Dear list, I'm using jags to run a number of binomial regression models with a largish number of covariates and several thousand observations. With module "glm" loaded, it still takes 1-2 hours to run 3 chains with an adequate number of iterations to ensure convergence on my Mac server (quad core with MacOS 10.6 and 64-bit R 2.12.2). On a once-off basic this is tolerable, however, for a large number of models this run time will be problematic. I asked on r-sig.mac about the possibility of using a parallel processing package to speed up the runs. As the model is 'embarrassingly parallel' I was advised to consider the "multicore" or "snow" packages. As I'm employing a package (jags) and not my own sampler, I'm not sure if this is feasible. I wondered if anyone on the BUGS list has attempted this with jags?
Many thanks for any advice/suggestions.
Alan Kelly
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail [log in to unmask]
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask]
If this fails, mail [log in to unmask], NOT the whole list
|