[ACT-R-users] the MindModeling infrastructure

[Harris, Jack J Civ USAF AFMC 711 HPW/RHAC]

David, Thanks for your question.

The MindModeling system is mostly composed of many 'consumer computers' with a 
few high-end computers and even a few super computers added in for good 
measure. So, the system is highly heterogeneous and there is a large 
distribution of clock speeds, core-counts and architectures available. Based 
on resource requirements (e.g., software requirements, operating system, etc.) 
of an experiment, the system distributes simulations to respective processors 
accordingly. In the last month while doing an experiment consenting an ACT-R 
model regarding fatigue, the system was sustaining rates at about 4 TFlops.

I hope this helps. Feel free to contact me directly if you would like to 
discuss your individual computational needs or if you have any further 
questions.
Thanks,
Jack

Jack Harris, PhD
Research Computer Scientist and Cognitive Scientist
Cognitive Models and Agents Branch (711 HPW/RHAC)
Wright-Patterson AFB, Ohio 45433
(937) 938-3937


From: -dp- [mailto:david at wooden-robot.net]
Sent: Saturday, October 20, 2012 10:56 AM
To: act-r-users at act-r.psy.cmu.edu <act-r-users at act-r.psy.cmu.edu>
Subject: [ACT-R-users] the MindModeling infrastructure


Although the MindModeling system seems highly distributed, and it might be 
difficult to cite typical performance, is there a typical per-thread clock 
speed that one could compare to that of CPUs in consumer computers?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 5624 bytes
Desc: not available
URL: <http://mailman.srv.cs.cmu.edu/pipermail/act-r-users/attachments/20121022/7990edf5/attachment.bin>