer scientists used BOINC to build a Grid using the approx. 500 computers of the different computer labs. Interesting here was that they already provide a simple web-driven interface for the scientist to generate work and receive the results. This is going to be made generic and hopefully can be implemented into BOINC or at least made Open-Source so other projects can use it too.
The second revelation was made by Michael Weber from the University of Marburg who is going to publish a new project focused on RNA research, in cooperation with the Rechenkraft.net non-profit association. This is not going to be a university-funded project but purely operated by volunteers.
During lunch break we where at the same place like the first day and it was great. In case you visit Barcelona and need a good place to eat: try the Carmelitas located in the Carrier del Carme.
In the afternoon there where some BOF sessions (Birds of Feather) discussing some issues regarding the BOINC framework and how it can be made more general and how to implement new stuff like virtual applications and easy to use interfaces for scientists. The main conclusions can be found on the wikipage, where the slides of the talks are also published (some are missing but should be added later).
The first day of the workshop here in Barcelona was very rainy and dull. But there was nothing of this inside the conference room. The day started with a talk by David Anderson the BOINC project leader.
He gave a review of volunteer computing in the last year since the workshop in Grenoble in 2008. The computing power since then increase by a factor of 3 for all volunteer computing projects. Mainly this is provided by the Folding@home project, which really early started to develop GPU applications that are able to utilize graphical processing units on graphic-cards. Approximately half of the overall computing power comes from GPU’s!
Besides this the volunteer base stayed at about 330,000 users. So volunteer computing is still unknown to the general public though the Progress Through Processors initiative launched by Intel und Grid Republic on facebook.
In the last 12 months no new academic projects started to use BOINC as a method to access a huge resource of unused computer power. Nevertheless the funding of BOINC is assured till mid 2011 through a grant from the National Science Foundation and another grant is on its way.
The development of BOINC is guided towards integration of GPU’s from ATI and NVidia at the moment and a new system for credit and runtime estimation is right around the corner. Details were not presented.
There where also presented many extension to the BOINC server that should make it easier for scientist to BOINCify there applications and create new projects that attract new volunteers to volunteer computing.
So much for the first day that’ll last till 6pm and end with a visit to nice Spanish restaurant.