MindModeling@home
Aus Rechenkraft
Das Ziel dieses Projektes ist eine Verbesserung der Kognitionswissenschaft durch Distributed Computing. Die Forschung fokussiert sich dabei auf eine rechengestützte Modellierung von Wahrnehmungsvorgängen zum besseren Verständnis der menschlichen Gedanken. Diese Abläufe sind deswegen wichtig, weil sie die menschliche Arbeitsleistung und Lernfähigkeit ermöglichen und steuern.
Inhalt |
Projektübersicht
 MindModeling@Home
| |
|---|---|
| Name | MindModeling@Home |
| Kategorie | Biologie |
| Ziel | Verbesserung der Kognitionswissenschaft |
| Kommerziell | nein |
| Homepage | mindmodeling.org/beta/ |
| Dieses Projekt wird in Arizona, USA durchgeführt. |
Projektstatus
| Projektstatus
| |
|---|---|
| Status | aktiv |
| Beginn | 2004 |
| Ende | pausiert, startet wieder 2010, weitere Infos |
Projektlinks
- Hintergrundinfos (Eng.)
- Hintergrundinfos
- Download
- FAQ
- Theorie des Wahrnehmungsvorgangs
- Neuigkeiten (RSS-Feed)
Beta Version:
- Neuigkeiten (RSS-Feed)
Statistiken
Clientprogramm
Betriebssysteme
| Windows | ||
| Linux | ||
| DOS | ||
| MacOS X | ||
| BSD | ||
| Solaris | ||
| Java (betriebssystemunabhängig) |
WU-Informationen
| Name | RAM | Dauer | Deadline | Speicherplatz | Download | Upload |
|---|---|---|---|---|---|---|
| ACT-R | 2,7 MB | bis 0:15 h (AMDX2 32bit 3,0 GHz) | 4 Tage | MB | MB | MB |
| DSST | MB | h ( GHz) | unbekannte | MB | MB | MB |
| PRP | 27 MB | 0:30 h (2,4 GHz) | 5 Tage | MB | MB | MB |
| PVT | MB | h ( GHz) | unbekannte | MB | MB | MB |
| SAST | MB | h ( GHz) | unbekannte | MB | MB | MB |
Die Deadline ist die Zeitspanne, in der die Work unit berechnet sein muss.
Installation
MindModeling@home benutzt die BOINC-Infrastruktur. Die Anmeldung, Installation und Konfiguration sind auf der allgemeinen BOINC-Seite beschrieben.
Veröffentlichte Versionen
- 12.01.2009: ACT-R cognitive modeling environment 3.45 (Windows, Linux, Linux 64bit)
- 11.09.2008: ACT-R cognitive modeling environment 3.40 (MacOS X)
RSS-Feed
- Changes
- As it has been several monthes since my last news post, I will attempt to address some of the growing concerns relating to the current state of this project. On the development side, the team is working hard to expand our IT infrastructure and implement new technologies that will allow for a more reliable and robust system. We're also taking great strides to develop a new backend portal for our modelers to make our project easier-to-use and include more features. What does this mean for our volunteers? A more reliable system with quicker debug times and plenty of available work. What does this mean for our project? Advances in cognitive modeling at a much faster pace and grander scale than ever before. Unfortunately, there is also some bad news. We are currently unable to host our MidModeling servers under the same agreement that we've been operating, and we will need to find a new home. We've got several options under way and should be hosting/serving MindModeling.org again sometime in early 2010. It is a setback, but it has also given us the time to make some necessary changess to our code and underlying architecture. Thanks for all your contributions and patience while we go through this period of readjustment. -Tom "It is not the strongest of the species that survive, nor the most intelligent, but the one most responsive to change." - Charles Darwin, English Naturalist [?]
- MindModeling.org Banner Contest 2009!!
- While our researchers are preparing new models to run on MindModeling.org, the MM team has decided to take some time to update and make improvements to the website. Part of this is the inclusion of a Mind Modeling banner that can be displayed at the top of our main webpage. Instead of simply making the banner ourselves, we decided to open up the design to all of our volunteers and hold a contest to select the best design. The banners will be voted on by the MindModeling team, and the winner will get to see their work displayed on MindModeling.org for at least the next year. We will keep the contest open until the end of July, 2009. Please submit all entries to our forum here: Banner Contest 2009 or via email. If there are any questions or concerns, please let me know through email. -Tom [?]
- Daemon shutdown and DSST6 model
- Hello crunchers! A model that we submitted to the volunteers last week was recently aborted, but all the workunits were not purged from the database. We're shutting down the server side daemons until all instances are removed. In the meantime, please abort any workunits with the prefix DSST, because generally they will not work on volunteer machines. Thanks! -Tom [?]
- Database Migration Complete!
- Hello crunchers! MindModeling.org has completed it's database migration, run two small tasks within the last couple of days, and is ready to submit large cognitive models to the volunteer community again. Our team and contributers are expanding, our research is growing, and our models are getting more complex and computationally intensive. We look forward to the active participation from our volunteers, and as always, we appreciate your help! Good luck and happy crunching! (So say we all) -Tom [?]
- Status Update
- After many hours of configurations, builds, and reboots, the new Mind Modeling database is finally up! We're still configuring all our authentication and networking scripts, so please be patient as we wrap up the final stages of the installation. If you notice any abnormalities or inconsistencies with your account, please let me know and I will resolve the matter immediately. -Tom [?]
- Server Updates
- Hi folks, MindModeling.org will be experiencing updates over the next couple days which may result in the server being shut down for extended periods of time. Please be patient and we will provide as much information as possible in regards to the status of the site. -Tom [?]
- MySQL Reboot
- There was a quick disconnect this afternoon between our database server and MySQL. The system is rebooting the data right now, and everything should appear normal again in just a few hours. -Tom [?]
- Mind Modeling Maintenanece
- We're installing a new database this week, and there may be short periods of disconnect while we make the transition. Consequently, we'll be pausing our current jobs until the installation is complete. Thanks for all your support and we look forward to your continued participation! :) -Tom [?]
- New Job! SASTNM_10 is ready for some heavy crunching!
- We've got a new job in the system, SASTNM_10, and it's big one; it should provide crunching for at least several weeks. There are some quirks to it though. Because it's such a large space, and becuase we're running 100 sample of the same parameter set, this job only includes one parameter set per WU. What does this mean for crunchers? Well, since our 'percent complete' calculation is based off the completion of the first paramter set, having only one parameter set per WU is going to throw our calculation off a bit. This isn't something we haven't seen before, but it can be misleading sometimes. Just know that each workunit will start off at around 90% complete, but will work for about 10-15 minutes before completing. If there are any other issues we will address them as they come up, but feel free to let me know via email or the message boards. And as always, Happy crunching! -Tom [?]